[Federal Register Volume 89, Number 195 (Tuesday, October 8, 2024)]
[Rules and Regulations]
[Pages 81620-81774]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2024-21867]
[[Page 81619]]
Vol. 89
Tuesday,
No. 195
October 8, 2024
Part II
Securities and Exchange Commission
-----------------------------------------------------------------------
17 CFR Part 242
Regulation NMS: Minimum Pricing Increments, Access Fees, and
Transparency of Better Priced Orders; Final Rule
Federal Register / Vol. 89 , No. 195 / Tuesday, October 8, 2024 /
Rules and Regulations
[[Page 81620]]
-----------------------------------------------------------------------
SECURITIES AND EXCHANGE COMMISSION
17 CFR Part 242
[Release No. 34-101070; File No. S7-30-22]
RIN 3235-AN23
Regulation NMS: Minimum Pricing Increments, Access Fees, and
Transparency of Better Priced Orders
AGENCY: Securities and Exchange Commission.
ACTION: Final rule.
-----------------------------------------------------------------------
SUMMARY: The Securities and Exchange Commission (``Commission'' or
``SEC'') is adopting amendments to certain rules of Regulation National
Market System (``Regulation NMS'') under the Securities Exchange Act of
1934, as amended (``Exchange Act'') to amend the minimum pricing
increments for the quoting of certain NMS stocks, reduce the access fee
caps, and enhance the transparency of better priced orders.
DATES: Effective Date: December 9, 2024. Compliance dates: See section
VI., titled ``Compliance Dates,'' for further information on
transitioning to the final rules.
FOR FURTHER INFORMATION CONTACT: Kelly Riley, Senior Special Counsel,
Johnna Dumler, Special Counsel, Steve Kuan, Special Counsel, Marc
McKayle, Special Counsel, Leigh Roth, Special Counsel, and Alba Baze,
Attorney-Advisor, at (202) 551-5500, Office of Market Supervision,
Division of Trading and Markets, Commission, 100 F Street NE,
Washington, DC 20549.
SUPPLEMENTARY INFORMATION: The Commission is adopting amendments to the
following rules under Regulation NMS:
------------------------------------------------------------------------
CFR citation (17
Commission reference CFR)
------------------------------------------------------------------------
Rule 600(b)(69)................................... Sec.
242.600(b)(69)
Rule 600(b)(89)................................... Sec.
242.600(b)(89)
Rule 600(b)(93)................................... Sec.
242.600(b)(93)
Rule 603.......................................... Sec. 242.603
Rule 610.......................................... Sec. 242.610
Rule 612.......................................... Sec. 242.612
------------------------------------------------------------------------
I. Introduction
A. Rule 612 Minimum Pricing Increments
1. Background
2. Proposed and Adopted Amendments
B. Rule 610 Fees for Access to Quotations and Transparency of
Fees
1. Background
2. Proposed and Adopted Amendments
C. Transparency of Better Priced Orders
1. Background
2. Proposed and Adopted Amendments
D. Overarching Comments on the Proposing Release
II. Equity Market Structure Initiatives and the Regulation NMS
Proposal
III. Final Rule 612 of Regulation NMS--Minimum Pricing Increment
A. Issues Raised in the Existing Market Structure Related to
Tick Sizes
B. Proposal To Amend Rule 612
C. Final Rule--Minimum Pricing Increments for Orders Priced
Equal to or Greater Than $1.00 per Share
1. General Comments and Discussion
2. Specific Comments on the Proposed Minimum Pricing Increments
3. Comments on the Number of Proposed Increments
4. Comments on Small- and Mid-Sized Stocks
5. Comments on Market Resiliency
6. Comments on Proposed Criteria for Assigning Minimum Pricing
Increments
7. Rule 612(a)--Definitions
8. Rule 612(b)(1)--Semiannual Operative Dates
9. Rule 612(c)--New NMS Stocks
10. Rule 600(b)(89)--Regulatory Data
D. Minimum Pricing Increment for Trades
IV. Final Rule 610 of Regulation NMS--Fees for Access to Quotations
A. Background
B. Issues Raised in the Existing Market Structure and the Need
for the Amendments
1. Amendments to Rule 612
2. Exchange Fee Models
C. Proposal To Amend 610(c)
D. Final Rule 610(c)
1. Comments on Proposed Rule 610(c)
E. Final Rule 610(d) Requiring That All Exchange Fees and
Rebates Be Determinable at the Time of an Execution
1. General Comments
V. Final Rule--Transparency of Better Priced Orders
A. Background
B. Final Rule--Round Lots
1. Round Lot Definition
2. Proposed Acceleration of Round Lot Definition
3. Comments and Response
C. Final Rule--Odd-Lot Information
1. Proposed Acceleration of Odd-Lot Information Definition
2. Proposed Amendment to Odd-Lot Information Definition for Best
Odd-Lot Orders
D. Display of Round Lots and Odd-Lot Information
1. Comments and Response
E. MDI Rules Implementation
VI. Compliance Dates
A. Final Rule 612 Compliance Date
B. Final Rule 610 Compliance Date
C. Final Compliance Date for Round Lot and Odd-Lot Information
VII. Economic Analysis
A. Introduction
B. Broad Economic Considerations
1. Liquidity and Spread
2. Economics of Minimum Pricing Increments
3. Economics of Access Fees
C. Baseline
1. Tick Sizes
2. Access Fees
3. Round Lots, Odd-Lots, and Market Data Infrastructure
4. Affected Entities and Markets
5. Amendments to Rule 605
D. Benefits, Costs, and Other Economic Effects
1. Modification of Rule 612 To Create a Half-Penny Tick
2. Lower Access Fee Cap
3. Exchange Fees and Rebates Determinable at the Time of
Execution
4. Acceleration and Implementation of the MDI Rules and Addition
of Information About Best Odd-Lot Orders
5. Compliance Costs
6. Interactions With Recently Adopted Rules
E. Effect on Efficiency, Competition, and Capital Formation
1. Efficiency
2. Competition
3. Capital Formation
F. Reasonable Alternatives
1. Tick Size Alternatives
2. Access Fee Alternatives
VIII. Paperwork Reduction Act
A. Summary of Collection of Information
B. Proposed Use of Information
C. Respondents
D. Total Annual Reporting and Recordkeeping Burden
1. Initial Burden Hours and Costs
2. Ongoing Burden Hours and Costs
E. Collection of Information Is Mandatory
F. Confidentiality
G. Revisions to Current MDI Rules Burden Estimates
IX. Regulatory Flexibility Act
A. Amendments to Rule 612--Final Regulatory Flexibility Analysis
1. Reasons for the Action
2. Small Entities Subject to the Rule
3. Reporting, Recordkeeping, and Other Compliance Requirements
4. Significant Alternatives
B. Amendments to Rule 610
C. Amendments to Rule 603 and Definitions Odd-Lot Information
and Regulatory Data Under Rule 600
D. Certification
X. Other Matters
Statutory Authority and Text of Rule Amendments
I. Introduction
Consistent with Congress's directive almost 50 years ago to
facilitate the establishment of a national market system,\1\ the
Commission is amending certain of its rules to respond to market
developments since those rules were adopted, so that those rules
continue to benefit investors and the markets. Specifically, the
Commission is taking the following actions to continue to fulfill
Congress's directive and advance the objectives of investor protection
and the maintenance of fair and orderly markets:
---------------------------------------------------------------------------
\1\ See Public Law 94-29 (S.249), June 4, 1975, Securities Acts
Amendments of 1975 (``1975 Amendments''). See also 15 U.S.C. 78k-1.
---------------------------------------------------------------------------
Reduce Transaction Costs for Investors by Reducing Minimum
Pricing Increments. The amendments will relax
[[Page 81621]]
existing restrictions on market-wide minimum pricing increments (``tick
sizes''), thus reducing transaction costs for investors and relaxing a
constraint on price discovery for certain stocks.
The reduced tick size will benefit investors and market
participants by: (i) allowing stocks to be priced more efficiently and
competitively, therefore lowering costs for investors to trade in those
stocks; and (ii) improving liquidity, competition, and price efficiency
in the markets.
Improve Market Quality for Investors by Reducing Access
Fee Caps and Increasing Transparency. The amendments will reduce the
maximum fees that trading centers (e.g., securities exchanges) are
allowed to charge investors for execution against protected quotations
(``access fee caps''). The amendments will also address the lack of
transparency around the cost of a transaction at the time of a trade
execution by requiring exchange fees and rebates to be determinable at
the time of the execution.
The amendments will benefit investors and market participants by:
(i) providing for access fee caps that accommodate the change in tick
sizes; (ii) providing quotations that are more accurate and reflective
of market forces; (iii) mitigating potential conflicts of interest
between broker-dealers and their customers, where a broker-dealer is
incentivized to route to the exchange offering the most favorable fees
or rebates, which can lead to potentially worse execution quality for
customers; (iv) reducing the complexity associated with the fees and
rebates models; and (v) increasing the transparency of transaction fees
and rebates.
Improve Transparency to Investors about Better Priced
Orders. The amendments will increase price transparency by accelerating
the implementation of previously adopted definitions of ``round lot''
and ``odd-lot information'' and by adding a data element for the best
odd-lot orders to buy and sell (``BOLO'') to the definition of ``odd-
lot information.''
These amendments will improve information available to investors
and other market participants about better priced orders in smaller
sizes that are available in the market.
In 1975, Congress explicitly granted the Commission ``broad
authority to oversee the implementation, operation, and regulation of
the national market system'' and the ``clear responsibility to assure
that the system develops and operates in accordance with
Congressionally determined goals and objectives.'' \2\ The 1975
Amendments and section 11A of the Exchange Act set forth Congress's
findings regarding the nation's securities markets and direct the
Commission to facilitate the establishment of a national market system
in accordance with specified Congressional findings and objectives.\3\
---------------------------------------------------------------------------
\2\ Senate Report on Securities Act Amendments of 1975, S. Rep.
No. 94-75 at 8-9.
\3\ In particular, Congress found that it is in the public
interest and appropriate for the protection of investors and
maintenance of fair and orderly markets to assure five objectives:
(1) economically efficient execution of transactions; (2) fair
competition among brokers and dealers and among exchange markets,
and between markets other than exchange markets; (3) the
availability to brokers, dealers and investors of information with
respect to quotations for and transactions in securities; (4) the
practicability of brokers executing investors' orders in the best
market; and (5) an opportunity, consistent with items (1) and (4),
for investors' orders to be executed without the participation of a
dealer. See 15 U.S.C. 78k-1(a)(1)(C). Congress also found that new
data processing and communications techniques could create the
opportunity for more efficient and effective market operations, and
that ``[t]he linking of all markets for qualified securities through
communication and data processing facilities will foster efficiency,
enhance competition, increase the information available to brokers,
dealers, and investors, facilitate the offsetting of investors'
orders and contribute to the best execution of such orders.'' See 15
U.S.C. 78k-1(a)(1)(B), (D).
---------------------------------------------------------------------------
Since 1975, the Commission has regulated the national market
system, adhering to the objectives of efficient, competitive, fair, and
orderly markets that are in the public interest and protect investors,
which are essential to meeting the investment needs of the public and
reducing the cost of capital for listed companies.\4\ The national
market system is premised on promoting fair competition among markets,
while at the same time assuring that all of these markets are linked
together, through facilities and rules, in a unified system that
promotes interaction among the orders of buyers and sellers in a
particular NMS stock.\5\
---------------------------------------------------------------------------
\4\ See Securities Exchange Act Release No. 51808 (June 9,
2005), 70 FR 37496, 37497 (June 29, 2005) (``Regulation NMS Adopting
Release''). In the nearly fifty years since the enactment of section
11A, the Commission has monitored the national market system and its
operation and has periodically reviewed certain of its rules to
address issues that have arisen in the markets with the goal of
ensuring that the regulatory framework continues to fulfill the
goals of section 11A. In each such case, the Commission has been
guided by the objectives embodied in section 11A. The Commission
also formed the Equity Market Structure Advisory Committee
(``EMSAC'') in 2015 to provide diverse perspectives on the structure
and operations of the U.S. equities markets, as well as advice and
recommendations on matters related to equity market structure. The
archives of these meetings are available at https://www.sec.gov/spotlight/emsac/emsac-archives.htm (``EMSAC Archives'').
\5\ See Regulation NMS Adopting Release, supra note 4, at 37498.
``NMS stock'' is defined under Regulation NMS as any NMS security
other than an option. 17 CFR 242.600(b)(65). An ``NMS security'' is
defined as any security or class of securities for which transaction
reports are collected, processed, and made available pursuant to an
effective transaction reporting plan, or an effective national
market system plan for reporting transactions in listed options. 17
CFR 242.600(b)(64).
---------------------------------------------------------------------------
In 2005, the Commission adopted Regulation NMS to modernize and
strengthen the regulatory structure of U.S. equity markets, including
requirements pursuant to which quotations and orders for NMS stocks,
and the markets on which they trade, can compete. These requirements
support the public interest and the protection of investors and help to
ensure fair and orderly markets for the execution of orders in NMS
stocks. Among other things, Regulation NMS provides explicit
requirements for the tick sizes of quotations and orders,\6\ the means
for market participants to access quotations in the national market
system, including a cap on the highest permitted level of fees a
trading center may charge for access to the best quotations of a
trading center,\7\ and how information about quotations and trades is
made widely available to investors, among others.\8\
---------------------------------------------------------------------------
\6\ See Rule 612 of Regulation NMS; 17 CFR 242.612.
\7\ See Rule 610 of Regulation NMS; 17 CFR 242.610.
\8\ See Rules 601, 602, and 603 of Regulation NMS; 17 CFR
242.601, 17 CFR 242.602, 17 CFR 242.603.
---------------------------------------------------------------------------
Nearly two decades later, the technology and economics of trading
have evolved significantly. Transaction volume in listed equities
doubled in the last five years and tripled in the last seventeen.\9\
Electronic trading now dominates equity markets, with latency measured
in microseconds. These changes call for improvements to assure an
efficient and transparent price discovery process, in order to continue
to fulfill Congress's directive and advance the objectives of investor
protection and the maintenance of fair and orderly markets. However,
some parts of Regulation NMS have not been revised since their 2005
adoption. Thus, the Commission is adopting the below described
amendments to certain rules under Regulation NMS.\10\ The following
[[Page 81622]]
subsections provide an overview of the amendments and the rationales
for each.\11\
---------------------------------------------------------------------------
\9\ See Cboe, ``Historical Market Volume Data,'' available at
https://www.cboe.com/us/equities/market_statistics/historical_market_volume/.
\10\ The Commission has amended several aspects of Regulation
NMS to address and reflect changes in the markets since its
adoption. For example, in 2018, the Commission adopted new order
handling disclosure requirements in Rule 606 in response to changes
in equity market structure and order handling and routing practices.
See Securities Exchange Act Release No. 84528 (Nov. 2, 2018), 83 FR
58338 (Nov. 19, 2018). In 2020, the Commission adopted rules to
update the national market system for the collection, consolidation,
and dissemination of equity market data in the national market
system to keep pace with technological developments concerning the
use of market data. See Securities Exchange Act Release No. 90610
(Dec. 9, 2020), 86 FR 18596 (Apr. 9, 2021) (``MDI Adopting
Release''). More recently, responding to changes in market
conditions caused by technological advancements and the increased
participation of individual investors in the equity markets, the
Commission adopted amendments to Rule 605 under Regulation NMS to
update the disclosure of order execution quality statistics reports.
See Securities Exchange Act Release No. 99679 (Mar. 6, 2024), 89 FR
26428, 26429 (Apr. 15, 2024) (``Rule 605 Amendments'') (adopting
amendments to rule 605 under Regulation NMS to update reports on
execution quality).
\11\ See generally Securities Exchange Act Release No. 96494
(Dec. 14, 2022), 87 FR 80266 (Dec. 29, 2022) (``Proposing Release''
or ``Regulation NMS Proposal'').
---------------------------------------------------------------------------
A. Rule 612 Minimum Pricing Increments
1. Background
One way that investors can buy or sell a stock is through the use
of limit orders, which are a type of order that specifies the price
(``limit price'') at which the investor is willing to buy or sell a
security.\12\ Limit orders serve a critical market function by helping
to set prices at which market participants are willing to trade,
revealing the supply and demand for a security, and providing liquidity
to the market. As such, limit orders play a key role in price discovery
and allow investors to participate in the price-setting process.\13\
---------------------------------------------------------------------------
\12\ Whether a limit order can be executed immediately depends
on the limit price in relation to the current market price. For
example, a buy order with a limit price of $10.00 means the investor
would like to buy as soon as possible, but only when the current
market price is at $10.00 or less. By contrast, a ``market order''
is a type of order by which the investor specifies that it wishes to
buy or sell a security at the current market price, regardless of
what the market price is. See generally, Securities Exchange Act
Release No. 96495 (Dec. 14, 2022), 88 FR 128, 132-33 (Jan. 3, 2023);
Regulation NMS Adopting Release, supra note 4, at 37505 n.53.
\13\ Limit orders may be ``marketable'' meaning that its
specified price allows an immediate execution because it matches a
contra-side order, or they may be ``non-marketable'' meaning that
its specified price does not allow for an immediate execution and
therefore it must wait until a contra-side order comes in to trade
with it. Non-marketable limit orders that are submitted to an
exchange are placed on the order book and, if displayable, the price
and size will be displayed in the national market system if it is
the best priced order to buy or sell for such exchange. The
Commission has recognized displayed limit orders as ``a critically
important element of efficient price discovery.'' See Regulation NMS
Adopting Release, supra note 4, at 37517.
---------------------------------------------------------------------------
Recognizing the value of limit orders, the Commission adopted Rule
612 under Regulation NMS, which requires that the prices of quotations
and orders in the national market system be reflected in a specified
minimum pricing increment, also known as the ``tick size.'' Rule 612
required, for quotations and orders of NMS stocks priced at or greater
than $1.00 per share, the minimum pricing increment to be $0.01.\14\ As
a result, subject to certain exceptions,\15\ the quotations and orders
of such NMS stocks are priced in penny increments: $10.00, $10.01,
$10.02, for example.
---------------------------------------------------------------------------
\14\ See preexisting 17 CFR 242.612(a). For quotations and
orders of NMS stocks priced less than $1.00 per share, Rule 612
required the minimum pricing increment to be $0.0001. See
preexisting 17 CFR 242.612(b). However, most exchanges require
stocks listed on their exchanges to maintain a price greater than
$1.00 per share, and consequently $0.01 is the prevailing tick size
for most quotes and orders for NMS stocks. See infra section
VII.C.1.a.
\15\ See infra section VII.C.1.a. (discussing retail programs).
---------------------------------------------------------------------------
The Commission adopted Rule 612 and minimum pricing increments to
address the concern that a market participant could gain priority over
existing limit orders by posting an economically insignificant price
improvement.\16\ For example, consider a market participant that posts
a limit order to buy an NMS stock at $10.00 per share. Without minimum
pricing increments, a second market participant could ``step ahead''
(also known as ``pennying'') of the first market participant by posting
a bid to buy at a price that is higher by an infinitesimally small
amount, such as $10.000001.\17\ This behavior disincentivizes market
participants from posting a limit order in the first place because
another market participant could always gain priority over that first
price by posting a limit order that is better by an economically
insignificant amount.\18\ This may lead to a decline in limit orders,
harm liquidity, and make it more costly to trade.\19\ This hypothetical
scenario illustrates the need for a minimum pricing increment that is
not too small.
---------------------------------------------------------------------------
\16\ When Rule 612 was adopted, the Commission stated that
``[g]reater use of limit orders will increase price discovery and
market depth and liquidity'' and that ``if orders lose execution
priority because competing orders step ahead for an economically
insignificant amount, liquidity could diminish.'' See Regulation NMS
Adopting Release, supra note 4, at 37505, 37553. The Commission was
concerned that stepping ahead of displayed limit orders by
insignificant amounts would deter the submission and display of
limit orders, which would negatively impact price discovery and
market depth and liquidity. See id. at 37553. See also infra section
VII.A (discussing the importance of minimum pricing increments).
\17\ But with the minimum pricing increment of a penny, that
same market participant would be required to post a bid of $10.01
instead.
\18\ See infra sections VII.A, VII.B.2, and VII.D.1.
\19\ See infra sections VII.A, VII.B.2, and VII.D.1.b.i for
additional analysis of pennying.
---------------------------------------------------------------------------
Too big of a minimum pricing increment is also problematic since it
would reduce the quality of price discovery by precluding price
competition for providing liquidity.\20\ More specifically, too large a
tick size can increase transaction costs for investors by artificially
widening the ``bid-ask spread''--the difference between the bid
(highest price a buyer is willing to pay) and the ask (the lowest price
a seller is willing to accept) prices.\21\ For example, consider a
hypothetical scenario where a liquidity provider is willing to bid
$10.121 to buy a stock and offer $10.124 to sell the stock. If the tick
size were $0.005, the resulting bid and offer from this liquidity
provider would be $10.120 and $10.125, respectively, with a spread of
$0.005. If the tick size were $0.01, the corresponding bid and offer
would be $10.120 and $10.130, with a spread of $0.01.\22\ In other
words, but for the requirement under Rule 612 that sets the tick size
to be $0.01 for quotes and orders in NMS stocks priced at or above
$1.00, a smaller tick size would have narrowed spreads in some
instances and allowed prices to better reflect the underlying economics
for certain NMS stocks. As explained below, up to 74.3% of the share
volume transacted in NMS stocks in 2023 may have bid-ask spreads that
are constrained by the current minimum pricing increments.\23\ These
widened bid-ask spreads increase transaction costs for investors.\24\
Conversely, a smaller tick size that allows for narrower bid-ask
spreads would benefit investors by reducing transaction costs.\25\
---------------------------------------------------------------------------
\20\ See infra section VII.B.2.
\21\ See infra section VII.B.2; see also 17 CFR 242.600(b)(16).
\22\ See infra section VII.B.2 (providing a similar example
showing how a minimum pricing increment could double the width of a
bid-ask spread).
\23\ See infra section VII.C.1.b (discussing percentage of share
volume likely to be tick-constrained). See also infra section
VII.B.2 (discussing the definition of ``tick-constrained'').
\24\ See infra section VII.B.2.
\25\ See infra section VII.B.2.
---------------------------------------------------------------------------
The minimum pricing increments in Rule 612 were adopted in 2005,
when the Commission adopted Regulation NMS, and it was an adjustment in
a long series of adjustments to the minimum pricing increments over
time. For many decades, the U.S. equity markets used fractions of a
dollar as minimum pricing increments (e.g., \1/8\, \1/16\, and \1/32\
of a dollar).\26\ Prior to 1997, the minimum
[[Page 81623]]
pricing increment on the New York Stock Exchange LLC (``NYSE'') for
stocks above $1.00 per share was \1/8\ of a dollar (or 12.5 cents).\27\
In 1997, NYSE and the Nasdaq Stock Market LLC (``Nasdaq'') revised
their rules to use the minimum pricing increment of 1/16 of a dollar
(6.25 cents).\28\ In January 2000, the Commission mandated decimal
pricing (i.e., moving from fractional increments to penny increments)
in certain securities,\29\ and by April 2001, the market had fully
converted to decimal pricing.\30\
---------------------------------------------------------------------------
\26\ See Staff Report to Congress on Decimalization, Commission
(July 2012) (``Staff Decimalization Report''), available at https://www.sec.gov/files/decimalization-072012.pdf, at 4. Staff reports,
Investor Bulletins, and other staff documents (included those cited
herein) represent the views of Commission staff and are not a rule,
regulation, or statement of the Commission. The Commission has
neither approved nor disapproved the content of these staff
documents, and, like all staff documents, they have no legal force
or effect, do not alter or amend the applicable law, and create no
new or additional obligations for any person.
\27\ See Self-Regulatory Organizations; New York Stock Exchange,
Inc.; Order Granting Approval to Proposed Rule Change Relating to
Trading Differentials for Equity Securities, 62 FR 42847, 42848 n.5
(Aug. 8, 1997). See also Division of Market Regulation, Market 2000:
An Examination of Current Equity Market Developments (1994),
available at https://www.sec.gov/divisions/marketreg/market2000.pdf,
at 37-38, fn. 43 (describing NYSE's tick size of \1/8\ of a dollar
in 1994).
\28\ See Staff Decimalization Report, supra note 26, at 4-5.
\29\ See Securities Exchange Act Release No. 42360 (Jan. 28,
2000), 65 FR 5003 (Feb. 2, 2000).
\30\ See Staff Decimalization Report, supra note 26, at 5-6.
---------------------------------------------------------------------------
Up to this point, minimum pricing increments for NMS stocks were
set by the individual trading venues. But in 2004, as part of
Regulation NMS and pursuant to the authority under the 1975 Amendments,
the Commission proposed Rule 612 to implement market-wide uniform
minimum pricing increments for quoting in NMS stocks.\31\ The
Commission stated that, while the benefits of decimal pricing had
justified the costs, there was a potential for costs to investors and
the markets to surpass the benefits if the minimum pricing increment
decreased beyond a certain level, and the proposed rule was designed to
address the scenario where market participants attempt to step ahead of
competing limit orders at the smallest economic increment possible.\32\
Thus, the Commission adopted Rule 612 in 2005, which established the
minimum pricing increments of $0.01 for quotations and orders of NMS
stocks priced at, or greater than, $1.00 per share, and $0.0001 for
quotations and orders of NMS stocks priced under $1.00 per share. The
Commission stated that, at the time, it did not believe that the
potential benefits of marginally better prices offered by allowing sub-
penny quoting in securities were likely to justify the costs of
permitting such quotes.\33\
---------------------------------------------------------------------------
\31\ See Securities Exchange Act Release No. 49325 (Feb. 26,
2004), 69 FR 11126, 11171 (Mar. 9, 2004) (``2004 Regulation NMS
Proposing Release'') (``the Commission is proposing a rule that
would prohibit every national securities exchange, national
securities association, ATS (including ECNs), vendor, broker or
dealer from ranking, displaying, or accepting from any person a bid
or offer, an order, or an indication of interest in any NMS stock in
an increment less than $0.01.'').
\32\ See Regulation NMS Adopting Release, supra note 4, at
37551-52 (citing 2004 Regulation NMS Proposing Release at 11165).
\33\ See Regulation NMS Adopting Release, supra note 4, at 37553
(``Even assuming that quoting in sub-penny increments would reduce
spreads, the Commission continues to believe, on balance, that the
costs of sub-penny quoting are not justified by the benefits.'').
---------------------------------------------------------------------------
When the Commission adopted Rule 612 in 2005, it acknowledged that
the markets could evolve over time and shift the balance of the costs
and benefits of the adopted tick size.\34\ Two decades later, the
market has evolved considerably, and amendments to Rule 612 are
necessary to continue to further the objectives of the Exchange Act.
Data analysis shows that stocks with sufficiently narrow bid-ask
spreads would trade better, namely it would be easier and less costly
for investors to transact, if they were allowed to quote at increments
smaller than one penny.\35\ Indeed, for these stocks, the risks of
``stepping ahead'' are lowered while the benefits of greater price
competition from relaxing the ``tick constraint'' are greater.\36\
---------------------------------------------------------------------------
\34\ Id. (``Nevertheless, the Commission acknowledges the
possibility that the balance of costs and benefits could shift in a
limited number of cases or as the markets continue to evolve.'').
\35\ See infra section VII.D.1.
\36\ See infra sections VII.B.2 and VII.D.1.b.
---------------------------------------------------------------------------
2. Proposed and Adopted Amendments
Accordingly, the Commission proposed amendments to Rule 612 to
introduce three minimum pricing increments that were less than $0.01
(i.e., $0.005, $0.002, $0.001) for quotes and orders priced $1.00 or
more for certain NMS stocks based upon each stock's time weighted
average quoted spread (``TWAQS'').\37\ The proposed amendments would
have assigned sub-penny minimum pricing increments to any NMS stock
that had a TWAQS of $0.04 or less. This proposed amendment was designed
to address the issues related to tick-constrained stocks described
above that have arisen since 2005. The Commission also proposed to
impose these minimum pricing increments for trades, subject to certain
exceptions.
---------------------------------------------------------------------------
\37\ See Proposing Release, supra note 11, at 80280.
---------------------------------------------------------------------------
As explained below, in response to commenters, the Commission is
adopting modified amendments to Rule 612 to introduce one minimum
pricing increment that is less than $0.01, i.e., $0.005, for quotes and
orders priced $1.00 or more for NMS stocks that have a TWAQS of $0.015
or less.\38\ The Commission is not adopting a minimum pricing increment
for trades.\39\
---------------------------------------------------------------------------
\38\ See infra section III.C.
\39\ See infra section III.D.
---------------------------------------------------------------------------
B. Rule 610 Fees for Access to Quotations and Transparency of Fees
1. Background
Trading centers \40\ can choose to charge an access fee, or pay a
rebate, to the participants--liquidity providers (market participants
with orders resting at the trading center) and liquidity takers (market
participants who submit incoming orders to execute against orders
resting at the trading center)--who trade at their venue. As discussed
in section VII.C.2.b, the predominant exchange fee structure is maker-
taker, in which an exchange charges a fee to liquidity takers and pays
a rebate to liquidity providers, and the rebate is typically funded
through the access fee.\41\
---------------------------------------------------------------------------
\40\ 17 CFR 242.600(b)(106) (providing a definition of the term
``trading center''). This discussion focuses on exchange fees
because, currently, exchanges are the only trading centers that have
quotations that are subject to the access fee caps under Rule
610(c). See infra note 367.
\41\ See also infra sections VII.B.3 and VII.C.2.c, table 5 and
table 6 (showing the predominance of both dollar and share exchange
trading volume occurs on maker-taker venues).
---------------------------------------------------------------------------
As adopted in 2005, Rule 610(c) set the access fee cap for
protected quotations \42\ priced at $1 or more at 30 cents per 100
shares (``30 mils'' per share) for NMS stocks. Rule 610(c) also applies
to any other quotation of a trading center that is the best bid or
offer of an exchange or association.\43\ The access fee cap was based,
in part, upon the prevailing fees that were charged by certain trading
centers at that time.\44\ For NMS stocks priced below $1, the fee cap
was set at 0.3% of the quotation price.\45\ Rule 610 was adopted at the
same time as Rule 611, the Order Protection Rule, which established
intermarket protection
[[Page 81624]]
against trade-throughs \46\ for all NMS stocks. Rule 610(c) was
designed to preclude trading centers that posted protected quotations
from raising their fees in an attempt to take improper advantage of the
trade-through protections adopted under Rule 611.\47\ The Commission
designed the access fee caps to preserve the benefits of both the
strengthened price protection under Rule 611 and the more efficient
linkages among trading centers that were developed under Regulation NMS
to access protected quotations because the benefits could be
compromised if substantial fees were charged.\48\
---------------------------------------------------------------------------
\42\ 17 CFR 242.610(c). A protected quotation is defined in Rule
600(b)(82) as ``a protected bid or protected offer.'' 17 CFR
242.600(b)(82). A protected bid or protected offer is defined as ``a
quotation in an NMS stock that: (i) Is displayed by an automated
trading center; (ii) Is disseminated pursuant to an effective
national market system plan; and (iii) Is an automated quotation
that is the best bid or best offer of a national securities
exchange, or the best bid or best offer of a national securities
association.'' 17 CFR 242.600(b)(81).
\43\ For purposes of this discussion, references to protected
quotations under Rule 610(c) also include manual quotations that are
the best bid or best offer of an exchange or association.
\44\ See Regulation NMS Adopting Release, supra note 4, at
37545.
\45\ See Regulation NMS Adopting Release, supra note 4, at 37544
n.406.
\46\ A trade-through occurs when a trading center executes an
order at a price that is inferior to the price of a protected
quotation that is displayed by another trading center. See 17 CFR
242.600(b)(105) for the definition of trade-through under Regulation
NMS.
\47\ See Regulation NMS Adopting Release, supra note 4, at 37544
and 37595.
\48\ See Regulation NMS Adopting Release, supra note 4, at
37544.
---------------------------------------------------------------------------
Since an access fee that is too high when compared to the tick size
can create pricing distortions, the access fee caps need to be adjusted
in conjunction with the reduction in tick size to prevent such
distortions.\49\ In addition, as discussed below, many exchanges charge
the maximum fee allowed to access protected quotes, and primarily use
those fees to pay rebates to market participants that provide
liquidity.\50\ This practice raises a number of concerns and may
interfere with section 11A's objectives of ensuring the fairness and
usefulness of quotation information.\51\
---------------------------------------------------------------------------
\49\ See infra sections IV.D.1 and VII.D.2.a. See also Proposing
Release, supra note 11, at 80348 (stating ``the access fee cap
should not be greater than \1/2\ of the tick size in order to
preserve coherence between net and nominal price rankings of trading
venues.'').
\50\ See infra sections VII.B.3 and VII.C.2.
\51\ See infra sections IV.D.1 and VII.B.3. See also Regulation
NMS Adopting Release, supra note 4, at 37545 (``For quotations to be
fair and useful, there must be some limit on the extent to which the
true price for those who access quotations can vary from the
displayed price.'').
---------------------------------------------------------------------------
First, the actual prices, inclusive of fees and rebates, for
investors and other market participants to trade a stock are not fully
transparent. In general, the higher the permitted level of access fees,
the higher the rebates, and the greater the potential discrepancy
between displayed quoted prices on the one hand, and actual prices on
the other.\52\
---------------------------------------------------------------------------
\52\ See infra sections IV.B.2, VII.D.2, and VII.E.1. In certain
cases, the disparity between market quotations and actual
transaction costs may be substantial. See, e.g., Proposing Release,
supra note 11, at 80328.
---------------------------------------------------------------------------
Furthermore, exchanges' use of fees and rebates creates a potential
conflict of interest between broker-dealers and their customers with
respect to broker-dealer order routing, by providing incentives for a
broker-dealer to route customer orders to certain exchanges to receive
higher rebates or avoid higher fees based on their own economic
interest.\53\ This potential conflict of interest is exacerbated if
broker-dealers do not fully pass on the fees and rebate to their
customers, since rebate-seeking by broker-dealers may come at the cost
of execution quality of customers.\54\ In addition, exchanges use
complex fee schedules. Generally, the higher the access fee cap, the
wider the range of possible fees and rebates, which results in more
complex pricing schedules. Such complexity makes it more costly for
market participants to design and implement order execution strategies.
---------------------------------------------------------------------------
\53\ See infra sections IV.B.2, IV.D and VII.D.3.
\54\ See text accompanying infra note 1518.
---------------------------------------------------------------------------
Finally, exchanges' fee and rebate schedules are typically
calculated at month's end, which requires market participants to make
trading decisions without the ability to determine their full trading
costs at the time of execution.\55\ In turn, this lack of transparency
impedes a market participant's ability to evaluate fully where to send
its orders because the market participant cannot calculate the fees and
rebates that will apply to the order contemporaneous with
execution.\56\ Concerns with such lack of price transparency are
exacerbated when various exchanges have different fee schedules, as it
is difficult for market participants to compare net prices across
markets.
---------------------------------------------------------------------------
\55\ See infra sections IV.E, VII.C.2, and VII.D.3.
\56\ See infra sections IV.E, VII.C.2, and VII.D.3.
---------------------------------------------------------------------------
2. Proposed and Adopted Amendments
Accordingly, the Commission proposed to amend Rule 610 in two ways.
First, to accommodate the proposed smaller minimum pricing increments
under proposed Rule 612, as well as to address the distortions that
have developed under the access fee caps, the Commission proposed to
reduce Rule 610(c)'s 30 mil cap for executions against protected
quotations priced $1.00 or more as follows: a $0.001 (or 10 mils)
access fee cap for NMS stocks that would have been assigned a minimum
pricing increment larger than $0.001; and a $0.0005 (or 5 mils) access
fee cap for NMS stocks that would have been assigned a $0.001 minimum
pricing increment. For protected quotations in NMS stocks priced under
$1.00 per share, the Commission proposed to reduce the 0.3% fee cap to
0.05% of the quotation price.
As discussed in detail below, in response to comments, the
Commission is adopting amendments to Rule 610(c) with modifications
from the proposal. Specifically, in light of the amendments to Rule
612, the Commission is adopting only the proposed 10 mil per share
access fee cap for all protected quotations priced $1.00 or more.\57\
For protected quotations priced less than $1.00, the Commission is
adopting an access fee cap of 0.1% of the quotation price per
share.\58\ As discussed in section VII.D.2.b, the adopted amendments to
the access fee caps will not impede the ability of exchanges to fund
their execution services.
---------------------------------------------------------------------------
\57\ See infra section IV.D.
\58\ See id.
---------------------------------------------------------------------------
Second, to facilitate the ability of market participants to
understand and calculate the total price of transactions at the time of
execution, the Commission proposed an amendment to Rule 610 to add
subpart (d) to require that all exchange fees charged, and rebates
paid, for the execution of an order in an NMS stock be determinable at
the time of execution. As discussed in detail below, the Commission is
adopting Rule 610(d) as proposed.
C. Transparency of Better Priced Orders
1. Background
The widespread availability of timely information with respect to
quotations for and transactions in NMS stocks (``NMS information'') is
critical to the ability of market participants to participate
effectively in the U.S. securities markets.\59\ NMS information is
currently disseminated within the national market system by the
exclusive plan processors (``exclusive securities information
processors'' or ``SIPs'').\60\
---------------------------------------------------------------------------
\59\ NMS information is made widely available to investors
through the national market system and ``serves an essential linkage
function by helping to assure that the public is aware of the best
displayed prices for a stock, no matter where they may arise in the
national market system.'' See Securities Exchange Act Release No.
61358 (Jan. 14, 2010), 75 FR 3594 (Jan. 21, 2010) (``Concept Release
on Equity Market Structure'') at 3600. The availability of NMS
information also ``enables investors to monitor the prices at which
their orders are executed and assess whether their orders received
best execution.'' Id.
\60\ There are three effective national market system plans that
govern the collection, consolidation, processing and dissemination
of quotation and transaction information for NMS stocks: the
Consolidated Tape Association Plan (``CTA Plan''); the Consolidated
Quotation Plan (``CQ Plan''); and the Joint Self-Regulatory
Organization Plan Governing the Collection, Consolidation, and
Dissemination of Quotation and Transaction Information for Nasdaq-
Listed Securities Traded on Exchanges on an Unlisted Trading
Privileges Basis (``UTP Plan'') (together the ``Equity Data
Plans''). Currently, the Securities Industry Automation Corporation
(``SIAC,'' an affiliate of the NYSE) is the exclusive SIP for the
CTA and CQ Plans, and Nasdaq is the exclusive SIP for the UTP Plan.
See MDI Adopting Release, supra note 10, at 18728. Each exclusive
SIP is the plan processor for one of the Equity Data Plans.
---------------------------------------------------------------------------
[[Page 81625]]
In 2020, the Commission adopted amendments to Regulation NMS to
modernize the NMS information provided within the national market
system for the benefit of market participants and to better achieve
section 11A's goals of assuring ``the availability to brokers, dealers,
and investors of information with respect to quotations for and
transactions in securities that is prompt, accurate, reliable, and
fair'' (``MDI Rules'').\61\ In light of delays in the implementation of
the MDI Rules, the Commission is accelerating the implementation of the
round lot and odd-lot information definitions adopted as part of the
MDI Rules so that investors will benefit sooner from greater
transparency and accessibility of better priced orders \62\ and
improved ability to assess the execution quality of their orders, as
explained below.\63\
---------------------------------------------------------------------------
\61\ See MDI Adopting Release, supra note 10.
\62\ ``Better priced orders'' refers to orders that are priced
superior to the national best bid and national best offer but are
not included in NMS information because they consist of too few
shares. See infra notes 66-68 and accompanying text. The MDI Rules'
round lot and odd-lot information definitions will allow better
priced orders to be included in NMS information so that market
participants that subscribe to the exclusive SIP feeds (that
otherwise would not be able to view these orders without purchasing
exchange proprietary feeds) will be able to view and access these
orders.
\63\ See infra sections V.C.1.a and VII.D.4.
---------------------------------------------------------------------------
Until the full implementation of the MDI Rules, NMS information
disseminated within the national market system by the exclusive SIPs
includes, for each NMS stock, the price, size, and exchange of each
last sale, each exchange's current highest bid and lowest offer and the
shares available at those prices (the best bid and best offer or
``BBO''), the national best bid and national best offer (``NBBO''),
odd-lot \64\ transaction information, and certain regulatory and
administrative data (``SIP data'').\65\ Information on NMS stock
quotations is provided in round lots, and, until the round lot
definition adopted in the MDI Rules is implemented, round lots are
defined in rules of the exchanges.\66\ For most NMS stocks, exchange
rules define a round lot as 100 shares.\67\ Market participants
interested in quotation data for orders that have a size less than a
round lot, i.e., odd-lots, must purchase individual exchange
proprietary feeds.\68\ This odd-lot order information is highly
relevant to market participants, including for investors who trade
small numbers of shares.
---------------------------------------------------------------------------
\64\ Odd-lot is defined in Rule 600(b)(68) as an order for the
purchase or sale of an NMS stock in an amount less than a round lot.
17 CFR 242.600(b)(68).
\65\ See Proposing Release, supra note 11, at 80294. Under the
decentralized consolidation model established by the MDI Rules, NMS
information will consist of ``consolidated market data,'' as defined
in the MDI Rules. 17 CFR 242.600(b)(24).
\66\ See Proposing Release, supra note 11, at 80294 n.328. A
``round lot'' is not defined in the Exchange Act and, prior to the
MDI Rules, it was not defined in Regulation NMS. Exchange rules
typically define a round lot as 100 shares, but they also allow the
exchange, or the primary listing exchange for the stock, discretion
to define it otherwise. See, e.g., NYSE Rule 7.5 (``A `round lot' is
100 shares, unless specified by the primary listing market to be
fewer than 100 shares.'').
\67\ According to NYSE Trade and Quote (``TAQ'') Data, as of
Nov. 28, 2023, 11 NMS stocks have a round lot size other than 100.
Nine NMS stocks have a round lot size of 10 and two NMS stocks have
a round lot size of one share.
\68\ See Proposing Release, supra note 11, at 80294; MDI
Adopting Release, supra note 10, at 18599.
---------------------------------------------------------------------------
The MDI Rules expanded the NMS information that will be made
available for dissemination within the national market system in order
to increase transparency about better prices available in the
market.\69\ The Commission, in the MDI Rules, amended Regulation NMS to
include a definition of ``round lot'' that assigns each NMS stock to a
round lot size based on the stock's average closing price. The round
lot definition, once implemented, will increase transparency about
smaller sized orders in higher priced stocks by assigning NMS stocks
priced over $250 to round lot sizes that are less than the predominant
100 shares.\70\ The Commission also adopted a definition of odd-lot
information as part of the MDI Rules.\71\ Once implemented, information
regarding the prices and sizes of odd-lot orders priced better than the
NBBO will be made available within the national market system and is
expected to be made widely available to investors.\72\
---------------------------------------------------------------------------
\69\ See Proposing Release, supra note 11, at 80270.
\70\ 17 CFR 242.600(b)(93). In the MDI Adopting Release, the
Commission stated that ``[d]efining smaller-sized orders in higher-
priced stocks as round lots, in addition to providing transparency
into such quotations, ensures that these smaller-sized orders can
establish the [national best bid and national best offer], receive
order protection, and invoke the applicability of several other
rules under Regulation NMS.'' See MDI Adopting Release, supra note
10, at 18613.
\71\ Preexisting 17 CFR 242.600(b)(69). ``Odd-lot information''
is defined as (1) odd-lot transactions, and (2) odd-lots at a price
greater than or equal to the national best bid and less than or
equal to the national best offer, aggregated at each price level at
each national securities exchange and national securities
association. Id.
\72\ The Commission stated that the inclusion of this odd-lot
quotation information would allow market participants ``to trade in
a more informed and effective manner,'' and that ``the new
definition of round lot and the increased availability of better
priced odd-lot information will provide investors with valuable
information about the best prices available and help to facilitate
more informed order routing decisions and the best execution of
investor orders.'' See MDI Adopting Release, supra note 10, at 18602
and 18613. Unlike orders in the round lot sizes adopted pursuant to
the MDI Rules, odd-lots are not ``protected quotations.'' See 17 CFR
242.600(b)(16), (81), (82).
---------------------------------------------------------------------------
For the reasons explained in the MDI Adopting Release, the MDI
Rules sequenced the implementation of these definitions in the later
stages of the implementation schedule.\73\ The implementation of the
MDI Rules began with the filing of amendments to the effective national
market system plan(s) as required under Rule 614(e) (``MDI Plan
Amendments'').\74\ The Operating Committees of the CTA/CQ Plan and UTP
Plan \75\ filed the proposed MDI Plan Amendments on November 5,
2021,\76\ which the Commission disapproved.\77\ As a result, the
participants to the effective national market system plan(s) will need
to develop and file new proposed amendments pursuant to Rule 608.\78\
---------------------------------------------------------------------------
\73\ See MDI Adopting Release, supra note 10, at 18698. Pursuant
to the implementation schedule of the MDI Rules, the round lot
definition was set to be implemented as part of the last phase and
odd-lot quotation information was set to be implemented during a
``parallel operation period.'' See id. at 18700-01. As originally
adopted, during the parallel operation period, the exclusive SIPs
would have continued to disseminate the data that they currently
disseminate and competing consolidators would have been permitted to
offer consolidated market data products, including odd-lot
information. Because the round lot definition would have been
implemented during a later phase, the exclusive SIPs and competing
consolidators would have collected, consolidated and disseminated
NMS information based on then current exchange definitions of round
lot. Id. at 18699-18701.
\74\ 17 CFR 242.614(e). The Commission's approval of amendments
to the effective national market system plan(s) filed pursuant to
rule 614(e) will be the starting point for the rest of the MDI Rules
implementation schedule, which includes a 180-day development
period, during which competing consolidators can register with the
Commission, and ends with the cessation of the operations of the
exclusive SIPs and testing and implementation of the changes
necessary to implement the round lot definition. See MDI Adopting
Release, supra note 10, at 18699-701; Proposing Release, supra note
11, at 80295.
\75\ See supra note 60.
\76\ See Securities Exchange Act Release Nos. 93615 (Nov. 19,
2021), 86 FR 67800 (Nov. 29, 2021); 93625 (Nov. 19, 2021), 86 FR
67517 (Nov. 26, 2021); 93620 (Nov. 19, 2021), 86 FR 67541 (Nov. 26,
2021); 93618 (Nov. 19, 2021), 86 FR 67562 (Nov. 26, 2021).
\77\ See Securities Exchange Act Release Nos. 95848 (Sept. 21,
2022), 87 FR 58544 (Sept. 27, 2022); 95849 (Sept. 21, 2022), 87 FR
58592 (Sept. 27, 2022); 95850 (Sept. 21, 2022), 87 FR 58560 (Sept.
27, 2022); 95851 (Sept. 21, 2022), 87 FR 58613 (Sept. 27, 2022).
\78\ On Sept. 1, 2023, the Commission ordered the exchanges and
the Financial Industry Regulatory Authority, Inc. (``FINRA'') to
file a new single national market system plan regarding consolidated
equity market data. See Securities Exchange Act Release No. 98271,
88 FR 61630 (Sept. 7, 2023). On Jan. 19, 2024, the Commission
published notice of filing of a National Market System Plan for
Consolidated Equity Market Data. See Securities Exchange Act Release
No. 99403, 89 FR 5002 (Jan. 25, 2024). On April 23, 2024, the
Commission instituted proceedings pursuant to Rule 608(b)(2)(i) of
Regulation NMS to determine whether to approve or disapprove the
proposed plan or to approve the proposed plan with any changes or
subject to any conditions the Commission deems necessary or
appropriate after considering public comment. See Securities
Exchange Act Release No. 100017, 89 FR 33412 (Apr. 29, 2024). On
July 11, 2024, the Commission extended the period within which to
conclude proceedings regarding the proposed plan to 240 days from
the date of publication of the notice. See Securities Exchange Act
Release No. 100500 (Jul. 11, 2024), 89 FR 58235 (Jul. 17, 2024).
---------------------------------------------------------------------------
[[Page 81626]]
2. Proposed and Adopted Amendments
In light of the delays in the implementation of the MDI Rules, the
Commission proposed to accelerate the implementation of the round lot
and odd-lot information definitions, to allow investors to benefit
sooner from greater transparency and accessibility of better priced
orders and improved execution quality.\79\ As discussed further below,
the Commission is accelerating the implementation of the round lot and
odd-lot information definitions but is providing the industry with more
time to make the necessary systems changes to implement these
definitions than what was proposed.\80\
---------------------------------------------------------------------------
\79\ See Proposing Release, supra note 11, at 80299; see also
infra sections V.B.2. and V.C.1. In addition, as discussed below,
the Commission is amending the definition of round lot so that the
frequency of round lot changes will be consistent with the frequency
of minimum pricing increment changes under amended Rule 612. See
infra section V.B.3.b. The Commission is not changing the
calculation used to assign round lots or the round lot tiers in the
round lot definition adopted in the MDI Rules.
\80\ See infra sections V.B.3, V.C.1, and VI.C.
---------------------------------------------------------------------------
Additionally, the Commission proposed to amend the definition of
odd-lot information to include a new data element for the best odd-lot
orders available in the market, which would be made available to
investors broadly. The Commission is adopting the best odd-lot order to
buy and the best odd-lot order to sell as part of odd-lot information
as proposed.\81\
---------------------------------------------------------------------------
\81\ See infra section V.C.2.
---------------------------------------------------------------------------
D. Overarching Comments on the Proposing Release
The Commission received comments from a variety of market
participants on the Proposing Release.\82\
---------------------------------------------------------------------------
\82\ The comment letters on the Proposing Release (File No. 7-
30-33) are available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
---------------------------------------------------------------------------
Many commenters broadly supported the Regulation NMS Proposal.\83\
Two commenters urged the Commission to promptly adopt the Regulation
NMS Proposal.\84\ One commenter urged the Commission to revise and
adopt the Rule 605 Proposal as well as the Regulation NMS Proposal
without delay.\85\ Another commenter suggested that the Commission
prioritize the adoption of the Regulation NMS Proposal \86\ stating
that, of the four EMS Proposals related to equity market structure, the
Regulation NMS Proposal ``is the least controversial and the least
interdependent on the other three, and so is the easiest one for the
Commission to move forward'' \87\ and ``has garnered the most consensus
and support from various market participants.'' \88\
---------------------------------------------------------------------------
\83\ See, e.g., Letters from Mark Rogers dated Mar. 30, 2023
(``I approve of the proposed changes to Regulation NMS''); Omar
Fakhro dated Mar. 28, 2023 (``I as a household investor strongly
support this rule for a better and fair market for EVERYONE'');
Danielle Ball dated Mar. 27, 2023 (``The proposed tick size regime,
variable minimum pricing increment model, and revised round lot
definition are important steps towards promoting fair and
transparent pricing across trading venues.''); Keith Noble dated
Apr. 1, 2023; Chris Miller dated Apr. 1, 2023; Kristen Palmer dated
Apr. 1, 2023; Amanda Kappes dated Apr. 1, 2023; Ian Rohel, dated
Apr. 1, 2023; Riley Hume dated Apr. 1, 2023; Matt Kelleher dated
Apr. 1, 2023; Keagan Wethington dated Mar. 31, 2023; J.W. Verret,
Associate Professor, George Mason University Antonin Scalia Law
School, dated Jan. 12, 2024 (``Verret Letter III'') at 26 (``. . .
the proposed amendments to Reg NMS rules regarding minimum pricing
increments and the proposed reforms to volume/access fees both
support the core principles of free market economics and will lead
to a more competitive, transparent, and efficient market
landscape.''); Eric Budish, Paul G. McDermott Professor of Economics
and Entrepreneurship, The University of Chicago Booth School of
Business, dated Jan. 18, 2024 (``Budish Letter'') at 1 (``. . . this
set of rules changes--primarily, a finer tick-size for tick-
constrained stocks, a lower access fee cap, and harmonization of
pricing increments for on-exchange and off-exchange trading--will
reduce both investors' costs and the overall complexity of U.S.
equity markets.''); Stephen W. Hall, Legal Director and Securities
Specialist, and Brady Williams, Legal Counsel, Better Markets, Inc.,
dated Mar. 31, 2023 (``Better Markets Letter I'') at 8-17; Joseph
Saluzzi, Partner, Themis Trading LLC, dated Mar. 31, 2023 (``Themis
Letter'') at 2-8; John Ramsay, Chief Market Policy Officer,
Investors Exchange LLC, dated Mar. 20, 2023 (``IEX Letter I'');
Letter Type A, of which 22 comments were received; Letter Type C, of
which 5 comments were received; Letter Type D, of which 255 comments
were received; Letter Type E, of which 14 comments were received;
Letter Type G, of which 652 comments were received; Letter Type H,
of which 853 comments were received; Letter Type I, of which 22
comments were received; Letter Type J, of which 15 comments were
received; Letter Type K, of which 22 comments were received; and
Letter Type L, of which 4 comments were received; available at
https://www.sec.gov/comments/s7-30-22/s73022.htm.
\84\ See, e.g., Letters from Tyler Gellasch, President & CEO,
Healthy Markets Association, dated Mar. 31, 2023 (``Healthy Markets
Letter I'') at 28, 31; J. W. Verret, Associate Professor, George
Mason University Antonin Scalia Law School, dated Sept. 20, 2023
(``Verret Letter I'') at 1-2, 4, 5.
\85\ See Healthy Markets Letter I at 28, 31.
\86\ See Verret Letter I at 1.
\87\ Verret Letter I at 1-2.
\88\ See Verret Letter I at 2 (stating that the Regulation NMS
Proposal is supported by ``a wealth of prior work by the Commission
in the form of a pilot tick size study, comments submitted to the
SEC regarding the transaction fee pilot, and numerous roundtables
and proceedings of the SEC's Investor Advisory Committee and SEC's
Equity Market Structure Advisory Committee.'').
---------------------------------------------------------------------------
Some commenters agreed that Rules 610 and 612 should be amended but
recommended that the proposed amendments be modified and that the
Commission consider more modest, incremental changes to minimize the
possibility of unintended consequences and to enable the Commission and
market participants to evaluate the impact of the changes on trading
and execution quality.\89\
---------------------------------------------------------------------------
\89\ See, e.g., infra note 92.
---------------------------------------------------------------------------
The issues related to the amended rules have been considered by the
Commission and market participants for several years.\90\ Further, the
Commission has analyzed data provided by market participants and
conducted its own data analysis to inform the amendments that were
included in the Proposing Release and in this release.\91\ The
Commission has evaluated the national market system and its operation
in light of changes in the market and has sought input from market
participants throughout this process.\92\ After considering the
comments, which are discussed in context below, the Commission is
adopting amendments to these rules with certain modifications from the
Proposing Release.
---------------------------------------------------------------------------
\90\ See, e.g., EMSAC Archives, supra note 4 (Rule 610 was
considered at the EMSAC), see also supra note 4 (discussing the
EMSAC); infra note 362 and accompanying text for a discussion of
previous considerations of Rule 610. For a discussion of previous
considerations of Rule 612, see Proposing Release, supra note 11, at
80272.
\91\ See infra sections V.B.1; V.B.3.b.iv and VII.D.
\92\ See also Proposing Release, supra note 11, at 80272
(discussing considerations of minimum pricing increments since Rule
612 was adopted) and 80287 (discussing considerations of access fee
caps since Rule 610 was adopted). See also IEX Letter I at 5
(describing steps taken by the Commission since the adoption of
Regulation NMS in 2005 to review the impact of Regulation NMS,
including the solicitation of input from stakeholders, further
stating, ``[t]he history shows that the Commission's current
Proposals do not arise in a vacuum. In fact, the Commission has
deliberately considered the views of multiple stakeholders over
years of review, and its current Proposals grow out of and build on
that ongoing review.'').
---------------------------------------------------------------------------
The Commission received several comments that addressed the
interaction between the different individual proposed rule amendments
that made up the Regulation NMS Proposal. One commenter stated that
adopting the proposed changes to the minimum pricing increments in
proposed Rule 612 along with the proposed acceleration of the round lot
definition and the proposed access fee caps in Rule 610 ``would impact
the value of providing liquidity on public markets and consequently
would raise costs for
[[Page 81627]]
investors,'' and urged the Commission to review how these changes would
together impact liquidity.\93\ The Commission has considered the impact
of the amendments on liquidity and does not believe that they will
raise costs for investors.\94\ On the contrary, as discussed further
below, the amendments will enhance the ability of market participants
to price their orders in a competitive manner, reduce the amount of
fees for accessing protected quotations, help to ensure that exchange
fees are knowable when an order is placed and provide transparency
about orders in the market that are priced better than the NBBO. These
changes will enhance the operation of the national market system and
provide significant benefits to investors.
---------------------------------------------------------------------------
\93\ See Letter from Naureen Hassan, President, UBS Americas,
Robert Karofsky, President, UBS Investment Bank, and Suni Harford,
President, UBS Asset Management, dated Mar. 31, 2023 (``UBS
Letter'') at 10. See infra section V.B.3.b.i and section VII.D.4.a
for discussions of the interaction between the round lot definition
and the proposed changes to the minimum pricing increments.
\94\ See infra section VII.D.4.a (explaining that the
interaction of the reduction in tick size and the MDI Rules' round
lot definition would not have a material impact on the NBBO for
affected stocks as such stocks would be exceptionally liquid, which
should protect their NBBO from material deterioration).
---------------------------------------------------------------------------
Another commenter stated that the Regulation NMS Proposal would
increase ``market data costs'' because retail brokers would have to
take in and store an increased amount of market data to comply with the
changing minimum pricing increments, the MDI Rules' round lot
definition, and the odd-lot information requirements and to update
their systems accordingly, and because the exclusive SIPs may cause
third-party data vendors to require additional hardware to support
higher message rates.\95\ As discussed below,\96\ the Commission is
adopting amendments to the minimum pricing increments with
modifications from the proposal, which will lessen the potential costs
identified by the commenter. Specifically, the Commission is adopting
one minimum pricing increment for a smaller universe of NMS stocks than
was proposed and is reducing the frequency of minimum pricing increment
updates from a quarterly to a semiannual basis.\97\ While this
additional minimum pricing increment will likely require market
participants to incur new technology costs to manage the new data,
fewer changes are being adopted than were proposed and these changes
are necessary and justified to address the issues related to
constraints that have developed with the $0.01 minimum pricing
increment.\98\ Further, the costs related to implementing the round lot
definition were considered as part of the MDI Rules and the
acceleration of the timing of implementation does not increase those
costs. Although the Commission is modifying the round lot definition
from the definition adopted in the MDI Rules, the modifications will
reduce ongoing round lot implementation costs because round lots will
be assigned less frequently, i.e., from a monthly basis to a semiannual
basis, which means that systems will have to be updated less
frequently. Synchronizing the dates of the changes to round lots and
minimum pricing increments should also lower ongoing implementation
costs for market participants by potentially decreasing the number of
updates needed for their trading systems.\99\ Finally, the costs
related to implementing the odd-lot information definition were
considered in the Proposing Release.\100\ The adopted amendments, which
will result in fewer systems changes than anticipated in the Proposing
Release, will result in lower implementation costs than were
contemplated in the proposal \101\ and reduce the amount of data
disseminated by the exclusive SIPs and any future competing
consolidators as compared to what was contemplated in the Proposing
Release.
---------------------------------------------------------------------------
\95\ See Letter from Derrick Chan, Head of Equities, Fidelity
Capital Markets, dated Mar. 31, 2023 (``Fidelity Letter'') at 17.
The commenter described ``market data costs'' as those related to
systems changes necessary to implement the new minimum pricing
increments, round lot definition, and odd-lot information
definition.
\96\ See infra section III.C.
\97\ See infra section III.C.7.a; section III.C.8; section
VII.D.1.d and section VII.F.1.c.
\98\ See infra section VII.A; section VII.D.1.c (responding to
comments raising concerns about increased message traffic increasing
costs and stating: ``[t]he Commission recognizes the potential for
these costs articulated by the commenters but, considering
additional information provided by commenters, expects these effects
to be mild--including the effect on CAT costs.'').
\99\ See infra notes 1594-1595 and accompanying text.
\100\ See Proposing Release, supra note 11, at 80334.
\101\ See infra section VII.D.5.
---------------------------------------------------------------------------
One commenter stated that the implementation of various components
of the Proposing Release at or around the same time (specifically
access fees, minimum pricing increments and round lot sizes) could
complicate the Commission's ability to assess the impact of a specific
change and ``whether other consequences will ensue.'' \102\ To the
specific concerns of this commenter, the Commission has carefully
considered the interacting effects of access fees, minimum pricing
increments, and round lot sizes, see section VII. While the Commission
acknowledges that staging amendments may make them easier to study, the
nature of the adopted amendments will still make such study possible,
even if implemented together. Namely, the set of stocks for which the
tick size change applies tends to differ from the set of stocks for
which round lot changes apply.\103\ Access fee changes apply to some
stocks that will not be directly affected by either round lot reform or
tick size changes. Further, staging the amendments would delay the
significant benefits of the amendments.\104\
---------------------------------------------------------------------------
\102\ See Letter from Rich Steiner, Head of Global Market
Structure, RBC Capital Markets, dated Mar. 31, 2023 (``RBC Letter'')
at 2. See also Letter from Nathaniel N. Evarts, Managing Director,
Head of Trading, Americas, and Kimberly Russell, Market Structure
Specialist, Global SPDR Business, State Street Global Advisors,
dated Mar. 30, 2023 (``State Street Letter'') at 5 (suggesting that
the amendments to reduce the access fee caps should be implemented
before the minimum pricing increments to isolate the impact of the
effects) and infra section VII.D.2.c (responding to the State Street
Letter).
\103\ See infra note 801 for analysis identifying only two
stocks that would have qualified for both the tick reduction and a
reduction in the round lot as of Nov. 30, 2023. See also infra
section VII.D.4.a for a discussion of the small overlap of the round
lot definition and the tick size change.
\104\ See, infra, section II. The Commission recognizes that
delaying the rule would likewise delay costs to affected parties.
---------------------------------------------------------------------------
Several commenters suggested implementing the proposed accelerated
implementation of the round lot and odd-lot information definitions so
that the effects of these definitions could inform other proposed
changes.\105\ Other commenters suggested that round lots should be
implemented before the proposed changes to the minimum pricing
increments, so that data based on the MDI Rules' round lots could
inform changes to the minimum pricing increments.\106\
---------------------------------------------------------------------------
\105\ See, e.g., Letters from Jennifer W. Han, Executive Vice
President, Chief Counsel & Head of Global Regulatory Affairs,
Managed Funds Association, dated Mar. 30, 2023 (``MFA Letter'') at
14; Sarah A. Bessin, Deputy General Counsel, and Nhan Nguyen,
Assistant General Counsel, Investment Company Institute, dated Mar.
31, 2023 (``ICI Letter I'') at 2, 7; Gerald O'Reilly, Co-CEO and
Chief Investment Officer, and Ryan Wiley, Global Head of Equity
Trading, Dimensional Fund Advisors LP, dated Mar. 31, 2023
(``Dimensional Letter'') at 2.
\106\ See Letter from Hubert De Jesus, Managing Director, Global
Head of Market Structure and Electronic Trading, and Samantha DeZur,
Managing Director, Global Public Policy Group, BlackRock, Inc.,
dated Mar. 31, 2023 (``BlackRock Letter'') at 17; Dimensional Letter
at 2.
---------------------------------------------------------------------------
While the dissemination of odd-lot information will result in the
display of narrower spreads based on odd-lots, the calculation of the
TWAQS for determining minimum pricing
[[Page 81628]]
increments is based on round lots.\107\ Therefore, odd-lot information
will not have an impact on determining minimum pricing increments under
Rule 612. Further, for the reasons discussed below, the interaction of
the reduction in tick size and the MDI Rules' round lot definition will
likely not have a material impact on the NBBO of affected stocks since
only the most exceptionally liquid stocks would have prices over $250
and a TWAQS equal to or less than $0.015.\108\ Therefore, it is not
necessary to postpone amending the minimum pricing increments until
data is analyzed using the MDI Rules' round lots.
---------------------------------------------------------------------------
\107\ See infra section III.C.7.b.
\108\ See infra section V.B.3.b.i (identifying only two stocks--
both highly liquid--that would have qualified for both a tick
reduction and a reduction in the round lot as of Nov. 30, 2023).
---------------------------------------------------------------------------
In addition, the dissemination of odd-lot information in
conjunction with the MDI Rules' round lot sizes will increase
transparency about better priced orders and therefore should be
implemented within a similar time frame.\109\ Odd-lot information will
be provided for all NMS stocks, not just those NMS stocks that may be
assigned a smaller round lot. As discussed below, the number of NMS
stocks that may be assigned a smaller round lot as of November 30, 2023
is 163 NMS stocks.\110\ Therefore, while the MDI Rules' round lot sizes
will provide transparency about some better priced orders in higher
priced stocks, they will not enhance transparency about those orders
that continue to be defined as odd-lots and will not increase
transparency for NMS stocks priced at $250 or less. This transparency
is important for investors as it will enhance their ability to assess
the current pricing in the market for certain NMS stocks. Therefore,
the odd-lot information definition and the round lot definition each
represents important, but different information that will enhance the
usefulness of quotation information.
---------------------------------------------------------------------------
\109\ See infra section VII.D.4.
\110\ Id.
---------------------------------------------------------------------------
Some commenters recommended implementing the round lot definition
but not the odd-lot information definition,\111\ stating that
implementing odd-lot information would be burdensome on the
industry,\112\ or would delay the implementation of the round lot
definition by increasing the development work needed to be performed by
the industry,\113\ or that implementation of the odd-lot information
definition ``could lead investors to expect prices that are not
available.'' \114\ For the reasons discussed above, the implementation
of both of these definitions is important to enhancing transparency for
investors. The Commission has provided more time for implementing these
data elements to accommodate the systems changes that will be
necessary, therefore lessening implementation and development burdens
on the industry.\115\ Further, as discussed below, market participants
may decide to provide information to their customers about the changes
that are being implemented, such as how to understand the different
prices, and how the changes may impact their order entry requirements.
Investor notification and education can help investors understand the
operation and impact of these data elements.\116\
---------------------------------------------------------------------------
\111\ See, e.g., Letters from Michael Blaugrund, Chief Operating
Officer, NYSE, Jason Clague, Managing Director, Head of Operations,
Charles Schwab & Co., and Joseph Mecane, Head of Execution Services,
Citadel Securities, dated Mar. 6, 2023 (``NYSE, Schwab, and Citadel
Letter'') at 2; Jason Clague, Managing Director, Head of Operations,
Charles Schwab & Co., Inc., dated Mar. 31, 2023 (``Schwab Letter
II'') at 6, 36; Ryan Kwiatkowski, Chairman of the Board, and James
Toes, President & Chief Executive Officer, Security Traders
Association, dated Apr. 3, 2023 (``STA Letter'') at 8; Adam Nunes,
Hudson River Trading LLC, dated Mar. 31, 2023 (``Hudson River
Letter'') at 2; Joanna Mallers, Secretary, FIA Principal Traders
Group, dated Mar. 31, 2023 (``FIA PTG Letter II'') at 4-5; BlackRock
Letter at 12. See also infra section V.C.1.a. for a discussion of
comments received on the accelerated implementation of the odd-lot
information definition.
\112\ See FIA PTG Letter II at 4-5; Hudson River Letter at 2.
\113\ See FIA PTG Letter II at 4-5.
\114\ Schwab Letter II at 36.
\115\ See infra section VI.C.
\116\ See infra section V.B.3.a.
---------------------------------------------------------------------------
II. Equity Market Structure Initiatives and the Regulation NMS Proposal
In December 2022, the Commission issued three other proposals
related to separate aspects of equity market structure and Regulation
NMS.\117\ A number of commenters provided comments on all four EMS
Proposals jointly.\118\ One commenter stated that adoption of the Rule
605 Proposal is not a prerequisite to adoption of the other equity
market structure proposals.\119\ However, some commenters stated that
the Commission should consider an incremental approach and stagger the
implementation of the four EMS Proposals because of the extent to which
the proposed changes could impact the market and investors.\120\ Some
[[Page 81629]]
commenters suggested implementing only some of the proposed equity
market structure changes, such as the Rule 605 Amendments or portions
of the Regulation NMS Proposal.\121\ Some commenters stated that the
Rule 605 Proposal should be implemented first and that data from the
changes implemented in the Rule 605 Proposal should be analyzed to
assess whether the changes proposed in the Regulation NMS Proposal
should be made.\122\ Some commenters stated that, in light of the
Commission's approval of the amendments to rule 605, the Commission
should defer or suspend action on the Regulation NMS Proposal (and the
two remaining EMS Proposals) and re-evaluate whether to proceed after
the amendments to rule 605 have been implemented and the data collected
following implementation has been analyzed.\123\ One commenter
suggested implementing the round lot and odd-lot information
definitions after implementation of the Rule 605 Proposal, and
thereafter pausing to assess the impact of the changes on the
markets.\124\
---------------------------------------------------------------------------
\117\ See Securities Exchange Act Release Nos. 96943 (Dec. 14,
2022), 88 FR 3786 (Jan. 20, 2023) (proposal to amend rule 605 of
Regulation NMS) (``Rule 605 Proposal''); 96945 (Dec. 14, 2022), 88
FR 128 (Jan. 3, 2023) (proposal to adopt a new rule under Regulation
NMS that would enhance competition for the execution of marketable
orders of individual investors) (``OCR Proposal''); and 96946 (Dec.
14, 2022), 88 FR 5440 (Jan. 27, 2023) (proposal to establish
Commission rule-based best execution standards) (``Best Execution
Proposal'') (together, with the Proposing Release, the ``EMS
Proposals''). The Rule 605 Proposal was adopted on Mar. 6, 2024. See
Rule 605 Amendments, supra note 10.
\118\ See, e.g., Letters from Thom Tillis, Bill Hagerty, Mike
Crapo, Cynthia Lummis, and Kevin Cramer, United States Senate, dated
Jan. 20, 2023 (``Tillis et al. Letter''); Ellen Greene, Managing
Director, Equity and Options Market Structure, Securities Industry
and Financial Markets Association, dated Feb. 8, 2023 (``SIFMA
Letter I''); Joanna Mallers, Secretary, FIA Principal Traders Group,
dated Feb. 15, 2023 (``FIA PTG Letter I''); Hope M. Jarkowski,
General Counsel, NYSE Group, Inc., dated Mar. 13, 2023 (``NYSE
Letter I''); John A. Zecca, Executive Vice President, Global Chief
Legal, Risk & Regulatory Officer, Nasdaq, Inc., dated Mar. 30, 2023
(``Nasdaq Letter I''); Stephen John Berger, Managing Director,
Global Head of Government & Regulatory Policy, Citadel Securities,
dated Mar. 31, 2023 (``Citadel Letter I''); Adrian Griffiths, Head
of Market Structure, MEMX LLC, dated Mar. 31, 2023 (``MEMX
Letter''); Mehmet Kinak, Vice President and Global Head of Equity
Trading, and Jonathan Siegel, Vice President and Managing Legal
Counsel (Legislative & Regulatory Affairs), T. Rowe Price
Associates, Inc., dated Mar. 31, 2023 (``T. Rowe Price Letter'');
Bill Foster, French Hill, Henry Cuellar, Bill Huizenga, Wiley
Nickel, Andy Barr, Ritchie Torres, Ann Wagner, Brittany Pettersen,
Dan Meuser, Josh Gottheimer, Mike Flood, Vicente Gonzalez, Byron
Donalds, Mike Quigley, Michael V. Lawler, David Scott, Andrew R.
Garbarino, Gregory W. Meeks, Monica De La Cruz, Sean Casten, Scott
Fitzgerald, Bradley S. Schneider, Erin Houchin, Jim Himes, Young
Kim, Steven Horsford, Ralph Norman, Gwen Moore, Tom Emmer, Marc
Veasey, and Zach Nunn, United States House of Representatives, dated
Sept. 26, 2023 (``Foster et al. Letter''). See also Form Letter Type
E, of which 14 comments were received, Form Letter Type F, of which
1,703 comments were received, and Form Letter Type G, of which 652
comments were received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
\119\ See Letter from John Ramsay, Chief Market Policy Officer,
Investors Exchange LLC, dated Oct. 13, 2023 (``IEX Letter III'') at
3-5 (explaining how adoption of the amendments to rule 605 should
not delay adoption of the access fee cap and minimum increment
amendments, and stating, ``the premise that Rule 605 updates must be
a precondition to any other changes looks more like a calculated
stall than an argument for careful, reasoned decision making'').
\120\ See, e.g., T. Rowe Price Letter at 3; BlackRock Letter at
17; FIA PTG Letter II at 2; Dimensional Letter at 1, 3; State Street
Letter at 1-2; Letters from Jameson Schriber, Managing Director,
Goldman Sachs & Co. LLC, dated Mar. 31, 2023 (``Goldman Sachs
Letter'') at 8-9; Kirsten Wegner, Chief Executive Officer, Modern
Markets Initiative, dated Mar. 24, 2023 (``MMI Letter'') at 2;
William Capuzzi, Chief Executive Officer, Apex Fintech Solutions,
Inc., dated Mar. 31, 2023 (``Apex Letter'') at 14, 19; Michael
Markunas, Deputy General Counsel, Chief Compliance Officer, B. Riley
Securities, Inc., dated Mar. 31, 2023 (``B. Riley Letter'') at 1;
Kristen Malinconico, Director, Center for Capital Markets
Competitiveness, U.S. Chamber of Commerce, dated Mar. 31, 2023
(``Chamber of Commerce Letter'') at 2; Ellen Greene, Managing
Director, Equity and Options Market Structure, Securities Industry
and Financial Markets Association, dated Mar. 31, 2023 (``SIFMA
Letter II'') at 2, 22-23; William C. Thum, Managing Director and
Assistant General Counsel, Securities Industry and Financial Markets
Association Asset Management Group, dated Mar. 31, 2023 (``SIFMA AMG
Letter I'') at 2; Peter D. Stutsman, Global Head of Equity Trading,
and Timothy J. Stark, Head of Equity Markets and Transaction
Research, The Capital Group Companies, Inc., dated Mar. 31, 2023
(``Capital Group Letter'') at 2, 5; Ann Wagner, United States House
of Representatives, dated Nov. 28, 2022 (``Wagner Letter'') at 2.
\121\ See, e.g., Letters from Stephen John Berger, Managing
Director, Global Head of Government & Regulatory Policy, Citadel
Securities, dated Mar. 31, 2023 (``Equity Market Structure Citadel
Letter'') at 21; Ellen Greene, Managing Director, Equities & Options
Market Structure, and Joseph Corcoran, Managing Director, Associate
General Counsel, Securities Industry and Financial Markets
Association, dated Aug. 24, 2023 (``SIFMA Letter III'') at 3; Steven
M. Greenbaum, Senior Vice President, General Counsel, TradeStation
Securities, Inc., dated Mar. 30, 2023 (``TradeStation Letter'') at
7; Gregory Davis, Managing Director and Chief Investment Officer,
and Matthew Benchener, Managing Director, Personal Investor, The
Vanguard Group, Inc., dated Mar. 31, 2023 (``Vanguard Letter'') at
2; Michael Camacho, Chief Executive Officer, Wealth Management
Solutions, George C.W. Gatch, Chief Executive Officer, J.P. Morgan
Asset Management, and Jason E. Sippel, Chief Executive Officer, J.P.
Morgan Securities LLC, JPMorgan Chase & Co., dated Mar. 31, 2023
(``JPMorgan Letter'') at 2; Ji[rcaron][iacute] Kr[oacute]l, Deputy
Chief Executive Officer, Global Head of Government Affairs,
Alternative Investment Management Association, dated Mar. 31, 2023
(``AIMA Letter'') at 3; John L. Thornton, Co-Chair, Hal S. Scott,
President, and R. Glenn Hubbard, Co-Chair, Committee on Capital
Market Regulation, dated Mar. 31, 2023 (``CCMR Letter'') at 46;
Douglas A. Cifu, Chief Executive Officer, Virtu Financial, Inc.,
dated Mar. 30, 2023 (``Virtu Letter II'') at 4; Andrew M.
Saperstein, Co-President, Morgan Stanley, dated Mar. 31, 2023
(``Morgan Stanley Letter'') at 2-3, 6 and 7; Steve Quirk, Chief
Brokerage Officer, Robinhood Markets, dated Mar. 31, 2023
(``Robinhood Letter'') at 46; MFA Letter at 14; FIA PTG Letter II at
2, 4, 7; NYSE Letter I at 10-11; SIFMA Letter II at 11, 23; State
Street Letter at 3; Chamber of Commerce Letter at 1; STA Letter at
10-11; T. Rowe Price Letter at 3; Verret Letter I at 1, 5, 11; MMI
Letter at 2-3; BlackRock Letter at 17; Capital Group Letter at 5;
UBS Letter at 1-2; Foster et al. Letter at 1, 2; Fidelity Letter at
2, 5.
\122\ See, e.g., Letters from David Howson, Executive Vice
President, Global President, Cboe Global Markets, Nathaniel N.
Evarts, Managing Director, Head of Trading, Americas, State Street
Global Advisors, Kimberly Russell, Market Structure Specialist,
Global SPDR Business, State Street Global Advisors, Mehmet Kinak,
Global Head of Equity Trading, T. Rowe Price, Todd Lopez, Americas
Head of Execution Services, UBS Securities LLC, and Douglas A. Cifu,
Chief Executive Officer, Virtu Financial Inc., dated Mar. 24, 2023
(``Cboe, State Street, et al. Letter'') at 1-2, 3; Michelle Bryan
Oroschakoff, Managing Director, Chief Legal Officer, LPL Financial
LLC, dated Mar. 31, 2023 (``LPL Financial Letter'') at 4; Schwab
Letter II at 6, 37; UBS Letter at 1-2; Apex Letter at 14-15; MFA
Letter at 6; SIFMA Letter II at 11, 22; SIFMA AMG Letter I at 2; T.
Rowe Price Letter at 3; Vanguard Letter at 2, 7; JPMorgan Letter at
2; AIMA Letter at 3; CCMR Letter at 46; UBS Letter at 1-2, 10; Virtu
Letter II at 4; Foster et al. Letter at 1, 2; Capital Group Letter
at 5; Morgan Stanley Letter at 2, 6-7; Fidelity Letter at 2, 5, 27;
Letter from Ann Wagner, Andrew R. Garbarino, Frank D. Lucas, Bill
Huizenga, Tom Emmer, Dan Meuser, Zach Nunn, Pete Sessions, French
Hill, Bryan Steil, Michael V. Lawler, Erin Houchin, United States
House of Representatives, dated June 27, 2024 (``Wagner et al.
Letter''). Some commenters suggested adopting only the Rule 605
Amendments and portions of the Regulation NMS Proposal and then
evaluating the impact of those changes on the market. See Letter
from Melanie Ringold, Head of Legal, Americas, and Will Geyer,
Global Head of Capital Markets, Invesco Ltd., dated Mar. 31, 2023
(``Invesco Letter'') at 2, 5; Hudson River Letter at 1-2;
TradeStation Letter at 7.
\123\ See, e.g., Letters from Barbara Comstock, Executive
Director, American Consumer & Investor Institute, dated May 20, 2024
(``ACII Letter II'') at 1 and 3; Ellen Greene, Managing Director,
Equities & Options Market Structure, SIFMA, and Joseph Corcoran,
Managing Director, Associate General Counsel, SIFMA, dated 14, 2024
(``SIFMA Letter IV''); Ellen Greene, Managing Director, Equities &
Options Market Structure, SIFMA, Joseph Corcoran, Managing Director
and Associate General Counsel, SIFMA, William C. Thum, Managing
Director and Associate General Counsel, dated Aug. 13, 2024 (``SIFMA
AMG Letter II'') at 1-2; Thomas H. Merritt, Deputy General Counsel,
Virtu Financial, Inc., dated June 21, 2024 (``Virtu Letter III'').
See also Letters from Dan Meuser, Ann Wagner, Frank Lucas, Pete
Sessions, Bill Huizenga, French Hill, Andrew Garbarino, Young Kim,
Byron Donalds, Michael V. Lawler, Zach Nunn, United States House of
Representatives, dated June 27, 2024 (``Meuser et al. Letter'') at
2; Michael V. Lawler, United States House of Representatives, dated
July 9, 2024 (``Lawler Letter'') at 1; Wagner et al. Letter at 1-2.
\124\ See State Street Letter at 3.
---------------------------------------------------------------------------
The Commission disagrees with comments urging delayed
implementation of the Regulation NMS Proposal, either in its entirety
or portions of it, as delaying these amendments will delay significant
benefits for investors.\125\ The amendments adopted in this release
revise several provisions of Regulation NMS to benefit investors. The
Commission is adopting amendments to Rule 612 that will benefit
investors and other market participants by allowing certain NMS stocks
to be priced in increments that are smaller than the preexisting rule
allowed, which will lower transaction costs and introduce greater
competition on price into the market. The adopted amendments to Rule
610 will lower costs for investors and other market participants by
reducing the access fee caps and will help to address distortions in
the market associated with the preexisting fee caps. Additionally, the
amendments will require all exchange fees charged and rebates paid for
the execution of an order to be determinable at the time of execution,
allowing investors and other market participants the ability to know
with certainty the costs of their transactions at the time of the trade
and to allow investors to more readily request details about the fees
and rebates applicable to their orders. Accelerating the implementation
of the MDI Rules' round lot and odd-lot information definitions will
provide investors and other market participants that use SIP data with
transparency about better priced quotes and orders that are available
in the market but only visible to subscribers of exchange proprietary
data feeds sooner than originally planned. The amendments provide
important investor benefits, which are discussed throughout. Therefore,
the Commission is not delaying adopting the amendments.
---------------------------------------------------------------------------
\125\ See supra notes 121-124 and accompanying text. See also
supra note 104.
---------------------------------------------------------------------------
With respect to the Rule 605 Amendments, the Commission does not
agree with commenters that stated that amended rule 605 data must be
analyzed before adoption of the changes in this release.\126\ The
amendments adopted in this release are not dependent on rule 605 data
nor is the data from rule 605 reports necessary before the Commission
makes changes to better protect investors and benefit the markets more
broadly.\127\ While the Rule 605 Amendments will bring improvements to
disclosures for order executions of NMS stocks,\128\ the Regulation NMS
amendments address other structural concerns relating to investors'
trading and the lack of transparency in the national market system. For
example, quoted spreads for NMS stocks could not get tighter than $0.01
under preexisting Rule 612 for all quotes and orders in NMS stocks that
were priced equal to, or greater than, $1.00 per share.
---------------------------------------------------------------------------
\126\ See supra note 122.
\127\ Although the amendments adopted in this release are not
dependent on the implementation of the Rule 605 Amendments, the
amendments adopted in this release will enhance the usability of
information in the recently amended rule 605 reports. See infra
section VII.D.6.a.ii.
\128\ See Rule 605 Amendments, supra note 10.
---------------------------------------------------------------------------
The Commission disagrees with the commenter that stated that the
Commission should implement the
[[Page 81630]]
round lot and odd-lot information definitions after the implementation
of the Rule 605 Amendments and then wait to assess the effects of these
changes.\129\ The commenter stated that it supported the round lot and
odd-lot information definitions but stated, without providing details
or any other support, that ``these changes could have unintended
impacts on price discovery, routing complexity, and trading costs.''
\130\ The Commission adopted the definitions in 2020 to provide
transparency about better priced orders that are available in the
market but are not fully transparent in NMS information. These
definitions will result in the provision to market participants of
important information about the prices at which market participants are
willing to trade and therefore will enhance price discovery. Market
participants may have to assess their order routing decisions based on
this enhanced transparency of better priced orders that are available
in the market.\131\
---------------------------------------------------------------------------
\129\ See State Street Letter at 3.
\130\ See State Street Letter at 3.
\131\ See Rule 605 Amendments, supra note 10, at 26482 (stating,
``Rule 605's price improvement statistics that are relative to the
best available displayed price will not be required to be reported
until six months after odd-lot order information needed to calculate
the best available displayed price is made available pursuant to an
effective national market system plan.'').
---------------------------------------------------------------------------
As discussed below, the data analysis performed by the Commission
and other market participants to assess changes in minimum pricing
increments and the access fee caps were not derived from rule 605
reports.\132\ While one commenter stated that rule 605 data should be
used to assess the amendments adopted in this release, the Commission
has utilized relevant and sufficient data other than rule 605 data that
fully and robustly support the amendments.\133\ One commenter states
that if this proposal were to be finalized along with the amendments to
Rule 605, ``it appears that market participants and regulators would be
unable to accurately assess the true impact of the market structure
changes contained in this Proposal, precluding an `apples-to-apples'
before-and-after comparison.'' \134\ However, market participants have
other data with which to analyze the effects of these amendments.
---------------------------------------------------------------------------
\132\ See infra section VII.D.6.a.iii (stating that the
Commission did not rely on rule 605 data in its analyses in the
Proposing Release and in this release).
\133\ See id. The Commission also has considered the interaction
of the compliance dates of the adopted amendments with the
compliance date of the Rule 605 Amendments. See infra section VI;
section VII.D.6.b.
\134\ See Citadel Letter I at 29.
---------------------------------------------------------------------------
Some commenters stated that the EMS Proposals would have an impact
on each other.\135\ Some commenters stated that the EMS Proposals
should have been analyzed together to assess how the proposals would
relate to, and operate with, each other.\136\ One group of members of
Congress recommended that no equity market structure rule ``should be
finalized or implemented'' until the Commission ``[c]onduct[s] a
comprehensive cost-benefit analysis of the aggregate impact of [these
rules] and seek[s] public comment on this analysis[,]'' and the
Commission proposes ``a reasonable, workable, and staggered schedule
for public comment on the adoption and implementation of the proposals,
considering their overlapping nature, significant compliance and
operational burdens, and if they may be insurmountable for smaller or
emerging firms.'' \137\
---------------------------------------------------------------------------
\135\ See, e.g., NYSE, Schwab, and Citadel Letter at 2; STA
Letter at 4, 10-11; T. Rowe Price Letter at 3; RBC Letter at 2 and
5; Nasdaq Letter I at 1, 6; Dimensional Letter at 1-2; FIA PTG
Letter II at 2; Schwab Letter II at 3, 37; Apex Letter at 14-15, 19;
JPMorgan Letter at 2-3; Chamber of Commerce Letter at 2; BlackRock
Letter at 3, 17; MMI Letter at 2-3, 9; B. Riley Letter at 2; Capital
Group Letter at 5; Letters from Ari Rubenstein, CEO, GTS Securities
LLC, dated Mar. 31, 2023 (``GTS Letter'') at 4, 9; Jatin
Suryawanshi, Managing Director, Head of Global Quantitative
Strategies, and Anna Ziotis Kurzrok, Managing Director, Head of
Market Structure, Jefferies, LLC, dated May 2, 2023 (``Jefferies
Letter'') at 1. See also Letters from Patrick McHenry, French Hill,
Frank Lucas, Pete Sessions, Bill Posey, Blaine Luetkemeyer, Bill
Huizenga, Ann Wagner, Andy Barr, Roger Williams, Tom Emmer, Barry
Loudermilk, Alexander X. Mooney, Warren Davidson, John Rose, Bryan
Steil, William Timmons, Ralph Norman, Dan Meuser, Scott Fitzgerald,
Andrew R. Garbarino, Young Kim, Byron Donalds, Mike Flood, Michael
V. Lawler, Zach Nunn, Monica De La Cruz, Erin Houchin, and Andy
Ogles, United States House of Representatives, dated Sept. 26, 2023
(``McHenry et al. Letter'') at 2; Ronald C. Parker, President and
CEO, National Association of Securities Professionals, dated Feb.
28, 2023 (``NASP Letter'') at 4; State Street Letter at 2.
\136\ See, e.g., SIFMA Letter I at 1; SIFMA Letter II at 3, 8-9,
11, 12-13; SIFMA AMG Letter I 4-5; GTS Letter at 4-5; Hudson River
Letter at 1; UBS Letter at 2; NYSE, Schwab, and Citadel Letter at 1;
Citadel Letter I at 2, 28-29; Schwab Letter II at 2-3, 37; Virtu
Letter II at 5, 19-20, 31-35, 55-57; MMI Letter at 2; Nasdaq Letter
I at 6-7; Invesco Letter at 2; Goldman Sachs Letter at 3; Robinhood
Letter at 7, 22, 24, 42, 44; Apex Letter at 14, 15; McHenry et al.
Letter at 1, 2; CCMR Letter at 46; Chamber of Commerce Letter at 3;
Equity Market Structure Citadel Letter at 13-14; Letters from JJ
Kinahan, President, Tastytrade, Inc., dated Mar. 30, 2023
(``Tastytrade Letter'') at 2; Jason Clague, Managing Director, Head
of Operations, Charles Schwab & Co., dated Mar. 22, 2023 (``Schwab
Letter I'') at 2; Eric J. Pan, President and CEO, and Susan Olson,
General Counsel, Investment Company Institute, dated Aug. 17, 2023
(``ICI Letter II'') at 2-3, 7-9; Mary Lou H. Ivey, Chairman of the
Boards and Independent Trustee, David J. Urban, Independent Trustee,
and Theo H. Pitt, Jr., Independent Trustee, Independent Trustees of
ETF Opportunities Trust and World Funds Trust, dated Mar. 31, 2023
(``Independent Trustees Letter'') at 1-2; Stephen John Berger,
Managing Director, Global Head of Government & Regulatory Policy,
Citadel Securities, dated Dec. 5, 2023 (``Citadel Letter II'') at 1,
10; Christopher A. Iacovella, President & Chief Executive Officer,
American Securities Association, dated Mar. 31, 2023 (``ASA
Letter'') at 2, 3; Seth A. Miller, President, Cambridge Investment
Research, Inc. dated Mar. 31, 2023 (``Cambridge Letter'') at 3;
Nicolas Morgan, Founder and President, Investor Choice Advocates
Network, dated Mar. 31, 2023 (``ICAN Letter'') at 2; Rebekah Goshorn
Jurata, General Counsel, American Investment Council, dated Aug. 8,
2023 (``AIC Letter'') at 2, 5, 10; James Angel, Associate Professor
of Finance, Georgetown University, dated Mar. 31, 2023 (``Angel
Letter'') at 2; see also Letter from Jonathan Kanter, Assistant
Attorney General, Doha Mekki, Principal Deputy Assistant Attorney
General, Maggie Goodlander, Deputy Assistant Attorney General, David
Lawrence, Policy Director, Karina Lubell, Chief, Competition Policy
& Advocacy Section, Ihan Kim, Attorney Advisor, Competition Policy &
Advocacy Section, and Owen M. Kendler, Chief, Financial Services,
Fintech & Banking Section, United States Department of Justice,
dated Apr. 11, 2023 (``DOJ Letter'') at 6.
\137\ See McHenry et al. Letter at 2.
---------------------------------------------------------------------------
As discussed below in the economic analysis, the Commission uses as
a baseline the world as it exists at the time of adoption, including
adopted rules but not proposed rules.\138\ Each release, like this
release and the Rule 605 Amendments (which were adopted prior to the
amendments in this release), explains fully the rationale for the
particular rulemaking and includes a robust economic analysis of the
rules being adopted, including the possible economic effects that
commenters raised with regard to specific interactions between the
amendments and the Rule. In addition, comments on how the adoption of
the amendments should affect the timing or sequence of the other EMS
Proposals will be considered if and when those rules are adopted. The
economic analysis considers potential economic effects arising from any
overlap in compliance dates between these amendments and other recent
amendments.\139\ Similarly, the effects of the amended rules are
measured against the existing regulatory baseline, which includes
recently adopted rules.\140\
---------------------------------------------------------------------------
\138\ See infra section VII.C.
\139\ See infra sections VII.C and VII.D.6.
\140\ The OCR Proposal and the Best Execution Proposal Release
mentioned by commenters remain at the proposal stage. To the extent
that the Commission takes final action on either of those proposals,
the baseline in each of those subsequent rulemakings will reflect
the regulatory landscape that is current at that time. See infra
section VII.C, note 1047.
---------------------------------------------------------------------------
Commenting on the Proposing Release together with the other EMS
Proposals, some commenters requested that the Commission publicly
release anonymized subsets of CAT data \141\
[[Page 81631]]
used in connection with the tables and figures in the EMS Proposals'
economic analyses.\142\ In the Proposing Release, unlike certain of the
other EMS Proposals, CAT data was not used in any tables and
figures.\143\ Rather, the Proposing Release used CAT data to determine
the numbers of affected broker-dealers in the baseline and compliance
cost discussion in the economic analysis, as well as to determine
statistics in a reasonable alternative to the proposed amendment that
would have imposed a minimum pricing increment for trades.\144\ The CAT
information used in this adopting release is narrower still.
Specifically, the Commission uses CAT information, consisting of lists
of firm names, including firm identifier numbers and account type
information, only to determine the numbers of affected firms. The
Commission is not releasing anonymized versions of the CAT information
used in this release because releasing an anonymized list of firm names
would provide no meaningful information beyond the total number of
affected firms, which is the same information provided in this release.
The Commission described in the Proposing Release and describes in this
release the CAT data and methodology used in connection with its
estimates.
---------------------------------------------------------------------------
\141\ The CAT database contains confidential market information.
See, e.g., Securities Exchange Act Release No. 67457 (Jul. 18,
2012), 77 FR 45722, 45782 (Aug. 1, 2012) (stating that maintaining
the confidentiality of customer and other information reported to
CAT ``is essential'' and that ``[w]ithout adequate protections,
market participants would risk the exposure of highly-confidential
information about their trading strategies and positions''); see
also Securities Exchange Act Release No. 84696 (Nov. 15, 2016), 81
FR 84696 (Nov. 23, 2016).
\142\ See, e.g., SIFMA Letter I at 1-2, 3-4; Letters from Thomas
M. Merritt, Deputy General Counsel, Virtu Financial, Inc., dated
Feb. 24, 2023 (``Virtu Letter I'') at 1, 2; SIFMA Letter II at 2-3,
11, 22; SIFMA AMG Letter I at 5; Schwab Letter II at 3-4; T. Rowe
Price Letter at 3; Chamber of Commerce Letter at 2-3; Robinhood
Letter at 8; Equity Market Structure Citadel Letter at 16-17;
Cambridge Letter at 4; Jefferies Letter at 1; SIFMA AMG Letter II at
5-7; and SIFMA Letter IV at 6.
\143\ See SIFMA Letter I at 7 (``Regulation NMS: Minimum Pricing
Increments, Access Fees, and Transparency of Better Priced Orders--
The following tables/figures within the Proposal use CAT data:
none.''). The Commission responds to specific comments on releasing
the CAT data used in the tables and figures of the specific EMS
Proposals in the relevant adopting release, where appropriate. See
Rule 605 Amendments, supra note 10.
\144\ See, e.g., Proposing Release, supra note 11, at 80316,
80340-41. A commenter identifies this limited use of CAT data in the
Proposing Release but does not identify specific additional
information the Commission should provide. See Equity Market
Structure Citadel Letter at 16-17.
---------------------------------------------------------------------------
III. Final Rule 612 of Regulation NMS--Minimum Pricing Increment
Rule 612 of Regulation NMS establishes minimum pricing increments
(also known as minimum price variations or tick sizes) for quotations
and orders in NMS stocks. Specifically, preexisting Rule 612 stated
that ``[n]o national securities exchange, national securities
association, alternative trading system, vendor, or broker or dealer
shall display, rank, or accept from any person a bid or offer, an
order, or an indication of interest in any NMS stock priced in an
increment smaller than $0.01 if that bid or offer, order, or indication
of interest is priced equal to, or greater than, $1.00 per share.''
\145\ Preexisting Rule 612(b) had similar language that applied to
bids, offers, orders, and indications of interest in any NMS stock
priced less than $1.00 per share and specified that the minimum pricing
increment could not be smaller than $0.0001. Preexisting Rule 612 of
Regulation NMS did not establish or include minimum pricing increments
for transactions.\146\
---------------------------------------------------------------------------
\145\ See 17 CFR 242.612.
\146\ As discussed in the Proposing Release, the Commission
granted exemptions from Rule 612 to various national securities
exchanges' retail liquidity programs (``RLPs'') as a way to allow
them to compete with over-the-counter (``OTC'') market maker sub-
penny price improvement. See Proposing Release, supra note 11, at
80271. Under the RLPs, exchanges can accept and rank certain quotes
and orders from certain participants in sub-penny increments as
small as $0.001.
---------------------------------------------------------------------------
A. Issues Raised in the Existing Market Structure Related to Tick Sizes
The Proposing Release contains an extensive discussion of the
development and the consideration by the Commission and market
participants of Rule 612 since its adoption.\147\ Since the adoption of
Rule 612, there has been a marked increase in the trading volume of NMS
stocks that would likely be priced with tighter spreads if their
pricing was not constrained by the uniform $0.01 minimum pricing
increment required by preexisting Rule 612 for quotes and orders all
NMS stocks priced equal to, or greater than, $1.00 per share. Easing
constraints on ticks for these NMS stocks will reduce transaction costs
for market participants, including investors, and allow prices to be
determined in a more competitive manner. In other words, the number and
volume of NMS stocks that could benefit from the ability to quote in a
minimum pricing increment that is smaller than $0.01 (i.e., sub-
pennies) has grown.
---------------------------------------------------------------------------
\147\ See Proposing Release, supra note 11, at 80272-80273.
---------------------------------------------------------------------------
In the Proposing Release, the Commission considered data to
evaluate and determine which NMS stocks, by number and by volume, would
benefit from a reduced minimum pricing increment for quotes and orders
that would allow for tighter spreads. While the Commission could not
estimate the number of stocks that would have a TWAQS of $0.008 or less
due to the preexisting Rule 612 requirement that all orders priced
equal to greater than $1.00 per share have a $0.01 minimum pricing
increment, the Commission could estimate that 1,707 stocks, which
represented approximately 64% of share volume and 37.9% of dollar
volume in January through May 2022, had TWAQS that were less than
$0.016.\148\ Additionally, 2,648 stocks, which represented
approximately 17.9% of share volume and 22.3% of dollar volume in
January through May 2022, traded with a spread that was greater than
$0.016 and less than or equal to $0.04. More recently, the Commission
analyzed NMS stocks in 2023 and identified 2,420 NMS stocks that had a
TWAQS of $0.015 or less; these NMS stocks represent about 74% of share
volume and about 47% of dollar volume.\149\
---------------------------------------------------------------------------
\148\ See Proposing Release, supra note 11, at 80280.
\149\ See infra section VII.D.1.b, table 3.
---------------------------------------------------------------------------
Prior to the Proposing Release, certain market participants
conducted data analyses on the effects of Rule 612 and concluded that a
$0.01 minimum quoting increment may not be appropriate for all NMS
stocks that are priced greater than or equal to $1.00.\150\ The
Commission discussed these data analyses in the Proposing Release.\151\
One of these market participants, Cboe, submitted updated data analysis
in two comment letters to the Proposing Release.\152\
---------------------------------------------------------------------------
\150\ See, e.g., The Tick-Constrained Stock Problem by Phil
Mackintosh (Jan. 20, 2022), available at http://www.nasdaq.com/articles/the-tick-constrained-stock-problem) (``Nasdaq Paper''). See
also Petition for Rulemaking to Amend Rule 612 of Regulation NMS to
Adopt Intelligent Tick-Size Regime, dated Dec. 16, 2019, submitted
by John A. Zecca, Executive Vice President, Chief Legal Officer &
Chief Regulatory Officer, Nasdaq Inc. available at https://www.sec.gov/rules/petitions/2019/petn4-756.pdf (``Nasdaq Intelligent
Tick Proposal''); The Impact of Tick-constrained Securities on the
U.S. Equity Market (available at http://www.nyse.com/publicdocs/Tick_Constrained_Stocks.pdf) (``NYSE White Paper'') (no date
available); and Cboe Proposes Tick-Reduction Framework to Ensure
Market Structure Benefits All Investors (available at https://www.cboe.com/insights/posts/cboe-proposes-tick-reduction-framework-to-ensure-market-structure-benefits-all-investors/) (``Cboe
Proposal'').
\151\ See Proposing Release, supra note 11, at 80274-80278.
\152\ See Letters from Angelo Evangelou, Cboe Global Markets,
Inc., dated Feb. 28, 2023 (``Cboe Letter I''); Patrick Sexton, EVP,
General Counsel & Corporate Secretary, Cboe Global Markets, Inc.,
dated Mar. 31, 2023 (``Cboe Letter II'') at Appendix A. See also
Letter from Hope M. Jarkowski, General Counsel, NYSE Group, Inc.,
dated Mar. 27, 2023 (``NYSE Letter II'') (submitting for the record
its paper entitled Price Improvement, tick harmonization & investor
benefit (Aug. 22, 2022). This paper was described in the Proposing
Release, supra note 11, at 80275; MEMX Letter, Appendix (submitting
Tick-constrained Securities (Aug. 2021). This paper was described in
the Proposing Release, supra note 11, at 80274. In the MEMX Letter,
MEMX also submitted Tick-constrained Securities, The Tick Size
Debate, Revisited (Jan. 2022) which analyzed a set of reverse splits
on certain low-priced ProShares exchange-traded products (``ETPs'')
and finding that the tick-constrained ETPs analyzed traded with
significantly lower spreads post reverse split. This paper was
described in the Proposing Release, supra note 11, at 80318.
---------------------------------------------------------------------------
[[Page 81632]]
B. Proposal To Amend Rule 612
The Commission proposed variable minimum pricing increments for
quotes and orders for NMS stocks priced at, or greater than, $1.00 per
share based on the TWAQS of a particular NMS stock. The Commission also
proposed that the minimum pricing increment for executions be the same
as, and correlate to, the minimum pricing increment for quoting on all
trading venues (i.e., on-exchange and OTC), subject to certain
exceptions.
Specifically, the Commission proposed that the minimum pricing
increments for quotations, orders and executions in NMS stocks that are
priced equal to or greater than $1.00 per share would be variable and
no smaller than: (1) $0.001 if the TWAQS \153\ for the NMS stock during
the Evaluation Period \154\ was equal to, or less than, $0.008; (2)
$0.002, if the TWAQS for the NMS stock during the Evaluation Period was
greater than $0.008 but less than, or equal to $0.016; (3) $0.005, if
the TWAQS for the NMS stock during the Evaluation Period was greater
than $0.016 but less than, or equal to, $0.04; and (4) $0.01 if the
TWAQS for the NMS stock during the Evaluation Period was greater than
$0.04.\155\ Further, as proposed, NMS stocks' TWAQS would have been
measured quarterly based on one month of trading data.\156\ In other
words, it was proposed that the assignment of minimum pricing
increments for the quoting and trading of NMS stocks priced equal to or
greater than $1.00 per share be done on a quarterly basis.
---------------------------------------------------------------------------
\153\ See infra section III.C.7.b. See also proposed Rule
612(a).
\154\ See infra section III.C.7.a. See also proposed Rule
612(a).
\155\ See proposed Rule 612(c).
\156\ See proposed Rule 612(a).
---------------------------------------------------------------------------
The Commission stated that it preliminarily believed that the
proposed Rule 612 amendments would promote: (1) fair and orderly
markets and economically efficient executions, particularly for tick-
constrained NMS stocks and retail order flow; and (2) fair competition
and equal regulation between OTC market makers, exchanges, and ATSs
that compete for retail liquidity by requiring that NMS stocks trade
with the same minimum pricing increment regardless of venue (i.e., on
or off-exchange).\157\ The Commission also stated that proposed Rule
612 would promote price discovery and price competition, particularly
for tick-constrained stocks and retail order flow, by permitting the
uniform quoting and trading of NMS stocks across trading venues, in
finer increments, based on objective criteria. The Commission
preliminarily believed that the proposed Rule 612 amendments would
result in the pricing of quotes and orders being more in alignment with
the principles of supply and demand.
---------------------------------------------------------------------------
\157\ See Proposing Release, supra note 11, at 80273 (discussing
the competitive dynamic among exchanges, ATSs and OTC market
makers).
---------------------------------------------------------------------------
C. Final Rule--Minimum Pricing Increments for Orders Priced Equal to or
Greater Than $1.00 per Share
After considering comments, and analyzing additional data in
response to those comments, the Commission is modifying and adopting
the proposed amendments to Rule 612. As adopted, Rule 612(b)(2)
provides that no national securities exchange, national securities
association, ATS, vendor, or broker or dealer shall display, rank, or
accept from any person a bid or offer, an order, or an indication of
interest in any NMS stock in an increment smaller than required
pursuant to either paragraph (i) or (ii) below if that bid or offer,
order, or indication of interest is priced equal to or greater than
$1.00 per share:
(i) $0.01, if the Time Weighted Average Quoted Spread for the NMS
stock during the Evaluation Period was greater than, $0.015; or
(ii) $0.005, if the Time Weighted Average Quoted Spread for the NMS
stock during the Evaluation Period was equal to or less than $0.015.
Rule 612(b)(3) provides that no national securities exchange,
national securities association, alternative trading system, vendor, or
broker or dealer shall display, rank, or accept from any person a bid
or offer, an order, or an indication of interest in any NMS stock
priced in an increment smaller than $0.0001 if that bid or offer,
order, or indication of interest is priced less than $1.00 per
share.\158\
---------------------------------------------------------------------------
\158\ Rule 612(b)(3) is the same as preexisting Rule 612(b).
---------------------------------------------------------------------------
Further, as amended, minimum pricing increments for quotes and
orders will be assigned on a semiannual basis using 3-months of trading
data to calculate each NMS stock's TWAQS.\159\ Therefore, as adopted, a
minimum pricing increment of either $0.01 or $0.005 will be assigned to
each NMS stock for quotes and orders that are priced equal to or
greater than $1.00 per share twice a year and will be operative for a
six-month period.\160\
---------------------------------------------------------------------------
\159\ See Rule 612(a)(1).
\160\ Some commenters suggested that the Commission consider
wider quoting increments. See, e.g., Nasdaq Letter I; ASA Letter at
4; MEMX Letter at 20; Cboe, State Street, et al. Letter at 2; BIO
Letter at 3; Invesco Letter at 3; Robinhood Letter at 39; Themis
Letter at 5; Dimensional Letter at 2; and Letter from Tim Gately,
Managing Director, Head of Equities Sales, Americas, Citigroup
Global Markets, Inc., dated Mar. 31, 2023 (``Citigroup Letter'') at
5. The Commission is not adopting a wider quoting increment for NMS
stocks or a subset of NMS stocks as part of these amendments. As
discussed throughout this release, the Commission is amending Rule
612 to address issues that developed related to the constraint that
results from the $0.01 minimum pricing increment. A wider quoting
increment would not address these specific issues.
---------------------------------------------------------------------------
The amendment differs from the proposal because rather than adding
three proposed smaller minimum pricing increments for quotes and orders
($0.005, $0.002, $0.001) to the current $0.01 increment, only one
additional minimum pricing increment ($0.005) for NMS stocks that have
a TWAQS of $0.015 or less will be added. In addition, the amendment
differs from the proposal as it (1) does not include a minimum pricing
increment for trades, (2) modifies the Evaluation Period, and (3)
provides for an implementation period.
1. General Comments and Discussion
The Commission received many comments on the proposal to amend Rule
612.\161\ Some commenters supported the need to amend Rule 612.\162\
Many individual commenters generally supported the proposed amendments;
\163\ while some individual
[[Page 81633]]
commenters agreed that Rule 612 should be amended but recommended that
the proposal be modified.\164\
---------------------------------------------------------------------------
\161\ See supra note 82.
\162\ See, e.g., Form Letter Type A, of which 22 comments were
received; Form Letter Type D, of which 255 comments were received;
Form Letter Type G, of which 652 comments were received, available
at https://www.sec.gov/comments/s7-30-22/s73022.htm; IEX Letter I at
6; Letters from David Mechner, Chief Executive Officer, Pragma, LLC,
dated Mar. 23, 2023 (``Pragma Letter''); Citigroup Letter at 4; MMI
Letter at 3; Cboe, State Street, et al. Letter at 2; Nasdaq Letter I
at 2; Managed Funds Letter dated March 30, 2023 at 11; letter from
Joseph Scafidi, Global Head of Trading, and Carlos Oliveira, Head of
Trading Analytics and Market Structure, Brandes Investment Partners,
L.P., dated Mar. 23, 2023 (endorsed by Adam Conn, Director, Baillie
Gifford (Overseas) Ltd. et al.) (``Brandes Letter'') at 1; Angel
Letter at 5; TradeStation Letter; Vanguard Letter at 4; B. Riley
Letter at 1; JPMorgan Letter at 4; and UBS Letter at 10.
\163\ See, e.g., Form Letter Type D, of which 255 comments were
received; Form Letter Type E, of which 14 comments were received;
and Form Letter Type G, of which 652 comments were received,
available at https://www.sec.gov/comments/s7-30-22/s73022.htm;
Letter from Bibambop RIP, dated Mar. 16, 2023; Letter from Binh
Tran, dated Mar. 4, 2023; Letter from Jerry Pang, dated Mar. 4,
2023; Letter from Charlie Chen, dated Mar. 1, 2023; Letter from
Daniel Song, dated Jan. 12, 2023; Letter from Deok Park, dated Dec.
26, 2023; and Letter from Clarissa West, dated Apr. 1, 2023.
\164\ See, e.g., Form Letter Type H, of which 853 comments were
received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
---------------------------------------------------------------------------
Broadly, many commenters stated that preexisting Rule 612 should be
amended in order to permit sub-penny quoting.\165\ One commenter stated
that for those stocks that are tick-constrained ``[t]he one-cent
increment for quoting can make it difficult for liquidity providers to
fill orders and often results in higher trading costs.'' \166\ Another
commenter stated that tick-constrained stocks experience wider quoted
spreads, which results in ``significantly increased transaction costs
for investors,'' and that these securities generally have longer queues
and trade with ``outsized notional liquidity at the NBBO.'' \167\
Several commenters stated that the ``one-size-fits-all'' requirement in
Rule 612 should be revisited.\168\ One commenter stated that Rule 612
impedes the ability of market participants to price some NMS stocks
that would naturally be priced within the penny spread.\169\ The
adopted minimum quoting increment of $0.005 will enable the targeted
NMS stocks to be more naturally priced based on the principles of
supply and demand within the penny spread.
---------------------------------------------------------------------------
\165\ See, e.g., Letter from Stephen W. Hall, Legal Director and
Securities Specialist, Better Markets, Inc., dated Oct. 31, 2023
(``Better Markets Letter II'') at 3; SIFMA Letter II; Brandes Letter
at 1; ICI Letter I; BlackRock Letter; B. Riley Securities Letter;
JPMorgan Letter at 4; Cambridge Letter at 6; Invesco Letter at 3;
UBS Letter at 10; Citigroup Letter at 4; TradeStation Letter at 6;
letters from individuals, including the Form Letter Type D, of which
255 comments were received; Form Letter Type G, of which 652
comments were received; and Form Letter Type H, of which 853
comments were received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
\166\ See ASA Letter at 4.
\167\ See MEMX Letter at 9.
\168\ See, e.g., SIFMA Letter II at 33; BlackRock Letter at 5;
Citigroup Letter at 4; and MMI Letter at 5; UBS Letter at 10; Letter
from Lawrence Harris, Ph.D., CFA, Professor of Finance and Business
Economics, U.S.C. Marshall School of Business, dated Dec. 18, 2023
(``Harris Letter'') at 8.
\169\ See Better Markets Letter II at 8.
---------------------------------------------------------------------------
Generally, comments from individuals supported the proposal without
any additional suggested changes.\170\ One commenter stated of the
proposal, ``[t]his means that the pricing of stocks will be more
precise and accurate, ensuring that I can get the best possible price
for my trades.'' \171\ Another commenter stated that ``[a]llowing for
sub-penny pricing will enable buyers to obtain lower prices from
willing sellers and sellers to obtain higher prices from willing
buyers, resulting in a more efficient market.'' \172\ Comments from
other market participants, including exchanges,\173\ broker-dealers,
and institutional investors \174\ recommended modifying the proposal to
Rule 612 to reduce the number of potential minimum quoting increments.
Some commenters stated that further reduction of the minimum pricing
increment for quotes and orders may be warranted for certain NMS stocks
``in the future'' but that a $0.005 increment should be implemented and
studied before any further reductions.\175\ For the reasons discussed
throughout, in response to commenters, the Commission is adopting
amended Rule 612. Compared to the initial proposal, the modified
amendments will be easier for market participants to implement and
adapt to.
---------------------------------------------------------------------------
\170\ See, e.g., Form Letter Type A, of which 22 comments were
received; Form Letter Type D, of which 255 comments were received;
Form Letter Type E, of which 14 comments were received; Form Letter
Type G, of which 652 comments were received; Form Letter Type I, of
which 22 comments were received; Form Letter Type J, of which 15
comments were received; and Form Letter Type K, of which 22 comments
were received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
\171\ Letter from John dated Feb. 23, 2023.
\172\ Letter from Nevin Varghese dated Dec. 26, 2022.
\173\ See IEX Letter I at 6; Cboe, State Street, et al. Letter
at 2; Nasdaq Letter I at 14; MEMX Letter at 18; and Cboe Letter II
at 3.
\174\ See Capital Group Letter at 4; ICI Letter I at 5-6;
Vanguard Letter at 4-5; Invesco Letter at 3; Schwab Letter II at 6;
T. Rowe Price Letter at 4; Fidelity Letter at 14; Brandes Investment
Letter dated March 31, 2023 at 2; Ontario Teachers, Alberta
Investment, CalSTRS, CalPERS, Canada Pension, and Texas Retirement
Letter dated Mar. 31, 2023 at 2 (``Ontario Teachers et al.
Letter''); BlackRock Letter at 5; Dimensional Letter at 2; B. Riley
Letter at 1; and Letter from Christopher P. Bowker Jr., Director of
Global Equity Trading, Boston Partners Global Investors, Inc., Joe
Mariano, Senior Vice President, Global Head of Trading, Calamos
Advisors LLC, Melissa F. Hinmon, Director of Equity Trading,
Glenmede Investment Management, Dan Royal, Global Head of Equity
Trading, Janus Henderson Investors US LLC, dated Apr. 6, 2023
(``Boston Partners, Calamos Advisors, Glenmede Investment, and Janus
Henderson Letter''); State Street Letter at 3; NYSE, Schwab, and
Citadel Letter at 2; Letter from John Zhu, Head of Trading, Optiver
US LLC, dated Mar. 15, 2023 (``Optiver Letter'') at 4; Pragma Letter
at 1; Cboe, State Street, et al. Letter at 2; Letter from Milan
Galik, Chief Executive Officer, Interactive Brokers Group,
Interactive Brokers LLC, dated Mar. 30, 2023 (``Interactive Brokers
Letter'') at 5; RBC Letter at 3; Morgan Stanley Letter at 3-4;
JPMorgan Letter at 4-5; Letter from at 2; Joe Wald, Managing
Director & Co-Head of Electronic Trading, Eric Stockland, Managing
Director, Global Markets, Brad A. Rothbaum, Managing Director & Head
U.S. Global Markets, Chief Operating Officer & Head of the U.S.
Branches, and Michael Forlenza, Managing Director & Head of U.S.
Capital Markets Compliance, BMO Capital Markets Corp., dated Mar.
31, 2023 (``BMO Letter''); Brandes Investment Letter dated March 23,
2023 at 2; B Riley Letter at 1; Themis Letter; UBS Letter at 10;
Citigroup Global Letter at 4-5; and Jefferies Letter at 3.
\175\ See, e.g., BlackRock Letter at 6 and B. Riley Letter at 1.
---------------------------------------------------------------------------
One commenter suggested that the Commission use its exemptive
authority to reduce minimum pricing increments and access fees in a
manner similar to that requested by MEMX.\176\ MEMX requested an
increment of $0.005 for NMS stocks that are ``tick-constrained''
(defined by MEMX as stocks that trade with an average quoted spread of
$0.011 or less).\177\ The commenter recommended this course of action
as a means to gather data on sub-penny pricing increments to help
determine whether, and to what degree, the proposed modifications were
warranted.\178\ The commenter also stated that using an exemption to
test a reduction of minimum pricing increments and the access fee caps
could include an expiration and a ``roll-back'' plan should unintended
consequences become apparent.\179\ Other commenters recommended that
the Commission reduce the minimum pricing increments for a sample of
stocks so that data could be gathered and evaluated before changes were
adopted on a more widespread basis.\180\ Finally, one commenter
recommended that the Commission establish a ``transparent structured
process to evaluate whether proposed changes to minimum pricing
increments and access fees are actually improving the execution
experience'' and that a ``clearly articulated off-ramp/kill-switch to
unwind these changes'' be in place to return to current minimum pricing
increments and the access fee caps.\181\ Another commenter stated that
if the Commission adopted a modified amendment to Rule 612 that such
modification should be re-proposed for public comment.\182\
---------------------------------------------------------------------------
\176\ See Jefferies Letter. See also Proposing Release, supra
note 11, at 80277 for a discussion of the MEMX request for
exemption.
\177\ See Proposing Release, supra note 11, at 80277 for a
discussion of the MEMX request for exemption.
\178\ See Jefferies Letter at 2.
\179\ Id. at 4.
\180\ See, e.g., Cboe, State Street, et al. Letter at 2; letter
from Carlo Passeri, Vice President Biotechnology Innovation
Organization (``BIO Letter''), dated Mar. 30, 2023; and State Street
Letter at 3; MMI Letter at 3-7.
\181\ See Citigroup Letter at 6. With regard to the comment
about an ``off-ramp/kill-switch,'' should the Commission observe
trends detrimental to investors, the Commission could take
appropriate action.
\182\ See Citadel Letter II at 3.
---------------------------------------------------------------------------
An exemption, other temporary course of action, such as a pilot or
sample reduction, or a re-proposal of the
[[Page 81634]]
adopted amendments is not warranted. The Commission and market
participants already have provided data and analyses that support
amending Rule 612 to address tick constraints.\183\ As discussed
throughout this release, the adopted amendments to Rule 612 will allow
NMS stocks that are experiencing tick constraints with the $0.01
minimum pricing increment to be priced more competitively (i.e., reduce
quoted spreads) and reduce transaction costs for liquidity demanders.
The amendments to minimum pricing increments are designed to
appropriately address significant concerns related to Rule 612.\184\
One of the primary goals of the proposal and the adopted amendments is
to alleviate tick constraints.
---------------------------------------------------------------------------
\183\ See, e.g., MEMX Letter, Pragma Letter; IEX Letter I; and
Nasdaq Letter I. See infra section VII.D.1.b.
\184\ See supra section III.A.
---------------------------------------------------------------------------
Reducing the minimum quoting increment for quotes and orders to
$0.005 for certain NMS stocks will enable such stocks to quote with
tighter spreads, which in return reduces the transaction costs of
investors.\185\ As discussed below, the Commission has conducted
analysis to show that quoted and effective spreads are likely to
decline such that costs of executing small and medium trades will
likely decline.\186\ Further, Rule 612, as amended, while simplified
compared to the proposal, continues to be designed to address
constraint concerns with respect to those NMS stocks. Market
participants and investors will be able to more easily adapt to the
amended tick regime because they will only need to accommodate, and
adjust for, one additional minimum pricing increment that is already
familiar for a limited, readily discernable, group of NMS stocks.\187\
The $0.005 minimum pricing increment for quotes and orders, one of the
three additional ticks proposed by the Commission, was widely supported
by commenters.\188\ Price improvement on exchanges and ATSs often
occurs through midpoint executions in an increment of $0.005.
Accordingly, $0.005 is an appropriate increment to introduce smaller,
sub-penny minimum pricing increments in the national market system for
quotes and orders priced equal to or greater than $1.00.
---------------------------------------------------------------------------
\185\ See infra section VII.D.1.b.ii.
\186\ See infra section VII.D.1.b.ii.
\187\ See infra section VII.D.1.a.
\188\ See, e.g., MEMX Letter at 15-16. See also note 219 and
accompanying text.
---------------------------------------------------------------------------
Some individual commenters did not support the proposal.\189\ One
of those commenters stated that the minimum pricing increment for
quotes and orders should be ``based solely on that which can be spent
in real life; no less than a single penny.'' \190\ Preexisting Rule 612
allowed quotes and orders in NMS stocks priced less than $1.00 per
share to be accepted, ranked and displayed in an increment as small as
$0.0001. Similarly, certain RLP Programs for national securities
exchanges have been granted Commission exemptions to permit quotes and
orders in NMS stocks priced equal to, or greater than, $1.00 per share
to be accepted, ranked and displayed in an increment as small as
$0.001. Sub-penny increments also existed in the market for many years,
even prior to the adoption of Rule 612 in 2005.\191\ Sub-penny
increments can allow market participants to better convey prices at
which they are willing to trade, which can promote better price
competition and lead to better price discovery. Further, as discussed
above, sub-penny trading occurs frequently, whether at the midpoint or
in other sub-penny increments.\192\ Thus, sub-penny increments are not
a novel concept. As discussed above, $0.005 is a common trading
increment because of the use of midpoint orders under current Rule 612,
and the ability to use such orders will not change under amended Rule
612. Nonetheless, the Commission understands that market participants
may decide to provide investor notice and education about the
availability of the new increment.\193\
---------------------------------------------------------------------------
\189\ See, e.g., letters from Joshua Russell dated Dec. 27,
2022; Matthew Gayvin Mutman dated Mar. 7, 2023; Aswin Joy dated Mar.
7, 2023.
\190\ See Letter from Joshua Russell dated Dec. 27, 2022. But
see letter from Anonymous dated Apr. 1, 2023 (stating ``[g]etting
more precise increment should be easy enough with our modern
computers. At the gas station I get charged down to the .000th
place, so why shouldn't our markets work the same? Seems fair to
me.'').
\191\ Prior to decimalization, quotes and orders were made in
increments that were fractions of a dollar, including \1/8\, 1/16
and 1/32, which resulted in sub-penny pricing.
\192\ See supra section III.A.
\193\ One commenter stated that to the extent the minimum
quoting increment is reduced, FINRA would need to update the Manning
Rule (FINRA rule 5320 which protects customer limit orders by
requiring a minimum amount of price improvement for a firm to
execute an order on a proprietary basis while holding an unexecuted
customer limit order--the minimum amount of price improvement is
currently $0.01 for orders equal to or greater than $1) in an
equivalent manner. See Citadel Letter I at 8. The compliance date of
the adopted rule provides sufficient time for FINRA to determine
whether it would want to amend the Manning Rule in light of the
amendments to Rule 612 and to file a proposed rule change pursuant
to section 19(b) of the Exchange Act and rule 19b-4 thereunder.
---------------------------------------------------------------------------
Another commenter stated that the proposed variable minimum pricing
increments were ``not an effective solution to address concerns related
to tick-constrained stocks'' and suggested a uniform $0.001 minimum
pricing increment for all NMS stocks.\194\ A uniform $0.001 minimum
pricing increment for all NMS stocks goes beyond what is necessary to
address the issues related to NMS stocks that are currently constrained
by the $0.01 tick. A $0.001 minimum pricing increment would be
significantly smaller than the current uniform $0.01 minimum pricing
increment for quotes and orders for NMS stocks that are priced equal
to, or greater than, $1.00 per share. A sub-penny increment for NMS
stocks that is too small would increase the incidence of stepping ahead
(i.e., pennying) \195\ and costs would not justify the benefits.
---------------------------------------------------------------------------
\194\ See Letter from Matthew Gayvin Mutman dated Mar. 7, 2023.
The commenter suggested a uniform $0.001 minimum pricing increment
for all NMS stocks. Comments related to the level of minimum pricing
increment are addressed in the next section.
\195\ See infra note 994 defining pennying. See also infra
section VII.D.1 for additional discussion of this topic.
---------------------------------------------------------------------------
2. Specific Comments on the Proposed Minimum Pricing Increments
A few commenters did not support the implementation of the smallest
proposed sub-penny increments (i.e., $0.002 and $0.001), and referenced
certain concerns, including stepping ahead of displayed orders, quote
flickering that occurs when the price of a trading center's best
displayed quotations changes multiple times in a single second, and
decreased depth.\196\ Each of these were articulated as concerns by the
Commission when Rule 612 was first adopted.\197\
---------------------------------------------------------------------------
\196\ See, e.g., Form Letter Type G Nasdaq Letter I; MFA Letter;
Letter from Douglas Friedman, General Counsel, Tradeweb Markets
Inc., dated Mar. 30, 2023 (``Tradeweb Markets Letter''); Virtu
Letter II; State Street Letter; RBC Letter; Invesco Letter; ICI
Letter I; Cboe Letter II; SIFMA Letter II; Vanguard Letter; JPMorgan
Letter; Hudson River Letter; T. Rowe Price Letter at 4; Goldman
Sachs Letter; Fidelity Letter; Citadel Letter I; Robinhood Letter;
GTS Letter; BlackRock Letter; Citigroup Letter; Fidelity Letter at
11; Themis Letter at 3; and Tastytrade Letter at 20.
\197\ See Regulation NMS Adopting Release, supra note 4, at
37551.
---------------------------------------------------------------------------
Some commenters stated that having ticks that are too small would
result in queue jumping \198\ and decreased depth.\199\ In the
Regulation NMS Adopting Release, the Commission discussed concerns
related to stepping ahead of displayed quotations with orders priced in
economically insignificant increments (i.e., to gain
[[Page 81635]]
execution priority) which can deter the display of aggressively-priced
limit orders that would narrow the spread.\200\ In light of these
comments, amended Rule 612 has been simplified compared to what was
proposed. Thus, the Commission is only adding the $0.005 minimum
pricing increment for quotes and orders for those NMS stocks that have
a TWAQS of $0.015 or less. Because the $0.005 minimum pricing increment
is based on the TWAQs of the NMS stock, the $0.005 minimum pricing
increment, relative to the spread, will be economically significant for
these stocks.\201\
---------------------------------------------------------------------------
\198\ See, e.g., MFA Letter at 11, State Street Letter at 3, and
RBC Letter at 3.
\199\ See, e.g., Nasdaq Letter I at 13; MFA Letter at 11, Virtu
Letter II at 15, State Street Letter at 3, and RBC Letter at 3.
\200\ See Regulation NMS Adopting Release, supra note 4, at
37551.
\201\ See infra section VII.D.1.b.ii and notes 1300-1303 and
accompanying text.
---------------------------------------------------------------------------
Some commenters stated that smaller tick sizes would cause
flickering quotations.\202\ In the Regulation NMS Adopting Release, the
Commission considered issues related to quote flickering.\203\ The
Commission stated that quote flickering can result in broker-dealers
having difficulties in satisfying their best execution obligations and
other regulatory responsibilities.\204\ Because computer algorithms and
ultra-fast connections dominate today's trading and quoting activities
such concerns are not as acute or prevalent as they were at the time of
the adoption of Rule 612.\205\ Today's quotations are calculated and
displayed in microseconds, which is significantly faster than in 2005
and while flickering quotations can exist today, computer systems are
much better able to process them such that they should not cause
compliance difficulties or investor confusion.\206\ Accordingly,
because of technological advancements, today's market structure,
compared to 2005, can more readily handle rapid changes to a trading
center's best bid or offer. Further, the concerns about the potential
for flickering quotes should be mitigated to some extent because the
amendments do not include the smaller proposed increments (i.e., $0.001
and $0.002) and are designed to have fewer ticks between the spread
which will lessen the potential price changes between the spread.
---------------------------------------------------------------------------
\202\ See, e.g., MFA Letter at 11, State Street Letter at 3, RBC
Letter at 3, and Invesco Letter at 3.
\203\ See Regulation NMS Adopting Release, supra note 4, at
37551.
\204\ Id. at 37552.
\205\ See MDI Adopting Release, supra note 10, for a discussion
about market data latencies. Flickering quotations is more of a
concern when there is quote latency, in other words, when the
displayed quotations do not reflect the actual quotations. For
example, when the quote is being updated faster than the quote can
be displayed, the price discovery mechanism may not be benefitted.
\206\ See Regulation NMS Adopting Release, supra note 4, at
37553-37554 (discussing the concerns with flickering quotes when
Rule 612 was adopted and acknowledging that the market could
evolve).
---------------------------------------------------------------------------
Other commenters stated that the proposed minimum quoting
increments of $0.002 and $0.001 were too small,\207\ would introduce
too many intra-spread ticks,\208\ and could harm trading by
substantially increasing fragmentation of liquidity.\209\ The
Commission also considered the impact of sub-penny quoting on market
depth,\210\ i.e., the number of shares available at the NBBO when it
originally adopted quoting increments.\211\ Decreased depth could lead
to increased transaction costs and fragmentation.\212\ Adopting only
one additional minimum quoting increment instead of the proposed four-
tier approach, should help address commenters' concerns with respect to
fragmented liquidity \213\ because there will be fewer price levels at
which liquidity aggregates, which will result in less fragmentation.
The modified amendment of Rule 612 does not include the proposed
smaller minimum pricing increments for quotes and orders of $0.001 and
$0.002, and thus commenters' concerns related to those increments
(e.g., decreased depth at the NBBO) are not applicable.\214\ As
discussed, the Commission has determined to take an incremental
approach in amending Rule 612 by only adding a $0.005 minimum pricing
increment for those NMS stocks that are constrained by the preexisting,
uniform minimum pricing increment based on an objective standard that
is designed to have fewer ticks between the spread than the
proposal.\215\ As adopted, those NMS stocks that are assigned the
$0.005 minimum pricing increment will result in three ticks intra-
spread, which falls in the middle of the 2 to 4 ticks intra-spread
suggested as potentially optimal by many commenters.\216\ Finally, the
Commission addresses its primary concern of relieving the constraint
related to the $0.01 increment for certain NMS stocks by only adding
the $0.005 minimum pricing increment and not adding minimum pricing
increments of $0.002 and $0.001. The $0.005 minimum pricing increment
for constrained NMS stocks will allow these stocks to quote more
naturally and efficiently, and thereby reduce transaction costs for
investors without the concerns that would attach if the minimum pricing
increments were smaller.
---------------------------------------------------------------------------
\207\ See, e.g., SIFMA Letter II at 33; Vanguard Letter at 5;
Schwab Letter II at 35; Fidelity Letter at 11; JPMorgan Letter at 4;
UBS Letter at 12; Citigroup Letter at 4; and Harris Letter at 7.
\208\ See, e.g., Pragma Letter, Robinhood Letter at 40; IEX
Letter I at 9; and Angel Letter at 6. The adopted $0.005 minimum
pricing increment will provide for at least three ticks intra-
spread. See infra section VII.D.1.
\209\ See, e.g., Interactive Brokers Letter at 4; Virtu Letter
II at 4; and Themis Letter at 3.
\210\ See infra section VII.D.1.b.
\211\ See Regulation NMS Adopting Release, supra note 4, at
37552.
\212\ See Regulation NMS Adopting Release, supra note 4, at
37552.
\213\ See Citadel Letter I at 7. See also Virtu Letter II at 2
and 6-7.
\214\ See infra section VII.D.1.b.i.
\215\ See infra section III.C.6.
\216\ See infra note 1299 and accompanying text.
---------------------------------------------------------------------------
3. Comments on the Number of Proposed Increments
Some commenters supported reducing the minimum pricing increment
for quotes and orders to address those NMS stocks that are tick-
constrained, but overall did not support the proposal's four minimum
quoting increments.\217\ Many commenters stated that the proposed
quoting increments were too numerous.\218\ Instead, a number of
commenters recommended that the Commission adopt a modified, simpler
amendment to Rule 612 and suggested only adopting one additional
minimum quoting increment of $0.005 for tick-constrained NMS
stocks.\219\ One commenter said that ``reducing the tick size to one-
half cent for stocks with narrower spreads will address the current
market need.'' \220\ Commenters opposed the proposed four minimum
quoting increments based on complexity for market participants to
program into their systems these increments,\221\ potential increased
costs for
[[Page 81636]]
investors,\222\ and potential investor confusion with respect to
minimum pricing increments that could change periodically as
proposed.\223\ Another commenter stated that the four-tier proposal
would favor ``high-frequency traders who have a long history of
leveraging complexity to their advantage and to the detriment of
ordinary investors.'' \224\ One commenter stated that the proposed
variable minimum pricing increments ``as small as $0.001 goes well
beyond what is necessary, and would also be cost prohibitive and
complicated to implement.'' \225\ One commenter questioned the impact
of smaller increments on Rule 611 of Regulation NMS and recommended
that if the Commission ``proceed[ed] with their sub-penny quoting
proposal. . . .'', it should consider amending Rule 611 to include all
displayed depth of book quotes.\226\
---------------------------------------------------------------------------
\217\ See, e.g., SIFMA Letter II at 34; AIMA Letter at 2; STA
Letter at 6-7; Citadel Letter I at 30; Citigroup Letter at 4;
Dimensional Letter at 2; BlackRock Letter at 3; Public Pension
Letters dated Mar. 31, 2023; MMI Letter at 3; Brandes Letter at 1;
Schwab Letter II at 35-36; Invesco Letter at 3; B. Riley Letter at
1; JPMorgan Letter at 4; Cambridge Letter at 6; and Tastytrade
Letter at 18.
\218\ See, e.g., MFA Letter at 12; Capital Group Letter at 3;
ICI Letter I ; Angel Letter at 6 ; Vanguard Letter at 5; and Meuser
et al. Letter at 1.
\219\ See id. See also Nasdaq Letter I; MFA Letter; MEMX Letter;
Capital Group Letter; ICI Letter I; Citadel Letter I; Citigroup
Letter at 4; BlackRock Letter; Apex Letter; Ontario Teachers et al.
Letter at 2; Citigroup Letter; GTS Letter; ICI Letter I; Invesco
Letter; Robinhood Letter; SIFMA Letter II; STA Letter; UBS Letter;
Vanguard Letter; TradeStation Letter at 6; Cboe Letter; IEX Letter;
Nasdaq Letter I; and NYSE Letter I; Brandes Letter at 2; Invesco
Letter at 2; Fidelity Letter at 14; Themis Letter at 6; B. Riley
Letter at 1; JPMorgan Letter at 4; Morgan Stanley Letter at 4; State
Street Letter at 3; Dimensional Letter at 2; BMO Capital Letter at
2; and Meuser et al. Letter at 1.
\220\ See ASA Letter at 5. See also TradeStation Letter at 6.
\221\ See, e.g., CTA/UTP Letter dated March 29, 2023; Nasdaq
Letter I; State Street Global Letter; RBC Letter; ICI Letter I;
Vanguard Letter; Cboe Letter II; SIFMA Letter II; Fidelity Letter;
Brandes Letter at 2; Robinhood Letter at 20; Morgan Stanley Letter
at 4; and Meuser et al. Letter at 2.
\222\ See, e.g., Dimensional Letter at 2.
\223\ See, e.g., Tastytrade Letter at 5, 18; SIFMA Letter II at
7; Morgan Stanley Letter at 3, 4; Fidelity Letter at 13; SIFMA
Letter II at 34; Better Markets Letter I at 14; Robinhood Letter at
20; Citadel Letter I at 8; and STA Letter at 5.
\224\ See Better Markets Letter II at 4. See also Fidelity
Letter at 12; Themis Letter at 6; Ontario Teacher et al. Letter at
2; and Harris Letter at 7.
\225\ See TradeStation Letter at 6.
\226\ See Themis Letter at 5. As discussed, the Commission is
adopting a modified amendment to Rule 612 to introduce only a $0.005
minimum pricing increment for certain NMS stocks, not the smaller
proposed increments of $0.002 and $0.001. Therefore, the commenter's
recommendation is no longer germane because without the proposed
smaller $0.002 and $0.001 increments, the liquidity would not be as
dispersed throughout the depth of the book which would not
necessitate protection of the full depth of the book.
---------------------------------------------------------------------------
After considering the comments and analyzing data,\227\ the
Commission is amending Rule 612 to only add one new minimum pricing
increment of $0.005 for those NMS stocks that have a TWAQS of $0.015 or
less, rather than also adopting the additional two $0.002 and $0.001
pricing increments as proposed. The Commission's basis for the new
minimum pricing increment of $0.005 is rooted by the current midpoint
increment when the NBBO is at its narrowest (or smallest) spread. The
midpoint increment of the current $0.01 minimum quoting spread is
calculated as (NBB plus NBO) divided by 2, and when the spread is at
its narrowest, the midpoint increment is equal to $0.005. For example,
if the NBB is 10.01 and the NBO is 10.02, the midpoint would be 10.015
((10.01 + 10.02)/2) = 10.015). Further, the new minimum quoting
increment is at a price level familiar to all market participants and
is already programmed into many computer systems. This modified
approach addresses the concerns raised by commenters related to the
proposed $0.002 and $0.001 minimum pricing increments. The adopted
amendments also address commenters' concerns about complexity and
potentially advantaging certain types of market participants by
reducing the number of new increments and the universe of NMS stocks
that may be eligible for a smaller minimum pricing increment. The
adopted $0.005 minimum pricing increment for those NMS stocks that have
a TWAQS of $0.015 will address the immediate concerns about the
constraints that have developed in the national market system as a
result of preexisting Rule 612.
---------------------------------------------------------------------------
\227\ See infra section VII.D.1.
---------------------------------------------------------------------------
4. Comments on Small- and Mid-Sized Stocks
A few commenters stated that the proposal to reduce minimum pricing
increments did not consider the impact on small and mid-sized
stocks.\228\ One commenter opposed the Regulation NMS Proposal because
of concerns that it did not ``address the needs and possible unintended
consequences for small and mid-sized stocks'' and that the Commission
should ``not take any action until such time as a pilot has been
launched and its effects studied and verified by a committee of market
participants and academics.'' \229\ Another commenter stated that the
proposed tick sizes were ``too granular'' for small to mid-sized stocks
and would result in fewer liquidity providers.\230\
---------------------------------------------------------------------------
\228\ See BIO Letter at 1-2, 3 and STA Letter at 5.
\229\ See BIO Letter at 1-2, 3.
\230\ See STA Letter at 5.
---------------------------------------------------------------------------
The assignment of the smaller minimum pricing increment is not
based on market capitalization because the economics of being tick-
constrained do not depend on market capitalization. Rather, whether a
stock is experiencing constraint depends on its spread. In other words,
since a stock's spread relative to the tick size does not depend on
whether it has a small or mid-sized market capitalization, such a stock
could still trade with a quoted spread constrained by $0.01 minimum
pricing increment. With respect to implementing a pilot program to
assess the needs and potential consequences of the proposal for small
and mid-sized stocks, the Commission previously conducted a tick size
pilot program for small- and mid-sized stocks to assess the impact of
wider minimum quoting and trading increments.\231\ The Commission
analyzed data from that pilot program for purposes of the
amendments.\232\ Another pilot program is not necessary because the
Commission and market participants have demonstrated with data the
issues related to tick constraints that have increased since the
preexisting rule was adopted.\233\ Further, the modified amendment will
not introduce increments that are ``too granular'' for any NMS stock;
only those NMS stocks that have a TWAQS of $0.015 or less will be
assigned the new $0.005 increment, or three ticks or fewer within the
spread. These NMS stocks are constrained by the preexisting increment
and the amendment will alleviate this regulatory constraint to allow
competitive forces of supply and demand to better establish bid and ask
prices.\234\
---------------------------------------------------------------------------
\231\ See Proposing Release, supra note 11, at 80272-73 for a
discussion of the tick size pilot program. See also Tick Sizes and
Market Quality: Revisiting the Tick Size Pilot by Yashar H.
Barardehi, Peter Dixon, Qiyu Liu, and Ariel Lohr, available at
https://www.sec.gov/dera/staff-papers/working-papers/dera_wp_tick-sizes-and-market-qualityrevisiting-tick-size-pilot.
\232\ See infra section VII.D.1.
\233\ See infra section VII.D.1.b.ii.
\234\ See also infra section VII.D.1.b.i and VII.B.2 for
additional discussion.
---------------------------------------------------------------------------
5. Comments on Market Resiliency
A few commenters raised concerns related to market resiliency
risks.\235\ The commenter stated that ``[b]ecause the Commission's
proposal would increase the number of ticks inside the weighted average
spread for many stocks, we could expect a significant increase in
message traffic that would result from the Commission's proposal.''
\236\ The commenter asked the Commission to consider the potential
increased message traffic that could result from the proposed minimum
pricing increments and stated that the proposal would result in a
significant increase in message traffic.\237\ The commenter recommended
the Commission take a measured and phased approach for reducing the
minimum pricing increment for quoting to apply the minimum quoting
increment initially to a limited number of stocks and additional groups
of stocks in subsequent phases, with review of market resiliency during
each phase.
---------------------------------------------------------------------------
\235\ See, e.g., Letter from Howard Meyerson, Managing Director,
Financial Information Forum, dated Mar. 31, 2023 (``FIF Letter'') at
6; and Goldman Sachs Letter at 8.
\236\ See FIF Letter at 7.
\237\ See FIF Letter at 7. See also Robinhood Letter at 41;
Morgan Stanley Letter at 3; UBS Letter at 12; Citigroup Letter at 4;
TradeStation Letter at 7; and Goldman Sachs Letter at 9.
---------------------------------------------------------------------------
The amendments modifying Rule 612 will result in less message
traffic, fewer systems changes and lower costs related to updating
ticks for NMS stocks compared to the original proposal and
[[Page 81637]]
therefore there should pose less of a concern related to market
resiliency. The modified amendment adopts a single sub-penny increment
that impacts a smaller universe of NMS stocks compared to the proposal,
which included three sub-penny increments that would have impacted more
NMS stocks. The need for a phased approach is significantly reduced
because fewer NMS stocks will be impacted by the one additional minimum
quoting increment, and there will be fewer ticks between the spread.
The commenter stated that the potential costs to industry members
from increased message traffic would include purchasing additional
computer hardware such as servers and that the costs would also apply
to production, backup, test, and development environments.\238\ The
commenter stated that the actual costs would be multiples of the
estimated costs from the proposal. However, the adopted amendment to
Rule 612 will result in less message traffic than the proposal because
it has fewer quoting increments. Consequently, the modified amendments
that are being adopted will reduce computer hardware and developmental
costs for the industry compared to the proposal. In the Proposing
Release, the Commission considered the message traffic of the options
markets, and the systems for the options markets that handle many times
more messages compared to (1) the current NMS stock market or (2) the
estimated additional message traffic from the adopted amendments.\239\
One commenter submitted data that supported this conclusion.\240\
---------------------------------------------------------------------------
\238\ See FIF Letter at 9. See also Citigroup Letter at 2. See
infra section VII.D.5.a.
\239\ See Proposing Release, supra note 11, at 80279, notes 196
and 197 (stating that in the second quarter of 2011, the average
peak message per second for Tapes A and B reported by the CTA/CQ
Plan was 1,015,000 and for Tape C reported by the UTP Plan was
408,300 versus 36.4 million reported by the Options Price Reporting
Authority (``OPRA'')). See also section VII.E.1.
\240\ See NYSE Letter I at 11-13.
---------------------------------------------------------------------------
The commenter also raised concerns that increased quote message
traffic could significantly increase the costs of the operation of the
CAT system.\241\ The commenter recommended that the Commission estimate
the potential increase in message traffic, provide those estimates to
CAT LLC, obtain estimates from the CAT LLC of the increased CAT costs
that would result from this increased message traffic, and factor the
estimated costs into the cost benefit analysis of the proposed minimum
pricing increments changes. Another commenter also stated that the
Commission failed to consider whether the increase in message traffic
will increase the CAT operating budget.\242\ The Commission estimates
the impact of the adopted amendments on message traffic, and thus on
the CAT operating budget in section VII.D.1.c. As discussed further
below, the Commission estimates the increase in CAT costs associated
with adopting the additional minimum pricing increment to be
approximately $4.1 million per year.\243\ The Commission does not
believe it is appropriate to delay action on Rule 612 to have CAT LLC
engage in its own analysis of the potential costs.
---------------------------------------------------------------------------
\241\ See FIF Letter at 10 (``FIF members are concerned that
increased message traffic could significantly increase the costs for
the operation of the CAT system as increased quote volumes
(including increased frequency of quote updates) would increase the
number of CAT-reportable events. 100% of these increased CAT costs
would be charged to broker-dealers and exchanges. The operating
expenses for CAT were $84.5 million for 2020 and $146.5 million for
2021. CAT LLC, the operator of the CAT system, has estimated the
total expenditures for CAT for 2022 at $178.9 million. These costs
are in excess of the costs that were contemplated in the CAT NMS
Plan.'').
\242\ See Citadel Letter II at 5. The commenter added that
increased message traffic increases costs for all market
participants, including higher fees charged by CAT and the exclusive
SIPs. See also Citadel Letter I at 9 and Virtu Letter II at 6-7.
\243\ See infra section VII.D.1.c.
---------------------------------------------------------------------------
Commenters raised the issue of increased market data volume on
competing consolidators, which are not yet in operation.\244\ Likewise,
the possible costs to potential competing consolidators will be reduced
vis-[agrave]-vis the proposal. The Commission recognizes that while the
costs may be lower than the proposed rule, the adopted rule could
nevertheless create increased message traffic than the preexisting
rule. It follows that more message traffic could lead to more possible
costs for competing consolidators. However, this new message traffic
should still be within the operational capacity of the existing
computer systems.\245\
---------------------------------------------------------------------------
\244\ See, e.g., Citadel Letter II at 9 (``A material increase
in total message traffic increases costs for all market
participants, including due to the resulting higher fees charged by
industry utilities, such as the [CAT] and the [SIP]'') and Virtu
Letter II at 6-7 (``The Commission has failed to analyze the impact
of the significantly increased volume of market data on competing
consolidators.'').
\245\ See infra section VII.D.1.
---------------------------------------------------------------------------
One commenter stated that even with the largest potential increases
in messages, equity messaging traffic would remain well below that of
the options market and that ``the increase in messaging activity from
adopting finer tick increments is now well within the industry's
capability.'' \246\ On the other hand, another commenter stated that a
larger number of ticks across a large number of stocks would lead to
increased message traffic, which would, in turn, increase data and
infrastructure costs and market latency.\247\ One commenter added that
increased message traffic would lead to increased latency, which would
harm market participants by disrupting trading strategies and impairing
market functionality and liquidity.\248\ As stated above, the adopted
amendment to Rule 612 is significantly less complex than the proposal
and will not result in the larger number of ticks across a large number
of stocks as the commenter suggested. The proposal's four minimum tick
increment has been simplified to one additional new tick at $0.005, and
the proposal's reduction of minimum pricing increments for NMS stocks
that had a TWAQS of $0.04 or less has been reduced to those NMS stocks
that have a TWAQS equal to or less than $0.015, which results in fewer
expected NMS stocks being assigned a smaller minimum pricing
increment.\249\ These adopted changes may result in significantly less
message traffic than under the commenter's assumption on the proposal.
While message traffic may increase over today's message traffic, any
increase in message traffic will be significantly less than in the
options market, and the options market participants have over the years
adjusted to increasingly higher message traffic.\250\
---------------------------------------------------------------------------
\246\ See NYSE Letter II at 11 (stating that OPRA handles many
times more messages than the equity markets).
\247\ See MFA Letter at 11.
\248\ See Tradeweb Letter at 2-3 (``Even trading platforms with
the most advanced technological infrastructure will need to expend
considerable amounts of time and resources to prepare the
accommodate increased message traffic, since any increase in latency
(even at the millisecond level) would disrupt trading strategies,
impair market functionality and liquidity, and, ultimately, harm
market participants.''); see also Virtu Letter II at 6 (``This
increase in message traffic. . . will significantly add to the
overall content of market data.''). See also NYSE Letter I at 6 and
Nasdaq Letter I at 9 (``Securities with too many ticks not only have
wider spreads, but they also have more odd lots, and more message
traffic, leading to a more fragile NBBO.'').
\249\ See infra section VII.D.1.a.
\250\ See Options Clearing Corporation Daily Volume report,
available at https://www.theocc.com/Market-Data/Market-Data-Reports/Volume-and-Open-Interest/Daily-Volume.
---------------------------------------------------------------------------
6. Comments on Proposed Criteria for Assigning Minimum Pricing
Increments
The Commission proposed to measure the TWAQS when determining the
appropriate minimum pricing increment for NMS stocks and proposed four
ranges of the TWAQS to determine the corresponding minimum pricing
increments. The four proposed TWAQS ranges were: (1) equal to or less
than
[[Page 81638]]
$0.008; (2) greater than $0.008 but less than or equal to $0.016; (3)
greater than $0.016 but less than or equal to $0.04; and (4) greater
than $0.04. Preliminarily, the Commission believed that NMS stocks with
a TWAQS of $0.04 or less would have benefited from smaller minimum
pricing increments. After considering the comments, the Commission is
retaining TWAQS as the measure to determine when an NMS stock will be
assigned smaller minimum pricing increment but has modified the
threshold to be equal to or less than $0.015.
Many commenters stated that tick-constrained stocks would benefit
from smaller minimum pricing increments.\251\ Commenters, however,
raised concerns about reducing the minimum pricing increment for NMS
stocks that were not experiencing tick constraint with the $0.01
minimum pricing increment.\252\ One commenter stated that ``a reduction
in tick sizes for those stocks that are merely near-tick-constrained
will not result in meaningful price-improvements and will not be worth
the increased risk of diminished liquidity to bids and offers being
spread too thinly across too many price points.'' \253\ As adopted, the
new $0.005 minimum pricing increment will be assigned to those NMS
stocks that have a TWAQS of $0.015 or less. These NMS stocks are
experiencing constraint with the $0.01 minimum pricing increment and
will benefit from being able to be quoted in the smaller increment. As
adopted, the Commission has modified the amendment so as not to assign
the smaller $0.005 increment to those NMS stocks that are not
necessarily experiencing constraint with the $0.01 minimum pricing
increment.
---------------------------------------------------------------------------
\251\ See, e.g., Pragma Letter at 6 (``tick-constrained stocks
will benefit from smaller tick sizes with narrower spreads.''); MEMX
Letter; NYSE, Schwab, and Citadel Letter; IEX Letter I at 7; Nasdaq
Letter I at 2 (``Nasdaq supports adjusting the minimum pricing
increment (``tick size'') to better reflect the trading dynamics of
Regulation National Market System (``Reg. NMS'') securities.'');
Brandes Letter at 2; Schwab Letter II at 35; and Robinhood Letter at
46.
\252\ See, e.g., IEX Letter I; Pragma Letter; Invesco Letter;
ICI Letter II at 14 (stating that the Commission should not apply
sub-penny increments to stocks that are not tick-constrained); ASA
Letter (``we strongly oppose the application of a one-half cent tick
size to any stock outside of the most liquid (narrower spread)
stocks.''); Nasdaq Letter I at 14 (``We propose that securities fall
into this new $0.005 tick bucket only if they are tick-
constrained.''); and Cboe Letter II.
\253\ See Invesco Letter at 3. See also e.g., ICI Letter II
(stating that there is no market failure or harm identified for
stocks that are not tick-constrained.) and Brandes Letter at 2
(favoring a reduction to $0.005 for those stocks that are
experiencing constraint with the $0.01 increment and stating that
the proposed reduction in a minimum pricing increment for stocks
that had a TWAQS of $0.04 or less was too broad).
---------------------------------------------------------------------------
Some commenters stated that the TWAQS of $0.011 should be used for
identifying NMS stocks that are experiencing tick constraint.\254\
However, one commenter recommended that NMS stocks that ``could easily
become tick-constrained'' should have their minimum pricing increment
reduced.\255\ Other commenters offered other recommendations as to the
TWAQS threshold for reducing minimum pricing increments, including a
TWAQS threshold of $0.02 or less,\256\ a TWAQS threshold of $0.016 or
less,\257\ and a TWAQS threshold of 0.015 or less.\258\
---------------------------------------------------------------------------
\254\ See, e.g., NYSE Letter I; Vanguard Letter; Cboe Letter I;
and Schwab Letter II at 35-36. But see also Invesco Letter at 3
(stating that $0.011 was overly broad and would result in
unnecessary tick reductions for stocks that are not tick-
constrained.).
\255\ See, e.g., IEX Letter I at 7 (``IEX agrees with the
premise that tick sizes should be reduced for stocks that are
currently ``tick-constrained'' or could easily become tick-
constrained because of the current one-cent limitation.'').
\256\ See IEX Letter I at 7, 13 (``We believe that reducing the
tick size and applying it to all securities with a TWAQS up to two
cents will substantially improve the efficiency of displayed trading
. . .'').
\257\ See BMO Capital Letter at 2 and Form Letter Type H, of
which 853 comments were received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
\258\ See Pragma Letter at 6 (``While perhaps not conclusive,
the lines of evidence from our analysis also suggest that the
Proposal's range of 4 to 8 ticks is too many and will force wider
spreads and higher trading costs on the market than necessary. This
leads to our primary recommendation: stocks should be moved to a
smaller tick size only when their average spread is less than \1/5\
in the preceding month; and moved to a larger tick size only when
their spread is greater than 4 ticks in the preceding month.'').
---------------------------------------------------------------------------
As discussed further below, the Commission is adopting the TWAQS
threshold of $0.015 or less in order to identify NMS stocks that will
be eligible for the $0.005 minimum pricing increment.\259\ This
amendment will generally result in these NMS stocks having a bid-ask
spread with one to three ticks, which will improve market quality.\260\
Data analysis supports that liquidity and market quality will improve
if NMS stocks with a TWAQS of $0.015 or less are assigned to the $0.005
minimum pricing increment.\261\ Commenters provided analysis and cited
studies that suggest that 2 to 4 ticks intra-spread is optimal for
trading, which is consistent with the results of the Commission's
analysis.\262\
---------------------------------------------------------------------------
\259\ See infra section VII.D.1.b for more discussion on TWAQS.
\260\ See infra note 1303 and accompanying text. See also Nasdaq
Letter I at 18 (stating that ``quoting outside of the optimal 2-3
tick spreads leads to queues for tick-constrained securities and
slower price formation for securities with overly-wide spreads.'').
\261\ See infra section VII.D.1.b.
\262\ See, e.g., Nasdaq Letter I at 8, 18; Pragma Letter at 1;
RBC Letter at 3; CCMR Letter at 23; Letter from Eric Swanson, Chief
Executive Officer, XTX Markets LLC, dated Mar. 30, 2023 (``XTX
Letter'') at 4; MMI Letter at 5; and Harris Letter at 7. See also
infra notes 1293-1299 and accompanying text.
---------------------------------------------------------------------------
Some commenters agreed that the TWAQS was the appropriate measure
for determining the relevant minimum pricing increment.\263\ Several
commenters stated that as many stocks as possible should be identified
as eligible for a smaller tick size.\264\ Other commenters suggested
that a multi-factor approach be taken in evaluating whether to reduce
the minimum pricing increment for certain NMS stocks.\265\ Commenters
suggested that such factors include average quoted size,\266\ ratio of
[[Page 81639]]
average quoted size to average traded size,\267\ daily traded
volume,\268\ queue length,\269\ quotes on multiple exchanges,\270\ or
stock price.\271\ One commenter recommended the inclusion of factors
such as large quoted displayed size and a relatively high level of
liquidity based on average daily trading volume.\272\
---------------------------------------------------------------------------
\263\ See, e.g., IEX Letter I at 7 (``[w]e agree that TWAQS is a
reasonable and appropriate measure to define which securities should
be subject to a narrower tick size.''); MEMX Letter; and BMO Capital
Letter.
\264\ See, e.g., Form Letter Type K, of which 22 comments were
received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm; Anonymous Letter dated Mar. 6, 2023; letter from Victor
Piousbox dated Mar. 6, 2023; letter from Jimit Raithatha dated Mar.
7, 2023; letter from Munib Mian dated Mar. 7, 2023; letter from
Peter Unum dated Mar. 19, 2023; letter from Anonymous dated Mar. 22,
2023; and letters from Chris and Donna Graves, dated Mar. 26, 2023;
Spencer Neukam dated Mar. 26, 2023; Samuel Cressy dated Mar. 24,
2023; and Zaf Khan dated Mar. 24, 2023.
\265\ See, e.g., Cboe Letter II at 3 (``The most critical step
in any tick-size regime reform is first establishing an objective
methodology designed to address truly tick-constrained securities.
In this regard, we recommend using a multi-factor methodology, such
as Cboe's Tick Size Reduction Framework.''); Themis Letter at 7
(supporting Cboe's methodology); Cboe, State Street, et al. Letter;
Optiver Letter (discussing the European Union's tick regime as
considering stock price and liquidity); NYSE Letter I; ICI Letter II
at 11 (stating that applying other factors would lessen concerns
about an overbroad tick reduction and mitigate concerns about an
adverse market outcome); BlackRock Letter at 5 (stating that quoted
spread is one-dimensional and does not provide sufficient context
for determining the optimal tick size); Citigroup Letter
(recommending a new $0.005 quoting increment for the most liquid
tick-constrained stocks); T. Rowe Price Letter (stating that a
multi-factor approach would allow the Commission to measure whether
a tick size is properly calibrated); STA Letter at 6 (recommending
that the Commission use a multifactor approach); Virtu Letter II at
6 (stating ``one must consider many factors, not just quoted
spread'' and describing methods proposed by Cboe and Nasdaq); NYSE,
Schwab, and Citadel Letter at 1 (``We define `tick-constrained' to
mean symbols that have an average quoted spread of 1.1 cents or less
and a reasonable amount of available liquidity at the NBBO.''); and
Cambridge Letter at 6 (stating that securities should have an
average quoted spread of 1.1 cents and be ``reasonably liquid'').
See also SIFMA Letter II at 36 (``SIFMA believes that a more robust
analysis is necessary to evaluate the most appropriate tick sizes
for purposes of achieving the best balance between available
liquidity at the inside quotation versus narrower spreads.'').
\266\ See, e.g., BlackRock Letter at 6 (`` . . . if material
size was present at the National best Bid and Offer (`NBBO') or a
significant proportion of executions were occurring at sub-penny
prices, this would be a clear indication of fierce order book
competition and interest to tighten the spread and trade in smaller
increments.''); T. Rowe Price Letter; and Citadel Letter I.
\267\ See, e.g., Cboe Letter II at 3 (``we started with the
complete universe of NMS securities, and applied three constraints--
quoted spread, quoted-size-to-trade-size ratio, and notional
turnover ratio--to arrive at a group of securities that are
quantifiably tick-constrained.'') and BlackRock Letter.
\268\ See, e.g., BlackRock Letter at 5 (``BlackRock recommends
that in addition to the time weighted quoted spread, the Commission
should incorporate other factors for designating tick sizes, such as
the average quoted size, ratio of average quoted size to average
traded size, daily traded volume, or stock price.''); Optiver
Letter; T. Rowe Price Letter; and Citadel Letter I.
\269\ See, e.g., T. Rowe Price Letter at 4 (``Other factors that
could be considered include queue length and quoted size at the top
of the order book, turnover, and whether the stock is quoted on
multiple exchanges.'') and Citadel Letter I.
\270\ See, e.g., T. Rowe Price Letter at 4.
\271\ See, e.g., BlackRock Letter at 5 and Optiver Letter
(``[w]e recommend that the Commission undertake further analysis of
the optimal level of tick granularity, leveraging price and volume
to define appropriate tick sizes.'').
\272\ See ICI Letter I at 11. See also Cambridge Letter at
(stating that minimum pricing increments should be reduced for those
stocks that have a TWAQS of $0.011 or less and ``are reasonably
liquid.'') and Citigroup Letter at 4.
---------------------------------------------------------------------------
One commenter suggested that in addition to the TWAQS, ``quote
stability'' should be measured.\273\ According to the commenter, quote
stability would be measured by looking at a change in a stock's quote
after execution; if the quote widens after an execution, ``the quoted
liquidity may not be sufficient for the liquidity demanded, suggesting
that the quote increment is not actually constraining quoting
activity.'' \274\ Another commenter suggested that the Commission
consider using ``spread leeway,'' which the commenter defined as equal
to the average quoted spread divided by the minimum tick size.\275\ The
commenter stated that spread leeway could ``effectively quantify the
extent to which bid-ask spreads are constrained by the minimum tick
size.'' \276\ Another commenter suggested that in addition to a TWAQS
of $0.011, there should be ``balance or near equilibrium of multiple
bids and offers at the top of the central order book'' as this would
``imply that market forces of supply and demand would naturally force
the bid/ask spread tighter through market competition.'' \277\ One
commenter that recommended a multi-factor approach to identify NMS
stocks suggested that in addition to average quoted spread, a high
quoted size to traded size ratio and a high average daily notional
turnover should be examined when identifying NMS stocks that are tick-
constrained.\278\ According to the commenter, a high quoted size to
traded size ratio is ``an objective signal that shows even though there
is an abundance of liquidity, the current $0.01 tick constraint
disincentivizes investors to cross the spread due to high costs,
resulting in a lack of trade executions.'' \279\ Further, the commenter
stated that a high average daily notional turnover is ``an objective
signal because it focuses the tick reduction effort on high turnover
securities that would benefit from the ability to be traded in finer
increments.'' \280\ Other commenters supported this approach.\281\
---------------------------------------------------------------------------
\273\ See, e.g., NYSE Letter I at 3. See also B. Riley Letter.
\274\ Id.
\275\ See MMI Letter (stating that spread leeway. . .
``quantifies the extent to which bid-ask spreads are constrained by
the minimum tick size . . . is equal to the average quoted spread
divided by the minimum tick size. Prior studies have suggested a
spread leeway of 3-9 as optimal for tick sizes to be neither too
small, nor too large.'').
\276\ Id. at 5.
\277\ See Invesco Letter at 3.
\278\ See Cboe Letter I and Cboe Letter II at 3. See also Cboe,
State Street, et al. Letter; State Street Letter at 3.
\279\ See Cboe Letter I at 2.
\280\ Id.
\281\ See, e.g., SIFMA Letter II at 40; Tastytrade Letter at 18;
and Themis Letter at 4.
---------------------------------------------------------------------------
After analyzing data to determine whether the suggested additional
factors would be helpful in eliminating NMS stocks that could be harmed
by a smaller minimum pricing increment,\282\ the Commission has
concluded that TWAQS is the appropriate measure to determine whether an
NMS stock should be eligible for a smaller minimum pricing increment.
Specifically, TWAQS provides a transparent and objective basis to
determine whether the $0.01 minimum pricing increment results in a
quoted spread that is too wide for a particular NMS stock. Other
possible factors, such as average quoted size, ratio of average quoted
size to average trade size, the average daily traded volume, queue
length, quotes on multiple exchanges or stock price, would add
unwarranted and additional complexity that would be difficult and
costly for market participants to monitor because some of these
measures require the purchase of proprietary data. Supplementing TWAQS
with quoted size, turnover calculations, quote stability, and the other
recommend criteria would similarly add additional complexity and
responsibilities to the primary listing exchanges assigned to
calculating TWAQS.\283\ The additional criteria suggested by commenters
are unnecessary because TWAQS is a sufficient, comprehensive and
objective way to determine whether NMS stocks are experiencing issues
of constraint related to the $0.01 minimum pricing increment for quotes
and orders.\284\ Specifically, the Commission concluded that harm is
unlikely to result if the other data factors suggested by commenters
(e.g., price, volume, or depth-based criteria) are not included.\285\
Accordingly, the Commission is not adopting factors other than the
TWAQS to measure which NMS stocks would be assigned a minimum pricing
increment of $0.005.
---------------------------------------------------------------------------
\282\ See infra section VII.D.1.b.iii. for further discussions
of alternative criteria.
\283\ See Proposing Release, supra note 11, at 80274.
\284\ See infra section VII.D.1.b.iii.
\285\ See infra section VII.D.1.b.iii.
---------------------------------------------------------------------------
One commenter suggested that issuers should be able to select the
minimum pricing increment for the quotes and orders of their
stock.\286\ The Commission disagrees. Rule 612 is an important rule
under Regulation NMS and serves to link the markets within the national
market system by establishing uniform minimum pricing increments for
all NMS stocks. This important linkage function would be undermined by
allowing increments to be individually assigned to each NMS stock in a
non-uniform manner. Rule 612, as originally adopted and as amended,
standardizes minimum pricing increments based on transparent and
objective criteria in order to ensure that minimum pricing increments
are applied uniformly. Introducing issuer choice would eliminate such
standardization and enable individual issuers to choose different
minimum pricing increments based on their specific, unique individual
preferences which would likely result in random and inconsistent
application of increments across NMS stocks that otherwise share
several relevant trading characteristics. Minimum pricing increment for
quotes and orders of NMS stocks priced greater than, or equal to, $1.00
per share based on unpredictable, opaque, non-standard criteria of
individual issuers would result in unnecessary complication, such as
varied minimum quoting increments
[[Page 81640]]
and investor confusion, to the national market system. Further, market
participants would likely incur additional costs related to, for
example, the monitoring and tracking of the minimum pricing increments
for issuers.
---------------------------------------------------------------------------
\286\ See Angel Letter at 5. But see Harris Letter at 8
(opposing suggestions that issuers should choose ticks).
---------------------------------------------------------------------------
7. Rule 612(a)--Definitions
As adopted, amended Rule 612(a) contains two definitions for
purposes of the rule--``Evaluation Period'' and ``Time Weighted Average
Quoted Spread.'' The primary listing exchanges will use these
definitions in identifying the required minimum pricing increments for
NMS stocks.
a. Evaluation Period
The Commission proposed to define ``Evaluation Period'' as the last
month of a calendar quarter (March in the first quarter, June in the
second quarter, September in the third quarter and December in the
fourth quarter) of a calendar year during which the primary listing
exchange shall measure the TWAQS of an NMS stock that is priced equal
to, or greater than, $1.00 to determine the minimum pricing increment
to be in effect for the next calendar quarter, as set forth by proposed
paragraph (c). In other words, the minimum pricing increment for quotes
and orders would have been evaluated every quarter based on one month's
worth of data and could have potentially changed once every quarter.
After considering the comments, the Commission is adopting a
revised definition of Evaluation Period. Rule 612(a)(1) defines
Evaluation Period as (i) the three months from January through March of
a calendar year and (ii) the three months from July through September
of a calendar year during which the TWAQS of an NMS stock shall be
measured by the primary listing exchange to determine the minimum
pricing increment for each NMS stock.
The Commission received comments on the proposed definition of the
Evaluation Period.\287\ Two commenters generally supported the
definition as proposed.\288\
---------------------------------------------------------------------------
\287\ See, e.g., IEX Letter I; Optiver Letter; NYSE Letter I;
Pragma Letter; MMI Letter; FIA PTG Letter II; BlackRock Letter;
SIFMA Letter II; T. Rowe Price Letter; Cboe Letter I and Cboe Letter
II; UBS Letter; and JPMorgan Letter at 4.
\288\ See, e.g., IEX Letter I and NYSE Letter I (stating that
the quarterly updates based on the last month of a quarter's data
would ``ensure that the next quarter's universe of tick-constrained
names is selected using the most recent and relevant basis and
allows for monitoring of other securities that are not yet tick-
constrained but maybe starting to exhibit tick-constrained
behavior.''). IEX, however, recommended that the second month of a
quarter be used for calculating the TWAQS. See also infra note 305
and accompanying text. See also Harris Letter at 7.
---------------------------------------------------------------------------
Several commenters stated that one month was too short a period for
measuring and calculating TWAQS.\289\ One commenter stated that an
analysis of one month's data ``has the potential to disproportionally
weigh systemic and idiosyncratic events (including corporate actions)
resulting in unrepresentative tick sizes.'' \290\ Another commenter
stated that ``longer evaluation periods will reduce the risk that
short-term aberrations will have an outsized impact on market
structure. Using too short of an evaluation period, especially during
periods of heightened volatility, could lead to unrepresentative price
variations that ultimately result in illogical minimum price
increments.'' \291\ A few commenters recommended providing a longer
period for conducting data analysis. Specifically, one commenter
suggested that the time frame be ``coterminous with the time between
tick size changes. In other words, if tick sizes are adjusted every
quarter, then the evaluation period should be every quarter . . .''
\292\ Another commenter suggested that the Evaluation Period should be
at least one quarter.\293\ Another commenter recommended that the
Evaluation Period be performed on an annual basis so as to reduce costs
and operational risks that may be created by needing to update relevant
systems.\294\ Finally, one commenter suggested that the evaluation of
NMS stocks be conducted on a semi-annual basis, based on six-months of
data, to reduce variability and complexity.\295\
---------------------------------------------------------------------------
\289\ See, e.g., Optiver Letter at 2; MMI Letter at 6; FIA PTG
Letter II at 2; and UBS Letter at 13. See also Cboe Letter I at 5
(stating that the framework for reevaluating the parameters for
revising tick changes according to their proposed methodology should
be quarterly or bi-annually so that the parameters ``remain nimble
to changing market conditions.'').
\290\ See Optiver Letter at 2.
\291\ See FIA PTG Letter II at 2-3. See also MMI Letter at 6
(``The evaluation period preceding the change should be at least one
quarter to avoid capturing instances of market volatility, or events
such as stock splits that may indirectly drive trading interest that
cause the behavior and characteristics of a stock to depart
dramatically from its history.'').
\292\ See FIA PTG Letter II at 2-3.
\293\ See MMI Letter at 6.
\294\ See UBS Letter at 13.
\295\ See JPMorgan Letter at 4.
---------------------------------------------------------------------------
A few commenters provided suggestions as to the length of time
between tick adjustments.\296\ One commenter stated that ticks should
be adjusted on a monthly basis rather than a quarterly basis.\297\ The
commenter stated ``[w]e would expect more frequent smaller updates to
reduce how often and how long a stock's tick stays outside the optimal
range.'' \298\ Another commenter, however, suggested that the
Commission align tick adjustments with other elements in the proposal.
Specifically, this commenter recommended that ``the Commission reduce
the frequency of changes and synchronize the intervals for revising
market structure parameters by updating both round lots and tick sizes
on a quarterly or semi-annual basis.'' \299\ Another commenter
suggested an annual consideration as a means to reduce burdens on
market participants and reduce operational risks.\300\
---------------------------------------------------------------------------
\296\ See Pragma Letter; UBS Letter; and BlackRock Letter. See
also SIFMA Letter II.
\297\ See Pragma Letter at 1.
\298\ See id.
\299\ See BlackRock Letter at 10. See also SIFMA Letter II
(commenting on three different elements--ticks, access fee caps and
round lots--that would have to be updated and stating ``[b]roker-
dealers will be required under the Tick Size Proposal to update
their systems to appropriately account for all three of these
variable changes, which carries inherent risks (and costs) of
inadvertent errors relative to today's environment where each of
these variables are static.''). See also section V.B.3.b.iii for a
discussion of the modifications to the round lot definition.
\300\ See UBS Letter at 12.
---------------------------------------------------------------------------
After considering the comments on the length of the Evaluation
Period, the amended rule will require that the TWAQS be measured over a
longer period of time than proposed, i.e., using three months' worth of
trading data instead of one month, and minimum pricing increments will
be assigned on a less frequent basis, i.e., every six months instead of
every three months. The Commission conducted analysis to evaluate the
length of the data analysis for the TWAQS and the length of time
between minimum pricing increment assignments.\301\ This revised
definition balances the concerns raised by commenters that a TWAQS
measured over too short a time frame could potentially be skewed by
high volatility or unique events, such as corporate actions,\302\ but
that a TWAQs measured over too long of a time period would increase the
probability of assigning a stale minimum pricing increment for quotes
and orders that does not reflect the prevailing trading
characteristics. Further, an annual evaluation would potentially cause
some NMS stocks to remain in a sub-optimal minimum pricing increment
for too long, while a monthly evaluation of NMS stocks would raise
concerns about investor confusion with frequent re-assignments
[[Page 81641]]
and increase operational risks due to the need for frequent systems
updates. The adopted semiannual evaluation addresses the potential
burdens and concerns of an Evaluation Period that is either too long or
too short.
---------------------------------------------------------------------------
\301\ See infra section VII.D.1.d for additional discussion on
the three-month period.
\302\ See supra note 290 and accompanying text. The suggestion
that minimum pricing increments be updated on a monthly basis would
raise these concerns.
---------------------------------------------------------------------------
Finally, two commenters recommended adding an implementation time
period between the calculation of the TWAQS and the potential change to
an NMS stock's minimum pricing increment.\303\ One commenter stated
that the proposed rule ``leaves little to no time for the industry to
communicate the change and update systems to reflect the new tick
sizes.'' \304\ Both commenters suggested that the rule should provide
one month between the end of the data collection and the effectiveness
of any new minimum pricing increments.\305\ One commenter stated that
one month between the calculation of the TWAQS and the implementation
of new tick sizes would ``give the industry adequate time to process
changes and minimize errors.'' \306\ Another commenter stated that the
quarterly changes with short transition time would raise operational
risk in the market and at individual firms.\307\ This commenter also
stated that ``[f]requent changes to tick sizes will require
considerable investor education.'' \308\ The Commission agrees with
commenters' suggestions and is also adopting an implementation period
for introducing new minimum pricing increments after an Evaluation
Period. As adopted, market participants will have one month to
implement any new minimum pricing increments. This will reduce concerns
about operational risk and will provide market participants with time
to inform investors of any changes in placing orders. One month is an
adequate time period for market participants to adapt and make any
required systems change to reflect the change, if any, in minimum
pricing increment of the quotes and orders of an NMS stock that is
priced greater than, or equal to, $1.00 per share. A longer period
could partially nullify the objectives of the adopted rule to
ameliorate issues related to the constraint of stocks that are quoting
at the $0.01 minimum pricing increment.
---------------------------------------------------------------------------
\303\ See T. Rowe Price Letter and IEX Letter.
\304\ See T. Rowe Price Letter at 4.
\305\ See IEX Letter I at 7 (suggesting that the Commission
provide ``one month between the end of data collection and the
beginning of trading with the reallocated tick sizes, in order to
avoid any unanticipated disruptions.'') and T. Rowe Price Letter.
\306\ See T. Rowe Price Letter at 4.
\307\ See Fidelity Letter at 13.
\308\ See Fidelity Letter at 13. See also Harris Letter at 7
(recommending mechanisms to ensure traders can determine relevant
ticks).
---------------------------------------------------------------------------
As discussed below, the Commission has aligned the semiannual
evaluation and implementation of minimum pricing increments and the
dates for implementing any minimum pricing increments with the timing
for changes to NMS stocks round lot assignment.\309\ The Commission
agrees with commenters who recommended that these two evaluations and
updates be conducted at the same time. This will lessen the burdens on
the primary listing exchanges and market participants of implementing
new minimum pricing increments and round lots. Further, it will lessen
operational risks associated with frequent system updates.
---------------------------------------------------------------------------
\309\ See also infra section V.B.3.b.iv.
---------------------------------------------------------------------------
Rule 612(a)(1) also requires that the TWAQS be measured for all NMS
stocks by the primary listing exchange. One commenter requested
clarification as to what would occur in volatile situations ``where a
low-priced stock (e.g., sub $1.00) suddenly jumps to a higher price
(e.g., $8.00).'' \310\ This situation could occur under preexisting
Rule 612 as the relevant minimum pricing increment is based on the
price of the order or quote. However, because orders in an NMS stock
can be submitted with prices at or above $1.00 and below $1.00
depending on its current market price, under the amended rule, each NMS
stock must have its TWAQS measured so that a minimum pricing increment
will be assigned for those orders that are priced at or above $1.00.
Therefore, as amended, all NMS stocks will be assigned a minimum
pricing increment based on its TWAQS and investors will be able to
understand the relevant minimum pricing increment for their orders when
priced at or over $1.00 and when priced under $1.00. Under the rule, as
adopted, quotes and orders in NMS stocks that are priced less than
$1.00 will continue to have a minimum pricing increment of
$0.0001.\311\ The operation of the amended rule is consistent with how
the preexisting rule operates in that quotes and orders for a
particular NMS stock may be required to be priced in a $0.01 or $0.005
increment when the price of an order is equal to or greater than $1.00
and may also be priced in a $0.0001 increment when the price of an
order is less than $1.00.
---------------------------------------------------------------------------
\310\ See SIFMA Letter I at 43.
\311\ See Rule 612(b)(3).
---------------------------------------------------------------------------
b. Time Weighted Average Quoted Spread
The Commission proposed to define TWAQS as the average dollar value
difference between the NBB and NBO during regular trading hours where
each instance of a unique NBB and a unique NBO is weighted by the
length of time that the quote prevailed as the NBB or NBO. The
Commission did not receive any comments on the definition of
TWAQS.\312\ The definition in Rule 612(a)(2) is adopted as proposed.
---------------------------------------------------------------------------
\312\ As discussed above, the Commission received comment on
whether there should be factors in addition to the TWAQS for
determining whether an NMS stock is tick-constrained. See supra
section III.C.6.
---------------------------------------------------------------------------
8. Rule 612(b)(1)--Semiannual Operative Dates
Rule 612, as adopted, contains amended paragraph (b)(1), which
defines the operative dates for the minimum pricing increments assigned
to each NMS stock and provides a month-long time period to implement
potentially new minimum pricing increments at the end of each
Evaluation Period. Specifically, minimum pricing increments for quotes
and orders will be operative on the first business day of May following
the Evaluation Period from January through March and the first business
day of November following the Evaluation Period from July through
September.\313\ In adopting these operative dates, the Commission seeks
to reduce the risk that market participants may not be fully staffed
during the time that technology changes are necessary to implement new
minimum pricing increments. Further, in addition to providing market
participants with adequate time to make necessary systems changes, the
implementation period will also provide adequate time for investors to
be notified about the minimum pricing increment for the quotes and
orders of NMS stocks that are priced equal to or greater than
$1.00.\314\
---------------------------------------------------------------------------
\313\ See supra section III.C.7.a, for a discussion of the
adopted Evaluation Period.
\314\ See supra note 303 and accompanying text.
---------------------------------------------------------------------------
Two commenters requested clarification as to how stock splits would
be handled.\315\ Once assigned under Rule 612(b)(1), minimum pricing
increments will remain operative until the next operative date (i.e.,
May or November). Therefore, a stock split will not impact an NMS
stock's minimum pricing increment until the next cycle. In order to
avoid the complexity and confusion that could occur if the minimum
pricing increment of NMS stocks were reassigned at unpredictable times,
minimum pricing increments will
[[Page 81642]]
not be changed during the time between operative dates.
---------------------------------------------------------------------------
\315\ See SIFMA Letter II at 42 and Virtu Letter II at 20.
---------------------------------------------------------------------------
9. Rule 612(c)--New NMS Stocks
Commenters asked for clarification on how new NMS stocks would be
handled under Rule 612.\316\ One commenter questioned how new NMS
stocks and IPOs would be assigned minimum pricing increments and stated
that allowing exchanges to assign different initial tick sizes could
lead to ``arbitrage of issuers choosing the exchange that offers the
most favorable initial tick size.'' \317\ Another commenter stated that
``the simplest and most intuitive alternative would be to use
specifications which were previously considered to be the standard
unit, such as a $0.01 tick size.'' \318\ The Commission agrees. New NMS
stocks will be assigned the same initial minimum pricing increment
under Rule 612(c), which requires all securities that become an NMS
stock to be assigned to the minimum pricing increment of $0.01.\319\
Thereafter, the TWAQS of the NMS stock will be calculated during the
next Evaluation Period to determine which minimum pricing increment
will be required under Rule 612(b)(2). Sub-penny increments are limited
to quotes and orders priced $1.00 or more for those NMS stocks that
have a demonstrated narrow TWAQS during the defined Evaluation Period;
new NMS stocks that become eligible for trading during an operative
period will not satisfy this requirement. New NMS stocks will have
their TWAQS calculated during the next Evaluation Period after they
start trading.
---------------------------------------------------------------------------
\316\ See SIFMA Letter II at 42; BlackRock Letter at 10; Virtu
Letter II at 18.
\317\ See SIFMA Letter II at 42; Virtu Letter II at 18.
\318\ See BlackRock Letter at 10.
\319\ See also section V.B.3.b.iii.
---------------------------------------------------------------------------
10. Rule 600(b)(89)--Regulatory Data
The Commission proposed to amend the definition of regulatory data
in Rule 600(b)(78) \320\ to require the primary listing exchange for
each NMS stock to calculate and provide to competing consolidators,
self-aggregators, and the exclusive SIPs an indicator of the applicable
minimum pricing increment required under Rule 612. The Commission is
adopting the minimum pricing increment indicator, as proposed, under
the definition of regulatory data in Rule 600(b)(89)(i)(F). A minimum
pricing increment indicator will be useful to market participants,
including investors, by providing important information about the
relevant minimum pricing increment for each NMS stock. This indicator
will help market participants, including investors, with submitting
orders in the relevant increment. Because the minimum pricing increment
can change on a semiannual basis depending on the TWAQS on an NMS
stock, this indicator will enable market participants to trade in a
more informed manner. The indicator will be included in SIP data that
is disseminated by the exclusive SIPs and consolidated market data
\321\ disseminated by competing consolidators, which will help to
ensure the wide availability of information about the applicable
minimum pricing increment for each NMS stock.
---------------------------------------------------------------------------
\320\ Preexisting Rule 600(b)(78) was subsequently renumbered to
Rule 600(b)(89) by the Rule 605 Amendments. See Rule 605 Amendments,
supra note 10.
\321\ See rule 600(b)(24).
---------------------------------------------------------------------------
One commenter supported the minimum pricing increment indicator
stating that it would make the new increments easier to implement.\322\
This commenter suggested that the exclusive SIPs publish the indicator
every morning, in a machine-readable format, and free of charge.\323\
---------------------------------------------------------------------------
\322\ See MMI Letter at 6.
\323\ See MMI Letter at 6.
---------------------------------------------------------------------------
The exclusive SIPs currently provide certain information that
comprises regulatory data as part of SIP data.\324\ Under the rule, the
exclusive SIPs will be required to collect and disseminate a new
regulatory data element, the minimum pricing increment indicator, and
collect and disseminate this regulatory data element as part of SIP
data.\325\ To the extent that the exclusive SIPs charge fees for this
new regulatory data element, such fees will be required to be filed
under rule 608 of Regulation NMS and must be fair and reasonable and
not unreasonably discriminatory.\326\ The Commission has not required a
specific format for SIP data, including regulatory data; such format
will be developed by the Operating Committees' for the Equity Data
Plans consistent with regulatory requirements.\327\
---------------------------------------------------------------------------
\324\ See MDI Adopting Release, supra note 10, at 18729 for a
description of the regulatory messages that are disseminated by the
exclusive SIPs.
\325\ See Rule 600(b)(89)(iv).
\326\ Sections 11A(c)(1)(C) and 11A(c)(1)(D) and Rule 603(a).
See MDI Adopting Release, supra note 10, at 18684.
\327\ The SIP data format will be available at the SIP's
website, https://www.ctaplan.com/index.
---------------------------------------------------------------------------
D. Minimum Pricing Increment for Trades
The Commission proposed to amend Rule 612 to introduce minimum
pricing increments for trades of NMS stocks where the minimum pricing
increment for trading NMS stocks priced at or above $1.00 would vary
and correlate to one of the four proposed minimum pricing increments
for quoting (i.e., $0.001, $0.002, $0.005 and $0.01), subject to
proposed exceptions for midpoint trades and benchmark trades.\328\ The
proposed minimum pricing increments for trades would have harmonized
trading increments (1) with the proposed variable minimum pricing
increments for quotes and orders (in this release, ``Quote and Trade
Harmonization''), and (2) across all trading venues (in this release,
``Venue Harmonization''). Specifically, Quote and Trade Harmonization
would have required all trading to occur in the same increments as
those required of quotes and orders subject to certain exceptions for
midpoint and benchmark trades. Venue Harmonization would have required
all trading on exchanges, ATSs and OTC to occur in the same pricing
increments.
---------------------------------------------------------------------------
\328\ The Commission also proposed to impose a minimum pricing
increment for trades for quotes and orders priced less than $1.00
that would have been the same as the minimum pricing increment for
quotes, i.e., $0.0001. See Proposing Release, supra note 11, at
80283.
---------------------------------------------------------------------------
In the Proposing Release,\329\ the Commission stated that it was
concerned about the competitive dynamic between exchanges, ATSs, and
OTC markets and that a potential contributing factor was the ability of
OTC market makers to execute orders in price increments that are
smaller than the price increments that exchanges and ATSs can
practically provide.\330\ The Commission stated that applying the
minimum pricing increment to trades across all venues would promote
equal regulation and fair competition among market participants such as
exchanges, OTC market makers and ATSs, particularly as it relates to
retail order flow; \331\ and that it was ``reasonable to assume that .
. . [applying] a minimum pricing increment to trades . . ., could
result in greater competition between exchanges and ATSs with other OTC
market makers, including wholesalers . . .'' \332\ However, the
Commission also stated that it could not anticipate how OTC market
makers would adjust to increased competitive pressure and whether a
market-wide trading increment would yield a
[[Page 81643]]
``positive, negative or neutral'' net effect on retail price
improvement.\333\
---------------------------------------------------------------------------
\329\ See Proposing Release, supra note 11, at 80268-69.
\330\ See Proposing Release, supra note 11, at 80283.
\331\ See Proposing Release, supra note 11, at 80283.
\332\ See Proposing Release, supra note 11, at 80303.
\333\ See Proposing Release, supra note 11, at 80326.
---------------------------------------------------------------------------
The Commission received several comments on the proposal to
establish minimum pricing increments for trades, including comments
related to Quote and Trade Harmonization,\334\ Venue
Harmonization,\335\ statutory authority,\336\ rationale,\337\ impact on
price improvement,\338\ exceptions to minimum pricing increment for
trades,\339\ and exchange RLP programs.\340\ After considering comments
and in light of changes from the proposal that the Commission is making
to amended Rule 612, the Commission has decided, consistent with one of
the Reasonable Alternatives set forth in the Proposing Release,\341\
not to adopt a minimum pricing increment for trades. As described
above, the Commission is amending Rule 612 to adopt one smaller minimum
pricing increment for quotes and orders that primarily focuses on those
NMS stocks that are experiencing constraint with the $0.01 minimum
pricing increment. A secondary impact of the changes to the minimum
quoting increment should be that it helps to partially address the
concerns related to fair competition between exchanges, ATSs, and OTC
markets that proposing a market-wide trading increment was designed to
address.
---------------------------------------------------------------------------
\334\ See, e.g., Nasdaq Letter I at 17; Citadel Letter I at 31;
State Street Letter at 4; Citigroup Letter at 5; MFA Letter at 7,
Letter from Nandini Sukumar, Chief Executive Officer, World
Federation of Exchanges, dated Mar. 30, 2023 (``World Federation of
Exchanges Letter'') at 5, STA Letter at 7, NYSE Letter I at 6;
Nasdaq Letter I at 17, NYSE, Schwab, and Citadel Letter at 2,
Brandes Letter at 2; Schwab Letter II at 6, Cambridge Letter at 6;
B. Riley Letter at 1, Vanguard Letter at 5, Robinhood Letter at 40,
55; JPMorgan Letter at 6.
\335\ See, e.g., Cboe Letter II at 9, IEX Letter I at 2, Nasdaq
Letter I at 2, Citigroup Letter at 5, Pragma Letter at 1, Luke
Peterson Letter; Chris Miller Letter, Amanda Kappes Letter.
\336\ See Citadel Letter I at 4 and Robinhood Letter at 28.
\337\ See Citigroup Letter at 5, Nasdaq Letter I at 16, World
Federation of Exchanges Letter at 4, Better Markets Letter I at 13,
Vanguard Letter at 4, IEX Letter III at 5, Drew Ferguson Letter at
1, Max Garrison Letter at 1, Lukas Boller Letter at 1, Trent Miller
Letter at 1, Phillip Worts Letter at 1, James Letter at 1, Andrew
Garley Letter at 1, Larry Douglas Letter at 1, Steve Sullivan Letter
at 1, Luke Czarnota Letter at 1, Charles S Letter at 1, Melisa
Virginillo Letter at 1, Melissa Hyer Letter at 1, Keagan Wethington
Letter at 1, DH Letter at 1, Alex Riley Letter at 1, Trevor
Capestany at 1, Marco Daeblitz at 1, Steven Sullivan Letter at 1.
See also Patrick Sexton, EVP, General Counsel & Corporate Secretary,
Cboe Global Markets, Inc., dated Aug. 23, 2023 (``Cboe Letter III'')
at 9.
\338\ See, e.g., SIFMA Letter II at 4, Citigroup Letter at 5,
TradeStation Letter at 6, CCMR Letter at 27, Virtu Letter II at 7,
Fidelity Letter at 13, BlackRock Letter at 9, Robinhood Letter at
20, JPMorgan Letter at 6, Morgan Stanley Letter at 4, TastyTrade
Letter at 20 and Nasdaq Letter I at 18.
\339\ See, e.g., JPMorgan Letter at 5; ICI Letter I at 17-18,
Vanguard Letter at 5; IEX Letter I at 17 and BlackRock Letter at 9.
\340\ See, e.g., Cboe Letter III at 10-11, IEX Letter I at 17-
18, JPMorgan Letter at 6, Ontario Teachers et al. Letter at 2, NYSE
Letter I at 6.
\341\ See Proposing Release, supra note 11, at 80339.
---------------------------------------------------------------------------
Amended Rule 612 will reduce the minimum pricing increment for
quotes and orders to $0.005 for certain NMS stocks that are priced
equal to, or greater than, $1.00 per share which in turn also
effectively reduces the increment that such stocks are able to trade
in. Under amended Rule 612, OTC markets will continue to be able to
trade more readily in comparatively smaller increments (e.g., $0.001 or
$0.0001) than exchanges and ATSs, however, exchanges and ATSs will now
be able to trade more regularly at smaller increments (i.e., $0.005 or
$0.0025), compared to preexisting Rule 612, for those NMS stocks that
are assigned the $0.005 minimum pricing increment. By effectively
reducing the trading increment for such NMS stocks, which represent a
significant amount of the daily trading volume (approximately 58%) and
dollar volume (approximately 43%),\342\ the potential trade pricing
discrepancy between exchanges, ATSs and OTC markets, while not
harmonized, will be reduced. Market participants will be able to better
compete based on the pricing of quotes and orders. Thus, because
reducing the minimum quoting increment for a significant amount of
volume of NMS stocks, whether measured by trading or dollars, also
effectively reduces the trading increments for those stocks, the
concerns related to the ability of OTC market makers to trade in
comparatively finer increments raised by the Commission in the
Proposing Release will be partially addressed so that the Commission
has determined not to adopt the minimum pricing increment for trading.
---------------------------------------------------------------------------
\342\ See infra section VII.D.1.a.
---------------------------------------------------------------------------
However, because the amendment to Rule 612 only partially addresses
the competitive dynamic between OTC market makers and exchanges and
ATSs described in the Proposing Release, and furthermore allows OTC
market makers to continue to execute trades in comparatively finer
increments, the Commission staff will continue to monitor sub-penny
trading to evaluate whether further action is appropriate for the
protection of investors and to assure ``fair competition among brokers
and dealers, among exchange markets, and between exchange markets and
markets other than exchange markets'' in the national market system.
The Commission's simplified, incremental approach to amending Rule
612 focuses on addressing issues that have developed regarding quoting
constraints for certain NMS stocks because of the $0.01 minimum pricing
increment. Further, the amendment to Rule 612 will facilitate the
transition of market participants and investors to the new tick size
regime and the wider use of sub-penny quoting. The amendment, as
adopted, also reduces the anticipated implementation costs compared to
the proposed amendments to Rule 612. Finally, under amended Rule 612:
(1) RLPs \343\ that operate pursuant to Commission exemptions that
either permit certain quoting and trading in increments of $0.001,\344\
or aggregate order flow at the midpoint,\345\ will be able to continue
to operate without interruption and without changes to exchange rules
or the grant of further exemptive relief by the Commission; (2) sub-
penny price improvement will continue to be permitted consistent with
the requirements of the rule; (3) and investors will continue to be
able to manage their order flow and implement trading strategies
through the use of midpoint orders and benchmark trades.
---------------------------------------------------------------------------
\343\ See supra note 146.
\344\ See, e.g., NYSE Rule 7.44 and BX Rule 4780.
\345\ See, e.g., NYSE Arca Rule 7.44-E and IEX Rule 11.232.
---------------------------------------------------------------------------
IV. Final Rule 610 of Regulation NMS--Fees for Access to Quotations
The Commission is adopting amendments to Rule 610(c)(1)(ii) as
proposed with technical modifications to remove the reference to the
minimum pricing increment. The Commission is not adopting proposed Rule
610(c)(1)(i) because it is unnecessary.\346\ Further, the Commission is
adopting amendments to Rule 610(c)(2) as proposed with modifications to
align the access fee caps for protected quotations in NMS stocks priced
below $1.00 and those priced $1.00 and above.\347\ Finally, the
Commission is removing outdated references to the ``The Nasdaq Stock
Market, Inc.'' in Rule 610(c), as proposed, because the Nasdaq Stock
Market is now a national
[[Page 81644]]
securities exchange and the language is redundant.\348\
---------------------------------------------------------------------------
\346\ The Commission is not adopting the proposed 5 mils access
fee cap because it is not adopting the $0.001 minimum pricing
increment. As proposed, all protected quotes in NMS stocks priced
$1.00 or more that would have been assigned a minimum pricing
increment other than $0.001 would have been subject to the proposed
10 mils access fee cap. The Commission is adopting this same model--
all protected quotes in NMS stocks priced $1.00 or more will be
subject to the 10 mils access fee cap.
\347\ See infra section VII.D.2.a.
\348\ See Proposing Release, supra note 11, at 80292.
---------------------------------------------------------------------------
Specifically, under Rule 610(c), as amended, a trading center \349\
will not be permitted to impose, or permit to be imposed, any fee or
fees for the execution of an order against a protected quotation of the
trading center or against any other quotation of the trading center
that is the best bid or best offer of a national securities exchange or
the best bid or best offer of a national securities association in an
NMS stock that exceed or accumulate to more than $0.001 per share if
the price of the protected quotation or other quotation is $1.00 or
more, and the fee or fees will not be permitted to exceed or accumulate
to more than 0.1% of the quotation price per share if the price of the
protected quotation or other quotation is less than $1.00.
---------------------------------------------------------------------------
\349\ 17 CFR 242.600(b)(106).
---------------------------------------------------------------------------
The Commission is also adopting Rule 610(d) as proposed to require
that all exchange fees and rebates be determinable at the time of
execution. For the reasons discussed below, these amendments to Rule
610 are appropriate for the modern national market system.
A. Background
Rule 610(c) was adopted in furtherance of the Congressional
directives in section 11A of the Exchange Act and was designed to
promote fair and non-discriminatory access to quotations displayed in
the national market system.\350\ Rule 610(c) seeks to ensure the
fairness and accuracy of displayed quotations by establishing an outer
limit on the cost of accessing such quotations \351\ and was designed
to help to ensure that orders placed in the national market system
reflect the best prices available. The access fee caps are necessary to
achieve the purposes of the Exchange Act, including section
11A(c)(1)(B) of the Exchange Act, which authorizes the Commission to
adopt rules assuring the fairness and usefulness of quotation
information. The Commission has stated that for quotations to be fair
and useful, ``there must be some limit on the extent to which the true
price for those who access quotations can vary from the displayed
price.'' \352\
---------------------------------------------------------------------------
\350\ The Commission also stated that by imposing a uniform fee
limitation of $0.003 per share, Rule 610(c) will promote equal
regulation of different types of trading centers. See Regulation NMS
Adopting Release, supra note 4, at 37595.
\351\ See Regulation NMS Adopting Release, supra note 4, at
37502.
\352\ See Regulation NMS Adopting Release, supra note 4, at
37545.
---------------------------------------------------------------------------
Rule 610 was adopted at the same time as Rule 611, the Order
Protection Rule, which established intermarket protection against
trade-throughs \353\ for all NMS stocks. Rule 610(c) was designed to
preclude trading centers that posted protected quotations from raising
their fees in an attempt to take improper advantage of the trade-
through protections adopted under Rule 611.\354\ The Commission
designed the access fee caps to preserve the benefits of the
strengthened price protection under Rule 611 and more efficient
linkages among trading centers that were developed under Regulation NMS
to access protected quotations that could be disrupted if substantial
fees were charged.\355\ At the time of adoption, the Commission
recognized the importance of protecting the best displayed and
accessible prices in promoting deep and stable markets that minimize
investor costs. In this regard, the Commission stated that Rule 611
would help to minimize investor transaction costs, which is ``the
hallmark of efficient markets'' and a ``primary objective of the
[national market system].'' \356\ Rule 610 is an important component in
supporting these goals.
---------------------------------------------------------------------------
\353\ A trade-through occurs when a trading center executes an
order at a price that is inferior to the price of a protected
quotation that is displayed by another trading center. See 17 CFR
242.600(b)(105) for the definition of trade-through under Regulation
NMS.
\354\ See Regulation NMS Adopting Release, supra note 4, at
37544 and 37595.
\355\ See Regulation NMS Adopting Release, supra note 4, at
37545.
\356\ See Regulation NMS Adopting Release, supra note 4, at
37498.
---------------------------------------------------------------------------
Market participants, including investors, need fair and efficient
access to the best priced quotations in the national market system.
Therefore, Rule 610(c) remains an important part of the national market
system to achieve the purposes of the Exchange Act, preserve the
benefits of price protection, and help ensure that displayed quotations
reflect something close to actual costs incurred for the
transaction.\357\
---------------------------------------------------------------------------
\357\ The access fee caps were calculated based upon the then
current fees that were charged by certain trading venues. See
Regulation NMS Adopting Release, supra note 4, at 37545 (stating
``the $0.003 fee limitation is consistent with current business
practices, as very few trading centers currently charge fees that
exceed this amount . . . [and those that do] do not account for a
large percentage of the trading volume.''). At the time the access
fee caps were adopted, the minimum pricing increment for quotes and
orders priced $1.00 or greater was $0.01 as Rule 612 was adopted at
the same time as Rule 610(c).
---------------------------------------------------------------------------
B. Issues Raised in the Existing Market Structure and the Need for the
Amendments
The national market system of 2024 is significantly different than
the national market system that existed when Regulation NMS was adopted
in 2005.\358\ Since Regulation NMS was adopted, new trading practices,
order types, and routing strategies have developed that did not exist
when Rule 610 was adopted.\359\ In addition, exchanges have developed
complex fee structures that charge the outer limits permitted for
accessing protected quotations and use those fees to fund rebates,
which have the effect of creating a discrepancy between displayed
prices and net prices.\360\ Finally, the national market system has
seen a proliferation of new exchanges, often within the same exchange
group, that implement varied pricing models to attract specific market
participants to their markets.\361\ The Commission has monitored these
developments and engaged extensively with market participants about the
impact of the modern fee structures on fair and efficient access to
protected quotations as well as the usefulness and accuracy of such
quotations.\362\
---------------------------------------------------------------------------
\358\ See Proposing Release, supra note 11, at 80289. See also
Letter from John Ramsay, Chief Market Policy Officer, Investors
Exchange LLC, dated Oct. 19, 2023 (``IEX Letter IV'') at 5-10;
Nasdaq Letter I at 4.
\359\ See also Letter from Theodore R. Lazo, Managing Director &
Associate General Counsel, SIFMA, to Brent J. Fields, Secretary,
Commission, dated Mar. 29, 2017, at 8 (stating exchanges have
developed order types primarily designed to avoid paying high fees,
market participants implement complex routing strategies (consistent
with their best execution obligations) to avoid paying high exchange
access fees in favor of lower costs ATSs and the market place has
seen a high level of fragmentation ``driven by each exchange group's
desire to provide a variety of pricing models within the wide
pricing range between 0 and 30 mils.'') (``SIFMA 2017 Letter'').
\360\ See infra section VII.C.2, note 1107 and accompanying text
and note 1457 (showing that the primary reason that access fees
remain near 30 mils on most exchanges is to fund rebates) and Panel
A of table 4, infra section VII.C.2.c.
\361\ See infra tables 4 and 5 showing within the large exchange
groups multiple exchanges each with different fee and rebate models.
See also e.g., SIFMA 2017 Letter, supra note 359 at 8 (stating ``the
high level of fragmentation . . . is in part driven by each of the
exchange group's desire to provide a variety of pricing models
within the wide pricing range between 0 and 30 mils.'').
\362\ For example, the EMSAC considered, among other things,
whether the access fee cap should be modified. See EMSAC Archives,
supra note 4. See also Concept Release on Equity Market Structure,
supra note 59.
---------------------------------------------------------------------------
Market participants have considered and suggested reductions in the
access fee caps for many years to address market distortions (as
further discussed
[[Page 81645]]
below) \363\ and better reflect evolutions in the market since
Regulation NMS was adopted.\364\ One commenter stated that ``[d]igital
innovations and efficiencies since 2005 have undoubtedly reduced the
costs of collecting, storing, processing, and transmitting
information'' and that ``despite reduced costs, increased efficiency,
and all the new data and computing power available, the access fee cap
has remained fixed at an inflated level that reflects the technology
capabilities of 2005.'' \365\
---------------------------------------------------------------------------
\363\ See, e.g., infra notes 371-376 and accompanying text and
infra section IV.B.2 and IV.D.
\364\ See, e.g., Letter from Theodore R. Lazo, Managing Director
& Associate General Counsel, SIFMA, to Brent J. Fields, Secretary,
Commission, dated May 24, 2018, at 2 (commenting on File No. S7-05-
18 ``Transaction Fee Pilot for NMS Stocks'') (stating ``On several
occasions, SIFMA has recommended that the Commission reduce the
access fee cap to no more than five cents per 100 shares because the
cap has not been adjusted to reflect market developments since
Regulation NMS was adopted more than a decade ago.'') (``SIFMA 2018
Letter''); Goldman 2018 Letter at 1 (stating ``a reduction in the
Fee Cap from $0.0030 to $0.0010 per share could be supported today
[2018] and would be better calibrated with the present-day trading
and execution costs, which have decreased substantially since
2005''); SIFMA 2017 Letter, supra note 359, at 3 (stating the
Commission should consider ``reducing the access fee cap to no more
than $0.0005 for all securities'' because ``[s]ince Reg. NMS was
adopted, spreads have narrowed and commissions have decreased,
making the existing cap of access fees outsized relative to today's
market realities.''); Letter from Theodore R. Lazo, Managing
Director & Associate General Counsel, SIFMA, to Mary Jo White,
Chair, Commission, dated May 24, 2015, at 2-3 (supporting reduction
in access fee to ``no more than five cents per hundred shares''
because access fees are ``an outsized element of transaction costs
that in turn distorts price discovery and contributes to market
complexity, both on- and off-exchange'' and further stating ``market
participants regularly implement complex order routing strategies .
. . that divide, route and re-route orders and parts of orders, when
possible, to market centers that enable them to avoid paying
excessive access fees'' and also stating ``access fees have
increased complexity on exchanges . . . through the proliferation of
exchange order types designed to avoid access fees.''); Letter from
Daniel Keegan, Managing Director, Head of Americas Equities,
Citigroup Global Markets Inc. to Elizabeth Murphy, Secretary,
Commission, dated Aug. 7, 2014, at 5 (``Citigroup 2014 Letter'')
(commenting on Concept Release on Equity Market Structure and
stating ``[t]he SEC should explore the impacts on the overall
markets of complex market structure issues such as maker/taker
pricing and access fees'' and ``[a]t a minimum, the cap on access
fees should be reduced to below 10 mils''); Letter from Theodore R.
Lazo, Director & Associate General Counsel, SIFMA, to Mary Jo White,
Chair, Commission, dated Oct. 24, 2014, at 2 (providing
``Recommendations for Equity Market Structure Reforms'' including
``that exchange access fees be significantly reduced, to no more
than five cents per 100 shares''); NYT Dealbook OpEd: How to Improve
Market Structure, Curt Bradbury, Chief Operating Officer, Stephen's
Inc. and Kenneth E. Bentsen, Jr., President and CEO, SIFMA (SIFMA
Market Structure Task Force Recommendations), dated July 14, 2014
(stating ``Access fees charged by exchanges and other venues should
be dramatically reduced, if not eliminated. While brokers are
legally required to route their orders to the exchange that is
quoting the best price--so called `protected quotes'--the exchanges
are permitted to charge relatively high fees for accessing these
quotes: currently 30 cents for every 100 shares. These fees have
distorted market pricing as they are a significant percentage of
overall trading costs and are several times higher than the fees
charged by off-exchange venues. As a result, brokers often avoid
routing their orders to exchanges. Exchanges also rebate most of
their access fee revenue through price structures such as `maker/
taker.' These developments have led to a proliferation of order
types designed to avoid access fees and capture rebates, and that
proliferation, in turn, adds complexity to the system, requires
continuing technology changes and creates potential for market
instability. The Securities and Exchange Commission should reduce
the current cap on access fees to no more than 5 cents per 100
shares, and indeed should consider eliminating access fees
altogether.''); Bradley Hope & Scott Patterson, ``NYSE Plan Would
Revamp Trading,'' WALL ST. J. (Dec. 17, 2014), available at http://www.wsj.com/articles/intercontinental-exchangeproposing-major-stock-market-overhaul-1418844900 (stating that since 2005, when the fee
cap of $0.003 per share was chosen, competitive and technological
advancements have led to decreased costs (spreads and commissions),
and as a result, access fees have become a larger portion of overall
transaction costs); ICE's Six Recommendations for Reforming
Markets,'' WALL ST. J. (Dec. 18, 2014), available at http://blogs.wsj.com/moneybeat/2014/12/18/ices-six-recommendations-for-reformingmarkets/ (recommending reduction in the access fee cap to
$0.0005 in conjunction with adoption of a ``trade at'' rule,
``eliminating maker-taker pricing'' and stating ``[w]ith myriad
different make-take and take-make pricing models in existence today,
we believe the potential conflicts and complexity that ensue from
the maker-taker models outweigh any perceived benefits. We believe
there are better options available to incentivize market-makers to
maintain two-sided quotes and reduce intraday volatility'' including
incentive programs that would obligate market makers to provide
liquidity); Joe Ratterman, Chief Executive Officer, & Chris
Concannon, President, BATS, ``Open Letter to U.S. Securities
Industry Participants Re: Market Structure Reform Discussion,'' at 1
(Jan. 6, 2015), available at http://cdn.batstrading.com/resources/newsletters/OpenLetter010615.pdf (stating access fee cap ``requires
a substantial reduction and restructuring'' and further stating the
cap ``has remained unchanged for far too long and has never been
reevaluated for potential market distortions given the substantially
altered broker models and reductions in commissions since the
implementation of Regulation NMS.'').
\365\ See IEX Letter IV at 8. See also e.g., Letter from Paul M.
Russo, Managing Director, Goldman Sachs & Co. LLC, to Brent J.
Fields, Secretary, Commission, at 3 (May 24, 2018), available at
https://www.sec.gov/comments/s7-05-18/s70518-3711788-162473.pdf
(``Goldman 2018 Letter'') (commenting on File No. S7-05-18
``Transaction Fee Pilot for NMS Stocks'') at 2 (stating in 2018 ``In
the thirteen years since the Commission adopted the Fee Cap, spreads
have considerably narrowed and commission rates have contracted.
However, the Fee Cap has remained unadjusted. There is a well-
developed, general consensus among market participants that a [30
mil] per share Fee Cap is an outdated benchmark for execution costs
in today's trading environment. As a limit, it creates an upper-
range that is simply too high and far from representative of true
prices in the marketplace.'').
---------------------------------------------------------------------------
1. Amendments to Rule 612
As discussed above, the Commission is adopting a smaller minimum
pricing increment of $0.005 for certain NMS stocks. Because the
Commission is adopting the smaller $0.005 increment, it is also
amending the preexisting access fee caps in Rule 610(c) to prevent
introducing new pricing distortions in the market.\366\ For protected
quotations priced $1.00 or more, the Commission is adopting a 10 mil
access fee cap and has decided that this level is appropriate based on
several additional considerations, as discussed in the following
sections.
---------------------------------------------------------------------------
\366\ See infra section VII.D.2.a.
---------------------------------------------------------------------------
2. Exchange Fee Models
The exchange fee structures in the national market system that have
developed under Rule 610 are complex and consist of fees charged and
rebates paid to market participants. As stated above, exchanges using a
``maker-taker'' pricing model, pay a rebate to a ``maker'' or provider
of liquidity, which is funded by the fees charged to a ``taker'' of
liquidity.\367\ The exchange earns as revenue the difference between
the fee paid by the taker and the rebate paid to the provider or
maker.\368\ For maker-taker exchanges, the amount of the taker fee is
limited by the access fee caps imposed by Rule 610(c). The Rule 610(c)
access fee caps apply to the fees assessed on an incoming order that
executes against a resting protected quote, but do not apply to the
rebates. However, the Rule 610(c) access fee caps indirectly limit the
average amount of the rebates that an exchange offers to about $0.0030
per share in order to maintain net positive transaction revenues. Thus,
an exchange may charge higher access fees to fund higher
[[Page 81646]]
liquidity rebates.\369\ Some exchanges state that rebates are necessary
in order for them to attract trading volume.\370\
---------------------------------------------------------------------------
\367\ See SRO fee schedules, which are available on each SRO's
website. See also infra section VII.C.2.c, table 4. This discussion
focuses on exchange fees because, currently, exchanges are the only
trading centers that display protected quotations. If an ATS or OTC
market maker displayed a protected quotation, its fees would be
subject to the access fee caps under Rule 610(c).
\368\ A few exchanges have adopted a ``taker-maker'' pricing
model (also called an inverted model), in which they charge a fee to
the provider of liquidity and pay a rebate to the taker of
liquidity. See, e.g., Nasdaq BX fee schedule available at http://www.nasdaqtrader.com/trader.aspx?id=bx_pricing (as of Feb. 2024);
NYSE National fee schedule available at https://www.nyse.com/publicdocs/nyse/regulation/nyse/NYSE_National_Schedule_of_Fees.pdf
(as of Jan. 1, 2024); Cboe BYX fee schedule available at https://www.cboe.com/us/equities/membership/fee_schedule/byx/ (as of Feb.
2024); and Cboe EDGA fee schedule available at https://www.cboe.com/us/equities/membership/fee_schedule/edga/ (as of Feb., 2024). See
also infra section VII.C.2.c, table 4. For taker-maker exchanges,
the amount of the maker fee charged to the provider of liquidity is
not bounded by the Rule 610(c) access fee cap because such fee is
not a charge to access the market's best bid/offer for NMS stocks,
but such fees typically are no more than $0.0030.
\369\ This was one of the concerns the Commission identified
when it approved the access fee caps. See Regulation NMS Adopting
Release, supra note 4, at 37545. (``[T]he fee limitation is
necessary to achieve the purposes of the Exchange Act. Access fees
tend to be highest when markets use them to fund substantial rebates
to liquidity providers, rather than merely to compensate for agency
services.'').
\370\ See, e.g., Letter from Patrick Sexton, EVP, General
Counsel & Corporate Secretary, Cboe Global Markets, Inc., dated Apr.
5, 2024 (``Cboe Letter IV'') at 2-5; Letter from Brett Kitt, Vice
President, Deputy General Counsel, Nasdaq, Inc., dated Mar. 25, 2024
(``Nasdaq Letter IV'') at 3-5; Nasdaq Letter III at 2-5.
---------------------------------------------------------------------------
In recent years, a variety of concerns have been stated about the
prevailing maker-taker fee model, and particularly the rebates paid by
the exchanges. Those include that the fee/rebate models: (1) undermine
market transparency since displayed prices do not account for exchange
transaction fees or rebates and therefore do not reflect the net
economic costs of a trade; \371\ (2) serve as a way to effectively
quote in sub-penny increments on a net basis when the effect of a
maker-taker exchange's sub-penny rebate is taken into account even
though the minimum quoting increment is expressed in full pennies;
\372\ (3) introduce unnecessary market complexity through the
proliferation of new exchange order types (and new exchanges) designed
solely to take advantage of pricing models; \373\ (4) drive orders to
non-exchange trading centers that do not display quotes as market
participants seek to avoid the higher fees that exchanges charge to
subsidize the rebates they offer to attract liquidity; \374\ and (5)
benefit sophisticated market participants like market makers and
proprietary traders at the expense of other market participants.\375\
Further, the prevailing access fee structure creates potential
conflicts of interest for broker-dealers, who must provide the best
execution to their customers' orders while facing potentially
conflicting economic incentives to avoid fees or earn rebates from the
trading centers to which they direct those orders for execution.\376\
---------------------------------------------------------------------------
\371\ See, e.g., Letter from Richard Steiner, Global Equities
Liaison to Regulatory & Government Affairs, RBC Capital Markets, to
Elizabeth Murphy, Secretary, Commission, at 2-3 (Nov. 22, 2013),
available at https://www.sec.gov/comments/s7-02-10/s70210-411.pdf
(``RBC Capital Letter'') (commenting on potential equity market
structure initiatives).
\372\ See, e.g., Larry Harris, ``Maker-Taker Pricing Effects on
Market Quotations,'' at 24-25 (Nov. 14, 2013).
\373\ See, e.g., Curt Bradbury, Market Structure Task Force
Chair, Board of Directors, SIFMA, and Kenneth E. Bentsen Jr.,
President and Chief Executive Officer, SIFMA, Opinion, ``How to
Improve Market Structure,'' N.Y. Times (July 14, 2014), available at
http://dealbook.nytimes.com/2014/07/14/how-to-improve-market-structure/?_r=0 (stating that the ``proliferation of order types
designed to avoid access fees and capture rebates . . . adds
complexity to the system, requires continuing technology changes and
creates potential for market instability'' and recommending access
fees charged by exchanges be ``dramatically reduced, if not
eliminated''); RBC Capital Letter at 2; and Letter from Haim Bodek,
Managing Principal and Stanislav Dolgopolov, Regulatory Consultant,
Decimus Capital Markets, LLC, dated Apr. 25, 2016, at 3 and 11
(``Decimus 2016 Letter''); Vanguard Letter at 6.
\374\ See, e.g., Menkveld, Albert J., Bart Zhou Yueshen, and
Haoxiang Zhu, ``Shades of darkness: A pecking order of trading
venues.'' Journal of Financial Economics 124, no. 3 (2017) at 503-
534, available at https://www.mit.edu/~zhuh/
MenkveldYueshenZhu_2017JFE_dark.pdf; RBC Capital Letter at 2.
\375\ See, e.g., RBC Capital Letter at 2-4; Letter from Mehmet
Kinak, Vice President--Global Head of Systematic Trading & Market
Structure, and Jonathan Siegel, Vice President--Senior Legal Counsel
(Legislative & Regulatory Affairs), T. Rowe Price, to Brent J.
Fields, Secretary, Commission, dated June 12, 2018, at 2, available
at https://www.sec.gov/comments/s7-05-18/s70518-3832746-162769.pdf
(sec.gov) (commenting on File No. S7-05-18 ``Transaction Fee Pilot
for NMS Stocks).
\376\ See, e.g., Stanislav Dolgopolov, ``The Maker-Taker Pricing
Model and its Impact on the Securities Market Structure: A Can of
Worms for Securities Fraud?'' 8 Va. L. & Bus. Rev. 231, 270 (2014),
available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2399821 (retrieved from SSRN Elsevier
database).
---------------------------------------------------------------------------
a. Transparency
The chart below illustrates with a hypothetical example the
Commission's concern that exchange fee/rebate models can undermine
price transparency. While some investors may invest heavily to fully
map out the fee schedules, these schedules are complex and thus doing
so would be costly, consequently it is likely that not all investors
fully map out fee schedules.\377\
---------------------------------------------------------------------------
\377\ See infra section VII.C.2.c and table 4 therein for
additional analysis and discussion of the complexity of fee and
rebate schedules.
[GRAPHIC] [TIFF OMITTED] TR08OC24.000
The Commission examined data for NMS stocks to demonstrate the
price variations that can occur under the exchange fee and rebate
schedules. The chart above shows a common scenario for a hypothetical
stock, with four exchanges all displaying the same best bid of $20.56.
However, due to differing fee schedules, each of the four represents a
different net bid to market participants. Furthermore, due to volume-
based price fee tiers, some exchanges' net bids differ for different
market participants. The fee schedules used to create this example are
taken
[[Page 81647]]
from various exchange fee schedules as of June 2024. Exchange A charges
the maximum allowed access fee of 30 mils, and so market participants
that trade with the Exchange A bid receive a net price of $20.557 per
share ($20.56 minus a $0.003 fee). Exchange B has an inverted ``taker-
maker'' fee schedule and so pays a rebate to participants who remove
liquidity, with differing rebates based on volume tiers. In this
example, a market participant that trades with the bid on Exchange B
receives a net price of $20.5616 if in the best tier ($20.56 plus a
$0.0016 rebate). Exchange C charges an access fee of 29.5 mils, meaning
that a market participant that trades with the Exchange C bid would
receive a net price of $20.55705 per share ($20.56 minus the $0.00295
fee). Lastly, Exchange D charges an access fee of 29 mils to those in
the best tier, so such a participant trading with Exchange D receives a
net price of $20.5571. Despite all four exchanges showing the same bid
of $20.56, the bids net of fees vary from a low of $20.557 to a high of
$20.5616, a substantial difference of $0.0046 per share (nearly half
the current minimum pricing increment).
b. Liquidity and the NBBO
The price of liquidity for investors in terms of buying and then
later selling a security is the spread between the best bid and the
best offer, which is reflected by the NBBO. For those market
participants that provide liquidity, such as market makers, this spread
similarly represents the market price for providing those liquidity
services at any given point in time. More recently, trading center
models that pay rebates to liquidity providers (which rebates are
funded on a transaction basis by charging an access fee to the taker of
liquidity) pay an additional return to the liquidity provider separate
from what would be captured though the spread, which may then lead
those liquidity providers to lower the spread (that is, the implicit
price for their liquidity provision services) more than they would
otherwise.\378\ These prices are reflected in the NBBO, which is
disseminated in the national market system.
---------------------------------------------------------------------------
\378\ See also infra section VII.B.3.
---------------------------------------------------------------------------
Others have stated that the maker-taker model has positive effects
by enabling exchanges to compete with non-exchange trading centers and
by narrowing quoted spreads through subsidizing posted prices,\379\ but
these potential benefits should be balanced against the market
distortions associated with the fee and rebate models mentioned
above.\380\ Further, rebates paid to liquidity providers under maker-
taker fee schedules may narrow displayed spreads in some securities by
subsidizing liquidity providers (i.e., by allowing a maker to post a
more aggressive price than it may have in absence of a rebate), and
these prices may not reflect the underlying economics for the NMS
stock.\381\ In turn, that displayed liquidity may establish the
NBBO,\382\ which is often used as the benchmark for marketable order
flow, including retail order flow, that is executed off-exchange by
either matching or improving upon those distortive prices.\383\
Accordingly, rebates may distort quotation prices that are displayed in
the national market system.\384\
---------------------------------------------------------------------------
\379\ See, e.g., Letter from John A. Zecca, Executive Vice
President, Global Chief Legal, Risk and Regulatory Officer, Nasdaq,
Inc., dated Aug. 9, 2023 (``Nasdaq Letter II'') at 5-6; Larry
Harris, ``Maker-Taker Pricing Effects on Market Quotations,'' at 5
(Nov. 14, 2013), available at https://en-coller.tau.ac.il/sites/nihul_en.tau.ac.il/files/media_server/Recanati/management/seminars/account/Maker.pdf; Letter from Richie Prager, Managing Director,
Head of Trading and Liquidity Strategies, BlackRock, Inc., to Mary
Jo White, Chair, SEC, at 2 (Sept. 12, 2014), available at https://www.sec.gov/comments/s7-02-10/s70210-419.pdf; Michael Brolley &
Katya Malinova, ``Informed Trading and Maker-Taker Fees in a Low
Latency Limit Order Market,'' at 2 (Oct. 24, 2013), available at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2178102.
\380\ As discussed throughout, these concerns include, for
example, undermining price transparency, introducing unnecessary
complexity through the proliferation of new exchange order types and
order routing strategies, added market fragmentation, creating or
exacerbating potential conflicts of interest between brokers and
their customers, and in NMS stocks that are tick constrained,
assessing a higher cost to liquidity demanders.
\381\ See also infra sections VII.B.3 and VII.C.2.
\382\ See also infra sections VII.B.3 and VII.C.2.
\383\ See, e.g., Concept Release on Equity Market Structure,
supra note 59 (evaluating broadly the performance of market
structure since Regulation NMS, particularly for long-term investors
and for businesses seeking to raise capital, and soliciting comment
on whether regulatory initiatives to improve market structure are
needed).
\384\ See infra section VII.A and infra notes 1733 and 1734 and
accompanying text.
---------------------------------------------------------------------------
High rebates also incentivize excessive intermediation,\385\
especially in very liquid securities, to the extent rebates induce or
exacerbate an oversupply of liquidity at the best bid or offer.\386\ As
discussed below, some stocks will trade with a quoted spread one tick
wide, whether the tick size is $0.01 or $0.005. In those cases, quoted
spread is unable to adjust lower due to the tick size, so the rebate
paid to liquidity providers, which is funded by an access fee paid by
liquidity takers, acts as a wealth transfer from liquidity takers to
liquidity providers. This wealth transfer in turn unnecessarily
incentivizes the provision of liquidity (i.e., encourages providing
liquidity in order to earn the rebate), thereby creating an environment
with too much liquidity supplied relative to liquidity demanded and
leading to rebates for faster liquidity suppliers and a higher cost to
liquidity takers.\387\ In other words, excessive quoting in tick-
constrained securities to earn rebates undermines price transparency
because displayed prices and the associated size at those prices do not
reflect the underlying economics of supply and demand, but rather
reflect the impact of fees or rebates.\388\
---------------------------------------------------------------------------
\385\ See infra note 1005.
\386\ See Proposing Release, supra note 11, at 80292. See also
infra note 1005 and accompanying text.
\387\ See infra section VII.D.2.
\388\ See infra note 1005 and accompanying text.
---------------------------------------------------------------------------
c. Potential Conflicts of Interest
As more fully discussed below and in the Economic Analysis, access
fees and rebates create potential broker-dealer conflicts of interest
in order routing, particularly to the extent the fees and rebates are
not passed through to the customer.\389\ For example, this structure
can create an incentive for a broker-dealer that fully absorbs
transaction costs or rebates potentially to route customer orders to an
exchange in order to avoid fees that are paid by the broker-dealer or
to receive the highest rebate paid.\390\ Lowering the access fees caps
will help alleviate potential conflicts of interest.\391\
---------------------------------------------------------------------------
\389\ See infra sections IV.D.1.b and VII.D.2.c. See also
Proposing Release, supra note 11, at 80330.
\390\ See, e.g., Vanguard Letter at 6.
\391\ See infra section VII.D.2.d.
---------------------------------------------------------------------------
d. Market Complexity
Exchange fee and rebate models under preexisting Rule 610(c) have
also led to the development of a variety of complex order types,
including those that allow market participants to avoid paying access
fees or to ensure that rebates are collected.\392\ Further, exchange
fee and rebate models are another way exchanges differentiate
themselves from each other and from other trading centers to attract
order flow. The
[[Page 81648]]
variability of fee and rebate models (often within the same exchange
group) introduces additional market complexity and fragmentation
because it encourages the creation of new exchanges that offer
different pricing structures to attract different types of market
participants or trading strategies. Moreover, the drive to establish
novel and competitive fee schedules results in frequent fee schedule
changes (typically on a monthly basis), which adds uncertainty and
complexity to the marketplace because market participants must
continually update their routing tables to reflect these price
changes.\393\ A higher cap allows for a wider range of possible access
fees (i.e., $0-$0.0030) and more variability in exchange fees, which
introduces additional complexity to the market.\394\
---------------------------------------------------------------------------
\392\ Examples of such complex order types include post-only
orders or add-liquidity only orders that seek to only provide
liquidity to gain a rebate and will not upon entry, execute against
a resting order on the other side of the market so as to avoid
paying a transaction fee. See, e.g., BZX Rule 11.9(c)(6) (defining a
BZX Post Only Order); MEMX Rule 11.16(i)(6) (defining Post Only);
Nasdaq Equity, Rule 4702(4)(A)(defining a Post-Only Order); NYSE
Rule 7.31(e)(2) (defining an ALO Order); NYSE Arca Rule 7.31(e)(2)
(defining an ALO Order). See also Staff Report on Algorithmic
Trading in the U.S. Capital Markets (Aug. 5, 2020), available at
https://www.sec.gov/files/Algo_Trading_Report_2020.pdf.
\393\ See infra note 1081 and accompanying text.
\394\ See supra note 361 and infra note 1764.
---------------------------------------------------------------------------
C. Proposal To Amend 610(c)
To accommodate the amendments to Rule 612 and to address the issues
that have developed with the fee schedules for protected quotes under
Rule 610(c), the Commission proposed to amend Rule 610(c) by reducing
the level of the access fee caps for all protected quotations in NMS
stocks. For protected quotations in all NMS stocks priced $1.00 or
more, the proposal introduced two lower access fee caps to accommodate
the proposed minimum pricing increments under proposed Rule 612 for
quotations priced equal to or greater than $1.00 per share.\395\
Specifically, for all protected quotations in NMS stocks priced $1.00
or more per share and assigned a proposed minimum pricing increment
greater than $0.001, the Commission proposed a 10 mil access fee cap;
and for all protected quotations in NMS stocks priced $1.00 or more per
share and assigned the proposed $0.001 minimum pricing increment, the
Commission proposed a 5 mil access fee cap. For all protected
quotations in NMS stocks priced less than $1.00 per share, the
Commission proposed to reduce the access fee cap to 0.05% of the
quotation price.\396\
---------------------------------------------------------------------------
\395\ See Proposed Rule 610(c); Proposing Release, supra note
11, at 80269.
\396\ See Proposed Rule 610(c); Proposing Release, supra note
11, at 80269.
---------------------------------------------------------------------------
The Commission received many comments on the proposed amendments to
Rule 610(c). The comments are discussed more fully below.
D. Final Rule 610(c)
The amended access fee caps reflect several considerations by the
Commission as well as input from commenters. For protected quotations
priced $1.00 or more, the Commission is adopting a 10 mil access fee
cap. This level is appropriate based on several considerations. First,
because the Commission is adopting the $0.005 increment for certain NMS
stocks under amended Rule 612, it is also reducing the level of the
preexisting access fee caps in Rule 610(c) in order to prevent the
price distortions that would occur if access fees were able to be set
at more than half of the minimum pricing increment (i.e., a protected
quotation with an access fee that exceeds half the minimum pricing
increment economically would be represented at the next less aggressive
pricing increment).\397\ A 10 mil access fee cap is sufficiently below
the smallest minimum pricing increment (i.e., $0.005) so as to not
create new pricing distortions.\398\
---------------------------------------------------------------------------
\397\ See Proposing Release, supra note 11, at 80290. See also
infra notes 1423-1424 accompanying text and sectionVII.D.2.a. As
discussed in greater detail below, the amendments to Rule 612 do not
include the proposed $0.001 minimum pricing increment, as proposed.
Therefore, although the Commission proposed a 5 mil access fee cap
to correspond to the $0.001 minimum pricing increment, the 5 mil
access fee cap is not necessary because the Commission is not
adopting the proposed $0.001 minimum increment. See infra section
IV.D.1.a.
\398\ See infra section VII.D.2.a.
---------------------------------------------------------------------------
Second, in adopting the 10 mil access fee cap, the Commission also
considered the impact of the reduction on the agency market business
model. A 10 mil access fee cap is at a level that will allow the
exchanges to maintain their current net capture for executions of NMS
stocks priced $1.00 or greater.\399\
---------------------------------------------------------------------------
\399\ See infra sections IV.D.1.c and VII.D.2.b.
---------------------------------------------------------------------------
The 10 mil access fee cap also should help to reduce distortions
and complexities under the fee structures that have developed under the
preexisting access fee caps levels.\400\ As discussed above, under the
preexisting access fee caps, many exchanges charge access fees at or
near the highest permitted levels under preexisting Rule 610(c) as a
means to fund rebates.\401\ Access fees charged at the highest level
permitted under the preexisting rule and the rebates they fund harm
price transparency because the displayed price does not accurately
reflect the underlying economics of a decision to post a protected
quotation at a particular price (i.e., the market participant's
assessment of the price of liquidity for the security is distorted by
the subsidy provided by the rebate), or to access a protected quotation
at a particular price (i.e., the displayed price does not reflect the
additional cost of the access fee).\402\ The negative impact on price
transparency is exacerbated when various exchanges have different fees
and rebates and make frequent updates to those rates, making the
comparison of net prices unnecessarily complex and difficult.\403\
---------------------------------------------------------------------------
\400\ See infra sections IV.D.1.c. and IV.D.1.d.
\401\ See Proposing Release, supra note 11, at 80288. See also
infra sections IV.D.1.b.and VII.C.2.c, specifically table 4. While
Rule 610(c) limits the fees assessed against an incoming order that
executes against a resting protected quote, it does not address the
rebates that may be paid.
\402\ See infra section VII.D.2.a.
\403\ See infra note 1096 and adjacent text and section
VII.C.2.c, table 4 (analyzing fee and rebate schedules and stating
the current structure of fees and rebates is complex and constantly
changing).
---------------------------------------------------------------------------
Further, reducing the level of the access fee caps to the adopted
levels will reduce complexity in the market because it will (1) reduce
the incentives to use certain complex order types that are designed to
avoid high fees/garner large rebates, (2) potentially reduce the number
of fee changes in the market and accompanying frequent changes to
complex order routing strategies, and (3) may discourage further market
fragmentation.\404\ Reducing the amount of rebates by reducing the
access fee caps to 10 mils also will reduce the magnitude of potential
conflicts of interest in the market.
---------------------------------------------------------------------------
\404\ See infra section VII.D.2.d.
---------------------------------------------------------------------------
Finally, in determining to adopt the 10 mil access fee cap, the
Commission has considered input from commenters, including market
participants. Commenters stated that the reduced 10 mil access fee cap
will better reflect current market rates and the increased efficiencies
from electronic trading and other market structure changes, all of
which have reduced trading costs since Rule 610 was originally
adopted.\405\
---------------------------------------------------------------------------
\405\ See, e.g., Better Markets Letter I at 16; Brandes Letter
at 3; Ontario Teachers et al. Letter at 1-2; IEX Letter IV at 7-8;
Verret Letter III at 5-6; Letter from John Ramsay, Chief Market
Policy Officer, Investors Exchange LLC, dated Apr. 19, 2024 (``IEX
Letter VI'') at 4. See also supra IV.D.1 (discussing comment
letters).
---------------------------------------------------------------------------
In addition, the Commission is retaining the uniform access fee cap
structure whereby the access fee caps are assigned based solely upon
the price of the protected quotation. In other words, the Commission is
adopting the 10 mil access fee cap for all protected quotes priced
$1.00 or more. Maintaining the preexisting uniform structure of the
access fee caps helps ensure that the requirements under Rule 610(c) do
not increase the fee structure complexity or introduce unintended
consequences (such as oscillations) that would create additional costs
for market participants.\406\
---------------------------------------------------------------------------
\406\ See infra section VII.D.2.d.
---------------------------------------------------------------------------
[[Page 81649]]
Further, the Commission is also reducing the level of the access
fee cap for NMS stocks priced under $1.00 to 0.1% of the quotation
price in order to maintain the preexisting structure as well as to
harmonize that cap with the adopted 10 mils cap for NMS stocks priced
at $1.00 or more. The harmonization will prevent different fees on
quotes above and below $1.00 that could negatively impact price
formation.\407\ These considerations are consistent with the analysis
and rationale the Commission used when it adopted the access fee caps
in 2005.\408\
---------------------------------------------------------------------------
\407\ See infra section VII.D.2.c.
\408\ See generally, Regulation NMS Adopting Release, supra note
4.
---------------------------------------------------------------------------
Lowering the access fee caps to these levels and maintaining the
preexisting uniform structure will promote the statutory objectives of
fair and efficient access to protected quotes and will help to ensure
the fairness and usefulness of protected quotations \409\ because the
amended access fee caps will lead to transaction pricing that is better
aligned with today's market dynamics.\410\ Recalibrating the level of
the caps will yield savings for investors and help to address
distortions in the markets while maintaining the ability of the trading
centers that display protected quotations to continue to provide
execution services, innovate and compete because they will be able to
retain the same fees (net of any rebates) for executions that are
priced $1.00 or more,\411\ notwithstanding the reduction of the fee cap
to 10 mils.\412\ The Commission has balanced the competing interests of
reducing the cap to address the amendments to Rule 612 and market
distortions associated with the fee/rebate models that have developed
under the preexisting fee caps, with the importance of preserving the
viability of multiple agency market business models. The amended 10
mils access fee cap will lead to improved market quality because it
will reduce distortions in the market and preserve the integrity of
displayed prices, which will support sufficient price discovery, and
will reduce costs for investors.\413\
---------------------------------------------------------------------------
\409\ Section 11A(c)(1)(B) of the Exchange Act. See Regulation
NMS Adopting Release, supra note 4, at 37545.
\410\ See infra section VII.D.2.b. As discussed below, the level
of the adopted access fee cap for protected quotes in NMS stocks
priced $1.00 or more is consistent with the current level charged by
some trading centers that do not post protected quotes and therefore
are not subject to the preexisting access fee caps under Rule
610(c). Specifically, the fees charged by some ATSs for execution
services are often in the range of 10 mils for access to their
liquidity. See infra note 1118 and accompanying text and infra
section VII.C.2. This level reflects a competitive rate for
providing access to liquidity, and hence represents a reasonable
level for amending the access fee caps for protected quotes. See
also e.g., Goldman 2018 Letter at 4 (stating ``a reduction of the
Fee Cap to $0.0010 per share is reasonable and would be better
calibrated with today's market pricing.''). According to one
commenter, while the ``overwhelming proportion of transaction volume
executed on national stock exchanges is subject to the maximum
access fee of 30 mils. . .volume executed on ATS's and other venues
outside of exchanges is typically subject to substantially lower
costs of access, in the range of ten mils and lower.''). IEX Letter
IV at 8.
\411\ See also infra sections VII.C.2 and VII.D.2, and table 14.
As discussed below, the Commission estimates for purposes of this
release that exchange net capture is 2 mils while also recognizing
that net capture can range from approximately 2 to 6 mils. See infra
note 1103.
\412\ See infra section IV.D.1.c. See also Proposing Release,
supra note 11, at 80290-91.
\413\ See infra section VII.D.2. See also Proposing Release,
supra note 11, at 80292 and at 80309 (stating because ``compensation
is above what would exist in a competitive market there is an
increased incentive to provide liquidity via limit orders, so queues
of limit orders tend to be longer, wait times to get a limit order
executed also tend to be longer, and, thus the likelihood that the
market moves away from an investor's limit order increases, leading
to lower overall fill rates for limit orders'') and at 80329
(stating ``[t]he primary beneficiaries of the reduction in the
access fee cap would be liquidity demanders. For stocks with narrow
spreads such as tick-constrained stocks, a 30 mil access fee can
increase the cost of demanding liquidity by as much as 60%.
Consequently, reducing the access fee significantly reduces the cost
of demanding liquidity in the predominant maker-taker trading
environment. This effect coupled with the expected decrease of
liquidity suppliers can be expected to decrease competition to
provide liquidity. Less competition to provide liquidity means that
queue lengths could decrease and fill rates increase because it
would be easier to get to the front of the order book. This effect
could allow non high frequency traders more opportunity to fill
orders using liquidity-providing instead of liquidity-demanding
transactions.''). See also IEX Letter IV at 2, 6-8.
---------------------------------------------------------------------------
1. Comments on Proposed Rule 610(c)
Many comments from a broad cross section of market participants
supported the need to reduce the access fee caps in preexisting Rule
610(c).\414\ Many individual commenters stated that the reduced access
fee caps would help to reduce trading costs.\415\ One individual
commenter stated ``[a]s a retail trader, I have experienced the high
costs of access fees, which can be a significant portion of my trading
costs. The proposed reduction in access fee caps will help to lower my
costs, which will allow me to take advantage of more trading
opportunities and ultimately benefit from a more efficient market.''
\416\ Other commenters also stated that a reduction in the access fee
caps would reduce trading costs for investors.\417\
---------------------------------------------------------------------------
\414\ See, e.g., IEX Letter IV at 1; IEX Letter VI at 1; Better
Markets Letter II at 1; We The Investors Letter dated Mar. 30, 2023
at 2; Letters from Chris Robinson, dated Mar. 3, 2023; William
Bledsoe, Jr., dated Feb. 28, 2023; Ryan Macarthur, dated Feb. 24,
2023; Julio Tello, dated Feb. 24, 2023; David Genco, Jr., dated Feb.
24, 2023; John, dated Feb. 23, 2023; Citigroup Letter at 3;
BlackRock Letter at 10-11; Better Markets Letter I at 16; ASA Letter
at 5; Vanguard Letter at 2; Invesco Letter at 2; JPMorgan Letter at
6; Ontario Teachers et al. Letter at 1-2; Budish Letter at 1; Mark
Rogers Letter, dated Mar. 30, 2023; Grant Medford Letter, dated Mar.
30, 2023 ; Jared Albert Letter, dated Mar. 28, 2023; Steven Tripari
Letter (``Tripari Letter''), dated Mar. 28, 2023; Peter McKornack
Letter, dated Mar. 29, 2023; Verret Letter III at 26.
\415\ See, e.g., Julio Tello, dated Feb. 24, 2023; Ryan
Macarthur, dated Feb. 24, 2023; David Genco, Jr., dated Feb. 24,
2023; John, dated Feb. 23, 2023; Budish Letter at 1.
\416\ Julio Tello, dated Feb. 24, 2023.
\417\ See, e.g., Ontario Teachers et al. Letter at 2; Brandes
Letter at 3; Boston Partners, Calamos Advisors, Glenmede Investment,
and Janus Henderson Letter.
---------------------------------------------------------------------------
Some commenters stated that the proposal would enhance transparency
\418\ and would reduce complexity.\419\ As discussed above, the reduced
access fee caps will enhance transparency of protected quotes and
reduce complexity in the market by reducing the need for complex order
types that have developed to accommodate the fee/rebate models.
---------------------------------------------------------------------------
\418\ See, e.g., Grant Medford Letter, dated Mar. 30, 2023;
Verret Letter III at 24; BlackRock Letter at 10-11.
\419\ See, e.g., Budish Letter at 1; Boston Partners, Calamos
Advisors, Glenmede Investment, and Janus Henderson Letter.
---------------------------------------------------------------------------
Among the commenters that supported a reduction in the access fee
caps,\420\ several stated that the amendments to Rule 612 to reduce
tick sizes would necessitate a reduction in the access fee caps for
those securities assigned a smaller tick,\421\ while many others stated
that the preexisting access fee cap should be lowered to 10 mils per
share for protected quotations in all NMS stocks priced at $1.00 or
more regardless of whether there was a reduction in tick size.\422\ As
discussed
[[Page 81650]]
below, the Commission has decided to maintain the uniform access fee
cap structure and continue to apply the caps to all protected quotes
based on price. The Commission has decided that reducing the caps for
all protected quotes, not just those that may be assigned the smaller
$0.005 minimum pricing increment, is appropriate to address the
distortions that exist in the national market system. Further,
maintaining the uniform structure will help to ensure that the rule
does not increase complexity in the national market system.
---------------------------------------------------------------------------
\420\ See, e.g., IEX Letter IV at 1; MEMX Letter at 22;
Biotechnology Innovation Organization Letter at 1; Brandes Letter at
3; Letter from Allison Bishop, President, Proof Services LLC, dated
Mar. 31, 2023 (``Proof Letter'') at 1; BlackRock Letter at 10-11;
Verret Letter I at 1, 2 and 4; MFA Letter at 13.
\421\ SIFMA Letter II at 39; Invesco Letter at 4; Nasdaq Letter
I at 19; Better Markets Letter II at 3-4.
\422\ IEX Letter IV at 1; IEX Letter I at 6 and 21; Better
Markets Letter I at 16; Better Markets Letter II at 4; Ontario
Teachers et al. Letter at 2; Brandes Letter at 3; Boston Partners,
Calamos Advisors, Glenmede Investment, and Janus Henderson Letter
(supporting Brandes' Letter); Healthy Markets Letter I at 24; Themis
Letter at 7-8; Letter from Andrew Hartnett, President and Deputy
Commissioner, Iowa Insurance Division, North American Securities
Administrators Association, Inc., dated Mar. 31, 2023 (``NASAA
Letter'') at 1 and 9; ASA Letter at 5; Vanguard Letter at 6;
JPMorgan Letter at 6; Capital Group Letter at 4; Pragma Letter at 7;
Invesco Letter at 4; Verret Letter I at 8; XTX Letter at 5; Letter
from Jeffrey P. Mahoney, General Counsel, Council of Institutional
Investors, dated Mar. 30, 2023 (``Council of Institutional Investors
Letter'') at 3; BMO Letter at 3; and BlackRock Letter at 10-11.
---------------------------------------------------------------------------
Several commenters stated that modifications to reflect the
significant evolution in market conditions since the caps were
established almost two decades ago are long overdue.\423\ One commenter
stated that the current access fee caps are ``antiquated.'' \424\
Another commenter stated that the access fee caps are ``outdated'' and
``counter to the interests of long term investors.'' \425\ The
Commission is reducing the access fee caps to accommodate the
amendments to Rule 612. Further, retaining a uniform access fee cap
structure will benefit the market by introducing less complexity and
help to address market distortions that have arisen under the current
fee caps. The current market structure has experienced significant
changes in trading dynamics and operates under a fee structure that is
different from when Rule 610(c) was adopted and problematic for the
reasons articulated throughout this release. As discussed above and in
the Economic Analysis, the amendment modernizes Rule 610(c) to reflect
current trading dynamics and mitigate distortions associated with the
preexisting caps while preserving its original objectives.
---------------------------------------------------------------------------
\423\ See, e.g., IEX Letter IV at 5; Brandes Letter at 3;
Citigroup Letter at 5; BlackRock Letter at 10-11; DOJ Letter at 5;
Harris Letter at 1; BMO Letter at 4.
\424\ IEX Letter IV at 5. See also IEX Letter III at 5.
\425\ Brandes Letter at 3.
---------------------------------------------------------------------------
Some individual commenters recommended that the proposal be
modified to go further to address distortions in the market related to
the current fee and rebate models.\426\ A few commenters stated that
the access fee caps should be eliminated entirely.\427\ One commenter
stated that ``competitive forces should inform access fees.'' \428\
Another commenter stated that the ``solution is to let brokers take
fees into consideration in their order routing . . . and route to the
market with the best all-in costs,'' which would obviate the need for
the Commission to ``get into the price control business.'' \429\
---------------------------------------------------------------------------
\426\ See, e.g., Tripari Letter, dated Mar. 28, 2023; Francisco
Gil, dated Mar. 28, 2023; We The Investors Letter dated Mar. 30,
2023 at 7; Adam Abreu, dated Apr. 30, 2023 at 2; Michael Dudek,
dated Mar. 31, 2023 at 4; Larry Douglas, dated Apr. 1, 2023 at 1 and
3; We The Investors Letter I dated Mar. 15, 20234; Betty Waters
Letter, dated Mar. 31, 2023; Harris Letter at 1 & 4.
\427\ See Cboe Letter II at 2 and 8; Angel Letter at 7.
\428\ Cboe Letter II at 2 and 8. See also Angel Letter at 7.
\429\ See Angel Letter at 7.
---------------------------------------------------------------------------
While some commenters suggested that access fees be further reduced
or eliminated altogether, the Commission is not eliminating fees for
access to protected quotes. Rule 610(c) establishes limits on the
amount of fees that can be charged for access and when the Commission
adopted Rule 610(c), the Commission recognized that agency market
trading centers have historically relied, at least in part, on charging
fees for access. Eliminating or prohibiting access fees entirely would
unduly harm the business model of agency market trading centers by not
allowing them to collect fees for the execution services they
provide.\430\ As discussed below, the national securities exchanges are
the only trading centers at this time that display protected quotations
and they should be able to continue to charge for the execution
services they provide.\431\
---------------------------------------------------------------------------
\430\ See Regulation NMS Adopting Release, supra note 4, at
37544-45 (stating the Commission considered market participants'
views stating that agency markets must be allowed to charge access
fees for their services, as well as those that stated that access
fees distort quotation prices and should be banned. In adopting the
30 mil access fee cap, the Commission recognized that ``agency
trading centers perform valuable agency services in bringing buyers
and sellers together, and that their business model historically has
relied, at least in part, on charging fees for execution of orders
against their displayed quotations.'' The Commission concluded that
``prohibiting access fees entirely would unduly harm this business
model.''). See infra sections VII.C.2.b and VII.D.2.b.
\431\ Id. See also infra section IV.D.1.c.
---------------------------------------------------------------------------
However, while recognizing the importance of the agency market
business model to the national market system, in adopting the access
fee caps the Commission also was mindful that ``[a]ccess fees tend to
be highest when markets use them to fund substantial rebates to
liquidity providers, rather than merely to compensate for execution
services'' and artificially high access fees (i.e., those that are used
primarily to fund rebates) can undermine price discovery because ``the
published quotations of such markets would not reliably indicate the
true price that is actually available to investors or that would be
realized by liquidity providers.'' \432\
---------------------------------------------------------------------------
\432\ See Regulation NMS Adopting Release, supra note 4, at
37545.
---------------------------------------------------------------------------
Further, notwithstanding commenters' statements to the contrary,
the access fee caps continue to be necessary in order to support the
objectives of fair and efficient access to protected quotations, which
``is necessary to support the integrity of the price protection
requirement established by the adopted the Order Protection Rule.''
\433\ In adopting Rule 611, the Commission stated that strong
intermarket price protection offers greater assurance that investors
who submit market orders will receive the best readily available prices
for their trades.\434\ Rule 611 was designed to ``strengthen the
protection of displayed and automatically accessible quotations in NMS
stocks.'' \435\ The Commission recognized that such objectives could be
undermined if ``outlier'' markets could charge high fees to market
participants who would be required to pay such high fees to access a
protected quotation because of Rule 611. The comments that suggest
eliminating the access fee caps and allowing a consideration by brokers
of all-in costs or relying solely on competition, do not address the
concern that led to the adoption of the access fee caps, namely that
outlier markets would take advantage of Rule 611 by imposing high fees
for access.\436\ The access fee caps remain necessary for the reasons
they were adopted.
---------------------------------------------------------------------------
\433\ See Regulation NMS Adopting Release, supra note 4, at
37502-37503 (stating ``protecting the best displayed prices against
trade-throughs would be futile if broker-dealers and trading centers
were unable to access those prices fairly and efficiently.'') See
also IEX Letter V at 2 (stating proposal is designed to ``prevent
high fees from undermining Regulation NMS's price protection
privileges afforded to exchanges'').
\434\ See Regulation NMS Adopting Release, supra note 4, at
37501.
\435\ See Regulation NMS Adopting Release, supra note 4, at
37501.
\436\ See supra note 355 and accompanying text.
---------------------------------------------------------------------------
Some commenters stated that the Commission should act to prohibit
rebates.\437\ While in many cases rebates
[[Page 81651]]
are funded by the access fees that are collected, Rule 610(c) does not
apply to rebates.\438\ The Commission is adopting amendments to Rule
610(c) to reduce the access fee caps for the reasons discussed herein,
but is not expanding its application to apply to rebates or to
eliminate them.\439\ The adopted amendments to Rule 610(c) maintain
fidelity to the original objectives of the rule, but recalibrate the
fee cap amounts to reflect current market structure. As a practical
matter, however, the reduced access fee caps in amended Rule 610(c)
will likely reduce the rebates paid \440\ and, as a result, the amended
access fee caps will reduce the distortions created by the existing fee
structures that use access fees as a means to fund the payment of
rebates in the market.\441\
---------------------------------------------------------------------------
\437\ See, e.g., Steven Tripani, dated Mar. 28, 2023; Michael
Dudek, dated Mar. 31, 2023 at 4; Betty Waters Letter dated Mar. 31,
2023. One commenter suggested that the Commission should prohibit or
restrict the use of CADV-tiers. We the Investors Letter, dated Mar.
30, 2023 at 7. As discussed below, the Commission is adopting Rule
610(d) as proposed which will enhance the transparency of fees and
rebates, including fees that may be tiered. The Commission notes
that it continues to assess volume-based exchange transaction
pricing. See Securities Exchange Act Release No. 98766 (Oct. 18,
2023), 88 FR 76282 (Nov. 6, 2023) (proposing new rule 6b-1 under the
Exchange Act, which would prohibit exchanges from offering volume-
based transaction pricing in connection with the execution of agency
or riskless principal orders in NMS stocks) (``Fee Tiers
Proposal''). As discussed below, Rule 610(d) will provide certainty
regarding the amount of the fee to be assessed and the rebate to
paid at the time of the time of the trade, which is separate and
distinct from the Commission's consideration of other regulatory
action regarding volume-based transaction pricing. See infra section
IV.E.
\438\ See supra section IV.B.2. See also infra section IV.D.1.b.
\439\ See infra note 590 and accompanying text.
\440\ See infra section VII.C.2.
\441\ See infra section IV.D.1.b and VII.D.2.
---------------------------------------------------------------------------
One commenter stated that the Commission should require the
exchanges to ``revert to traditional fees imposed on buyers or sellers
or both without regard to the maker-taker status.'' \442\ Rule 610(c)
does not require any particular fee structure, like a flat fee
structure suggested by a commenter.\443\ Instead, the access fee caps
set an upper limit on the amount of fees that can be charged for access
to protected quotations. Within this construct, trading centers can
continue to develop fee structures that are consistent with Rule 610 as
well as any other regulatory requirements that may be relevant to a
particular trading center.\444\
---------------------------------------------------------------------------
\442\ See Harris Letter at 1 and 4.
\443\ See Harris Letter at 4.
\444\ National securities exchanges establish and amend their
fee schedules by filing proposed fee rule changes, pursuant to
section 19(b) of the Exchange Act and rule 19b-4 thereunder, for
Commission review. See 15 15 U.S.C. 78f(b)(4) and (5)(requiring the
rules of the exchange provide for the equitable allocation of
reasonable dues, fees, and other charges among members, and issuers
and other persons using its facilities and not be designed to permit
unfair discrimination).
---------------------------------------------------------------------------
a. Access Fees and Minimum Pricing Increments
Several commenters stated the Commission should lower the access
fee caps regardless of whether any changes are made to the minimum
pricing increments and stated that such changes are warranted even if
the Commission elects not to proceed with the proposed tick
changes.\445\ One commenter stated ``the current harms associated with
the $0.003 access fee cap and maker-taker pricing models exist at the
current tick sizes. Accordingly, the Commission should consider
reducing the access fee cap even if it ultimately decides not to
proceed with the proposed tick size changes.'' \446\ Another commenter,
however, ``strongly disagreeing'' with commenters' suggestion that
regulatory reform of exchange fees could proceed independently, stated
that ``[c]alls for regulatory mandated reductions in exchange access
fee caps fail to recognize that exchange access fee caps were adopted
and justified to facilitate effective intermarket linkages.'' \447\
---------------------------------------------------------------------------
\445\ See, e.g., Healthy Markets Letter I at 24; Vanguard Letter
at 6. See also supra note 422.
\446\ See Vanguard Letter at 6.
\447\ Cboe Letter III at 1-2.
---------------------------------------------------------------------------
Other commenters stated that the levels of the access fee caps and
the minimum pricing increments are connected.\448\ One commenter stated
``[a] reduction in the minimum tick size without reducing access fees
could permit fees to become a higher percentage of the minimum pricing
increment, which would almost certainly undermine price transparency.''
\449\ Commenters stated that the access fee caps should be reduced to
help ensure that access fees do not become an outsized portion of
displayed quotes in light of proposed changes to the minimum pricing
increments.\450\ One commenter stated that it was ``recommend[ing]
keeping fees strictly less than \1/2\ of the tick size'' in order to
prevent ``price instability and quote flickering'' and stated ``[i]f
fees reach \1/2\ the tick size, it means that the same effective price
point can be achieved multiple ways on an all-in basis with different
nominal prices (e.g., an offer at 10.000 with a $0.0005 rebate, or a
bid of 10.001 with a $0.0005 rebate).'' \451\ Another commenter stated
``if the access fee cap were to exceed half of the tick size, the paper
trail can be not only confusing but can literally misrank trades.''
\452\
---------------------------------------------------------------------------
\448\ See, e.g., SIFMA Letter II at 39 (stating ``the Commission
appropriately recognized that tick sizes and access fees are linked
with each other''); Nasdaq Letter I at 19 (``support[ing] adjusting
the access fee cap to accommodate new tick sizes''); Better Markets
Letter II at 3-4; MEMX Letter at 22 (``access fees and tick sizes
are inherently linked'').
\449\ Better Markets Letter II at 3-4.
\450\ See, e.g., Invesco Letter at 4; NASAA Letter at 9;
JPMorgan Letter at 6; Better Markets Letter II at 4; Healthy Markets
Letter I at 22; Pragma Letter at 7; Budish Letter at 6; AIMA Letter
at 4.
\451\ Pragma Letter at 7.
\452\ Budish Letter at 6.
---------------------------------------------------------------------------
As recognized by several commenters, access fees and tick sizes are
related in certain instances.\453\ Specifically, an access fee that is
too high when compared to the tick size can create pricing
distortions.\454\ Therefore, because the Commission is reducing the
minimum pricing increments for certain NMS stocks as set forth in Rule
612, the Commission is also reducing the Rule 610(c) access fee caps to
prevent introducing pricing distortions that can occur if an access fee
is greater than one-half of the tick. Maintaining an access fee cap
that is less than one-half of the tick size will preserve coherence
\455\ and result in lower transaction costs for investors.\456\
---------------------------------------------------------------------------
\453\ See, e.g., MEMX Letter at 22; SIFMA Letter II at 39;
Better Markets Letter II at 3-4; Nasdaq Letter I at 19. See also
infra section VII.D.2.
\454\ See infra section VII.D.2.a.
\455\ See infra section VII.D.2.a.
\456\ See id.
---------------------------------------------------------------------------
Further, lowering the access fee cap to 10 mils for those NMS
stocks assigned a lower tick will address the distortionary effect on
price transparency that would result absent adjustment to the access
fee cap.\457\ To illustrate, if the access fee cap remained at $0.003,
which would fund rebates at a similar level, and the tick size was
adjusted to $0.005, the effect on the price of a stock could be as
follows. An executed trade could be displayed at a price of $10.005
followed by another executed trade at a price of $10.010. Many
investors would interpret this as a sign that a stock was increasing in
value. However, with an access fee of $0.003, the net price of the
first order if it represents a market order to buy would be $10.008
(the buyer pays $10.005 and pays $0.003 in fees), whereas the net price
of the second order if it is a market order to sell would be $10.007
(the seller receives $10.010 and pays $0.003 in fees). In this example,
the price has fallen, not risen. Lowering the access fee cap to 10 mils
will mitigate this problem.\458\
---------------------------------------------------------------------------
\457\ See id.
\458\ See id.
---------------------------------------------------------------------------
The Commission also agrees that the access fee caps should be
lowered for all NMS stocks regardless of whether a stock is assigned
the lower pricing increment of $0.005 or retains a $0.01 minimum
pricing increment to address market distortions attributable to the fee
structures that have developed under the access fee caps and align the
fee caps with current market dynamics. The Commission is reducing the
access fee caps for all NMS stocks and maintaining the structure that
was originally adopted, i.e., assigning an access fee cap based on the
price of the protected quotation. And, as discussed above, maintaining
a uniform access fee cap structure will help to ensure that the
requirements under Rule 610(c) do not increase the fee structure
complexity.
Some commenters were generally supportive of an access fee cap
[[Page 81652]]
reduction because of the changes to the minimum pricing increment, but
offered no or few specifics as to the amount or percentage of the
reduction.\459\ Other commenters urged the Commission to reduce the
access fee caps in a manner that is proportionate to any reduction in
the tick (e.g., for a $0.005 tick, the access fee cap should be 15
mils).\460\
---------------------------------------------------------------------------
\459\ See, e.g., AIMA Letter at 4; Cambridge Letter at 6; ICI
Letter I at 16; STA Letter at 8.
\460\ See, e.g., Pragma Letter at 7; MFA Letter at 13; Schwab
Letter II at 6 and 36; SIFMA Letter II at 45; MEMX Letter at 22;
Jefferies Letter at 2; NYSE Letter I at 7; Nasdaq Letter I at 29;
NYSE, Schwab, and Citadel Letter at 1-2.
---------------------------------------------------------------------------
Some commenters stated that the access fee caps should be 30% of
the minimum pricing increment in order to remain consistent with the
percentage level under the preexisting rule.\461\ These commenters
recommended maintaining the current 30% ratio between tick size and
access fee cap because they were primarily concerned that the proposal
would result in access fees that were 50% of the tick (an increase in
fee to tick ratio) for the smallest proposed tick size (i.e., the
proposed $0.001 tick would have been assigned a 5 mils access fee cap).
Other commenters expressed concern regarding application of a two-
tiered access fee cap structure to a four-tiered reduction in minimum
pricing increments.\462\ These commenters' concerns regarding both the
increase in fee to tick ratio and asymmetrical structure of the
proposal, however, have been obviated because the Commission is neither
adopting the $0.001 tick, nor the corresponding 5 mils cap.
---------------------------------------------------------------------------
\461\ See, e.g., FIA PTG Letter II at 3; Hudson River Letter at
4; MMI Letter at 7; Robinhood Letter at 46, 56-59. See also MEMX at
22.
\462\ See, e.g., JPMorgan Letter at 6; MFA Letter at 13.
---------------------------------------------------------------------------
Other commenters stated that the access fee caps should be reduced
only for NMS stocks that are assigned a smaller minimum pricing
increment and only in proportion to the amount of a stated
corresponding decrease in the tick size, i.e., NMS stocks that were
assigned a smaller $0.005 tick size would be subject to a 15 mils
access fee cap.\463\ Similarly, several exchange groups commented on
the proposal and offered alternative approaches to modify Rule 610(c)
to reflect the change in minimum pricing increments.\464\ One exchange
group commenter supported the need to adjust the access fee caps to
accommodate the proposed new tick sizes, but stated its view that the
proposal went ``far beyond what is needed'' to achieve that purpose
``to the detriment of market quality and the NBBO'' because reducing
the caps would implicitly reduce rebates, which would impede exchanges'
ability to attract liquidity and encourage tighter spreads.\465\ As an
alternative to the proposal, the commenter recommended the Commission
adopt a fee cap of 15 mils for NMS stocks assigned to a $0.005 minimum
pricing increment and retain the preexisting 30 mils access fee cap for
NMS stocks that retain the $0.01 minimum pricing increment.\466\
According to this commenter, this alternative ``would cut access fees
by half for securities in the $0.005 tick bucket, while preserving room
for exchanges to continue [to] offer rebates that are needed to bolster
market quality and the NBBO.'' \467\ This commenter further stated that
although the Commission intends the proposal to help the exchanges
compete for retail order flow by reducing the cost for broker-dealers
to access liquidity on the exchange, compressing the access fee caps
would make it more expensive to provide liquidity to the exchanges and
thus any benefit would be undermined.\468\
---------------------------------------------------------------------------
\463\ See, e.g., Fidelity Letter at 14-15; Cambridge Letter at
6; NYSE, Schwab, and Citadel Letter at 2; MEMX Letter at 23;
Jefferies Letter at 2; Schwab Letter II at 6.
\464\ See Nasdaq Letters I and II; NYSE Letter I; and Cboe
Letters I-IV.
\465\ Nasdaq Letter I at 2 and 19.
\466\ Nasdaq Letter I at 2, 19 and 29. See also NYSE Letter I at
7.
\467\ Nasdaq Letter I at 2.
\468\ Nasdaq Letter I at 20.
---------------------------------------------------------------------------
Several commenters stated that there should be a uniform fee cap to
``avoid any additional market complexity.'' \469\ Some commenters
stated that the proposed tiered access fee caps and the proposed
variable minimum pricing increments would add unnecessary
complexity.\470\ One commenter stated that it ``strongly favor[ed] a
single, consistent standard, rather than multiple caps tied to
different ticks, which would create unnecessary complexity.'' \471\
Another commenter stated that applying a uniform cap across all NMS
stocks would help to address market distortions such as routing
conflicts arising from the maker-taker fee model.\472\ Another
commenter stated that continuing to apply a uniform cap will more
effectively achieve the objectives of Rule 610 because absent such
adjustment, ``the ability of exchanges to abuse their status as
protected markets will be no less for stocks that are assigned a higher
tick increment.'' \473\ One commenter stated that setting different fee
caps based on tick size would ``allow exchanges to impose a `penalty
fee' for participants looking to access quotes in stocks that are less
actively traded'' and further stated that ``there is no justification
in logic or regulatory purpose to make that distinction.'' \474\
However, other commenters disfavored a ``one-size-fits-all''
model.\475\ One commenter suggested the Commission consider a ``dynamic
tick size approach with the access fee cap proportionally tied to both
smaller and larger tick sizes'' \476\ and recommended the access fee
caps be a certain percentage of the minimum pricing increment, but did
not propose a particular percentage that should be adopted.\477\
---------------------------------------------------------------------------
\469\ Better Markets Letter I at 16. See also e.g., Better
Markets Letter II at 4; Brandes Letter at 3; BlackRock Letter at 11;
JPMorgan Letter at 6; Invesco Letter at 4.
\470\ See, e.g., JPMorgan Letter at 6; Brandes Letter at 3.
\471\ Brandes Letter at 3.
\472\ Capital Group Letter at 4.
\473\ See, e.g., IEX Letter VI at 2.
\474\ IEX Letter VI at 2.
\475\ See, e.g., T. Rowe Price Letter at 4; BlackRock Letter at
10-11; Citigroup Letter at 5-6; Letter from Phil Mackintosh, Nasdaq,
Inc., dated May 7, 2024, at 2 (``Nasdaq Letter V'').
\476\ Morgan Stanley Letter at 3.
\477\ Morgan Stanley Letter at 3-4.
---------------------------------------------------------------------------
Other commenters disagreed that the level of the access fee cap
should be tied to the pricing increment assigned. One exchange
commenter stated that the ``original justification for access fee caps
had nothing to do with tick sizes'' and instead ``centered around
ensuring that transaction fees did not unduly distort the price of a
quote that the Commission was protecting by rule [611].'' \478\ This
commenter also stated that ``[m]odifications to access fee caps should
only be discussed in the context under which they were conceived''
which was ``to ensure that market centers displaying the best price did
not impose access fees that compromised the value of the better
price.'' \479\
---------------------------------------------------------------------------
\478\ See Cboe Letter IV at 2.
\479\ See Cboe Letter IV at 2.
---------------------------------------------------------------------------
The Commission agrees that one of the purposes of the access fee
cap was, and remains, to help to ensure that transaction fees do not
unduly distort the price of protected quotations. However, the
Commission does not agree that goal is best achieved by adopting
certain commenters' recommendation to reduce the access fee caps
proportionally (i.e., to 15 mils) and only for those NMS stocks that
are assigned a smaller minimum pricing increment. A proportional
reduction for a limited universe of NMS stocks would allow higher
access fees and rebates along with related market distortions to
continue for the NMS stocks that retain the $0.01 increment and would
perpetuate unwarranted complexity (e.g., complex orders types, market
[[Page 81653]]
fragmentation, complex fee and rebate schedules and frequent changes to
complex order routing strategies to adjust to fee changes) to the
current market structure.\480\ Applying a uniform 10 mil access fee cap
to access protected quotations in all NMS stocks priced $1.00 or
greater will avoid injecting complexity \481\ in the market and will
continue to guard against ``outlier'' markets undermining the
objectives of Rule 611.
---------------------------------------------------------------------------
\480\ See infra section VII.D.2.
\481\ See infra section VII.D.2.d.
---------------------------------------------------------------------------
While the Commission agrees with commenters that tick size and
access fees are relational in so far as the access fee cannot be more
than half of the minimum pricing increment (for the reasons discussed
above), maintaining the current proportionality of the access fee to
tick could perpetuate distortions in the market. As stated by one
commenter, ``when the cap was set in 2005, neither the Commission nor
commenters ever suggested that the cap should be exactly equal to a
fixed proportion of the tick size.'' \482\
---------------------------------------------------------------------------
\482\ IEX Letter V at 7.
---------------------------------------------------------------------------
Further, lowering the access fee cap for only those NMS stocks that
are assigned a lower minimum pricing increment (i.e., those NMS stocks
that are constrained by the $0.01 minimum pricing increment) and
maintaining the preexisting (30 mils) fee cap for all other NMS stocks
priced $1.00 or greater could increase the probability that some stocks
will oscillate from one tick size to another rather than settling on an
appropriate tick. This oscillation creates additional cost for market
participants, introduces complexity in the markets and creates
operational risk.\483\ In addition, continuing to apply a 30 mil access
fee cap to those NMS stocks that continue to be assigned a $0.01
pricing increment would ignore the efficiencies in trading \484\ that
have been realized in the intervening 19 years since the caps were
adopted and would not address the distortive effects of access fee
structures that assess access fees at or near the current cap in order
to maximize the amount of the rebate that can be offered.\485\
---------------------------------------------------------------------------
\483\ See infra section VII.F.2.a.
\484\ See supra note 405 and accompanying text (discussing
trading efficiencies due to technology changes and reduced costs),
infra note 598, and infra section IV.D.1.d.
\485\ See infra section IV.D.1.d.
---------------------------------------------------------------------------
In addition, as discussed below, and supported by some commenters,
fees and rebates which are currently benchmarked against the 30 mil cap
have a negative impact on price transparency and routing
practices.\486\ According to one commenter, ``there is evidence that
exchanges that pay the highest rebates often provide worse execution
quality.'' \487\ Another commenter provided data it said demonstrates
that ``poor execution quality is directly linked to high access fees.''
\488\
---------------------------------------------------------------------------
\486\ See, e.g., IEX Letter IV at 2-4; Better Markets Letter I
at 16.
\487\ Better Markets Letter I at 16. See also IEX Letter VI at
7; NASAA at 9.
\488\ IEX Letter VI at 7.
---------------------------------------------------------------------------
Lowering the cap to 10 mils for all NMS stocks, both those that are
assigned a $0.005 tick and those that retain the $0.01 tick, will
benefit market participants, including investors, by lessening the
incentives to route to a market in order to receive a rebate.
The Commission proposed a two-level access fee cap structure for
access to protected quotes in NMS stocks priced at $1.00 or more to
accommodate the proposed variable minimum pricing increment structure
and specifically to prevent the access fee caps from creating pricing
distortions with the smallest proposed minimum pricing increment (i.e.,
5 mils access fee cap for NMS stocks that would have been assigned a
$0.001 pricing increment and 10 mils cap for NMS stocks that would have
been assigned a pricing increment greater than $0.001).\489\
Specifically, the proposed 5 mil access fee cap was necessary to
accommodate the proposed lowest $0.001 minimum pricing increment
because imposing the proposed 10 mil access fee cap on the $0.001
minimum pricing increment would have created distortions in quoting and
negatively impact pricing transparency.\490\
---------------------------------------------------------------------------
\489\ See Proposing Release, supra note 11, at 80348 (stating
``the access fee cap should not be greater than \1/2\ of the tick
size in order to preserve coherence between net and nominal price
rankings of trading venues. This would not be possible with an
access fee cap of $0.001 and a lowest possible proposed tick size of
the same amount, as would be the case for the smallest tick size
tier from the proposal.'').
\490\ See Proposing Release, supra note 11, at 80267 & 80289-90
(stating ``[a] reduction in the minimum pricing increment without
reducing the access fee caps could permit fees to become a higher
percentage of the minimum pricing increment, which could potentially
undermine price transparency and exacerbate the other concerns with
maker-taker fees.'').
---------------------------------------------------------------------------
Several commenters raised concerns about the proposed 5 mil access
fee cap and its ratio as compared to the minimum pricing
increment.\491\ One commenter stated that ``[a]t 50% of the minimum
pricing increment, a round trip buy and sell trade could result in
access fees equal to the spread'' and expressing concern that ``at 50%
of the spread, rebates of greater than half of the minimum pricing
increment could lead to market distortion.'' \492\ Because the
amendment to Rule 612 does not include the $0.001 minimum pricing
increment, the Commission is not adopting the 5 mils access fee cap.
Specifically, with the elimination of the proposed $0.001 minimum
pricing increment,\493\ the proposed 5 mil access fee cap is
unnecessary.\494\ Accordingly, the Commission is adopting the proposed
10 mils access fee cap as proposed.\495\ The Commission is removing the
proposed tiered approach to the access fee caps and instead maintaining
the preexisting single, uniform access fee cap structure for protected
quotes priced $1.00 or more.\496\ The amendment will introduce fewer
variables, less complexity and lower cost and operational risk as
compared to the proposed two-level access fee cap structure.\497\ More
specifically, as is the case today, there will be one access fee cap
for NMS stocks priced at $1.00 or more and a separate access fee cap
that applies to NMS stocks priced below $1.00.
---------------------------------------------------------------------------
\491\ See, e.g., Schwab Letter II at 6 and 36; SIFMA Letter II
at 45; Citadel Letter I at 23; Hudson River Letter at 4; AIMA Letter
at 4; Robinhood Letter at 57-58.
\492\ Hudson River Letter at 4.
\493\ See supra section III.C and supra note 346.
\494\ This approach is consistent with some commenters'
recommendations. See, e.g., Better Markets Letter I at 16 (stating
``the Commission should just dispense with the $0.001 tick size
altogether'' because doing so would eliminate ``the need for a
separate [5 mil] access fee cap.'' This commenter stated that
proceeding in this manner would maintain a single uniform cap for
all stocks and avoid introducing additional complexity).
\495\ See supra note 346.
\496\ Several commenters expressed support for expanding the
application of the access fee caps in certain ways. See, e.g., infra
notes 614-616. Expanding or altering the structure of the cap would
add complexity to the national market system. As discussed above, in
response to commenters, amended Rule 610(c) introduces fewer
variables and less complexity into the national market system.
Expanding the application of Rule 610(c) and/or modifying its
structure as these commenters suggest would be inconsistent with
this approach.
\497\ See Regulation NMS Adopting Release, supra note 4, at
37595.
---------------------------------------------------------------------------
b. Impact on Liquidity and the NBBO
Some commenters stated that lowering the access fee cap to 10 mils
would not impinge on the exchanges' ability to offer incentives and
attract liquidity and instead stated that reducing the caps would
likely draw liquidity back to the exchanges.\498\ According to one
commenter, ``ATSs and other off-exchange venues generally charge rates
much lower than the access
[[Page 81654]]
fees imposed by most exchanges. Because their cost of access is so much
higher than on other venues, exchanges become the venue of `last
resort.' '' \499\ This commenter further stated that ``modernizing the
access fee cap and bringing exchange access fees in line with off-
exchange trading venues will reduce the need for exchange avoidance and
naturally result in a better experience for liquidity providers, one
that will not need to be `offset' by rebate payments.'' \500\ Another
commenter stated that ``[b]rokers' avoidance of these fees is a
significant contributor for brokers often choosing to internalize or
first route to ATSs or OTC market makers, rather than to exchanges with
their customers' orders.'' \501\ Another commenter stated that ``the
30-mil cap is among the factors driving the shift away from displayed
trading.'' \502\ One commenter stated ``reduced displayed trading is a
problem because it impedes the fair and transparent distribution of
pricing and transaction information that Congress directed the
Commission to protect . . . [and] the 30-mil cap is among the factors
driving the shift away from displayed trading.'' \503\ And another
commenter stated that lower access fees will impose lower costs on
investors, ``removing a disincentive for trading on exchanges.'' \504\
---------------------------------------------------------------------------
\498\ See IEX Letter IV at 18-19, 22; Themis Letter at 8; Better
Markets Letter I at 16; BMO Letter at 3-4; ASA Letter at 5.
\499\ IEX Letter IV at 18.
\500\ IEX Letter IV at 18-19.
\501\ Healthy Markets Letter I at 21.
\502\ IEX Letter IV at 6. See also ASA Letter at 5.
\503\ IEX Letter IV at 6.
\504\ Better Markets Letter I at 16. See also IEX Letter VI at
1.
---------------------------------------------------------------------------
Other commenters stated that lowering the access fee caps would
address other concerns regarding market distortions associated with the
payment of high rebates.\505\ One commenter stated that ``[r]ebates
distort supply and demand and harm the price discovery process.'' \506\
One commenter stated that reducing the cap to 10 mils would ``provide
ample room for exchanges to create incentives, charge premium or
discounted prices, and earn a profit, all while lowering the distortive
effects they have on the equity market.'' \507\ In addition, some
commenters stated that reducing the cap would alleviate the potentially
distortive effects of the maker-taker pricing model.\508\ According to
one commenter, ``the current fee levels foster and enable significant
market distortions in today's marketplace'' and ``the fees charged by
exchanges often serve as powerful disincentives for market participants
to access that liquidity.'' \509\ Another commenter ``recognize[d] that
access fees and the rebates that they fund serve an important function
in incentivizing liquidity provision for thinly-traded securities and
compensating market makers for adverse selection,'' but also stated
that ``[p]rudent regulation must appropriately . . . balance the
beneficial effect of access fees on liquidity against the potential for
market distortions'' associated with maintaining the 30 mil cap.\510\
According to this commenter, ``lowering fees would mitigate the
detrimental effect of access fees on order routing, price transparency,
and market quality in many securities.'' \511\ Further, another
commenter stated that when the Commission adopted the preexisting caps,
``its focus was on limiting the distortive impact of disproportionate
access fees, not on facilitating the ability of markets to pass them
through as rebates.'' According to this commenter, the only way in
which the Commission viewed access fees and rebates as related was that
``a fee limit was needed to avoid distortive pricing of the type that
occurs when access fees are primarily passed through to other
participants in the form of rebates.'' \512\ This commenter further
stated that the ``bulk of executions against displayed quotes pay the
maximum fee, with the overwhelming share of that revenue being passed
through as rebates.'' \513\
---------------------------------------------------------------------------
\505\ See IEX Letter IV at 10-11, 16; Themis Letter at 7.
\506\ Themis Letter at 7. See also IEX Letter IV at 16; Verret
Letter III at 11, 13-14.
\507\ BMO Letter at 4. See also Tripari Letter; Verret Letter
III at 13; Proposing Release, supra note 11, at 304.
\508\ See, e.g., BMO Letter at 4; Brandes Letter at 3; Healthy
Markets Letter I at 21; Themis Letter at 7; Council of Institutional
Investors Letter at 3. See also infra section VII.D.2 (discussing
market distortions).
\509\ Healthy Markets Letter I at 21.
\510\ BlackRock Letter at 10-11.
\511\ Id. at 10.
\512\ IEX Letter IV at 5.
\513\ Id.
---------------------------------------------------------------------------
Finally, one commenter stated ``the pricing distortions the
Commission was concerned about when it adopted Regulation NMS have
become acute today due to changed market conditions'' resulting in
brokers being incentivized to ``route orders away from best-displayed
exchange quotes in order to avoid the high fees--precisely the result
the Commission sought to avoid when it first adopted the cap.'' \514\
This commenter also stated that ``the introduction of `inverted' venues
that pay rebates to access rather than provide displayed orders, and
the use of highly-skewed rebate tiers, has created even more price
distortion and misaligned incentives.'' \515\
---------------------------------------------------------------------------
\514\ IEX Letter IV at 5.
\515\ IEX Letter IV at 5.
---------------------------------------------------------------------------
Other commenters opposed any changes to the existing access fee
caps because they stated that reducing the caps would limit the
exchanges' ability to offer rebates to incentivize liquidity providers,
as access fees typically fund such rebates and this could negatively
impact liquidity on exchange markets.\516\ Some commenters stated that
the reduction in the access fee caps, which would reduce rebates, would
result in wider spreads, and less quoted size which would increase
trading costs.\517\ One commenter stated that the ``cost of widening
spreads that would result from removing fees and rebates would cost
retail investors . . . as much as $687 million per year.'' \518\ Some
commenters stated that the reduction of rebates would impact spreads,
which (according to those commenters) suggests that rebates have an
impact on displayed pricing.\519\ Other commenters stated that the
reduction in rebates would have a negative impact on market
liquidity.\520\ One commenter stated that the current access fee cap
levels ``help[ ] improve liquidity and provide narrower quotes than
otherwise would be available in the marketplace.'' \521\ Similarly,
another commenter stated support for further ``examining changes to the
access fee cap,'' but cautioned ``that wholesale reductions,
particularly when combined with other changes . . . will disincentivize
liquidity provision, reduce market maker support, widen bid-ask
spreads, and increase volatility in thinly-traded securities.'' \522\
Some commenters stated that certain securities may ``require rebates
larger than 10 mils to incentivize tight quotes.'' \523\ However,
another commenter stated that claims of ``hidden costs to investors, in
the form of worse NBBO prices, wider spreads, higher costs for retail
investors, in the
[[Page 81655]]
form of worse NBBO prices, wider spreads, higher costs for retail
investors and less liquidity for thinly-traded securities'' did not
have a factual basis and did not account for the ``tangible cost
reductions that would arise from lower access fees.'' \524\
---------------------------------------------------------------------------
\516\ See, e.g., World Federation of Exchanges Letter at 4;
Virtu Letter II at 8, 16-17; Citadel Letter I at 22.
\517\ See, e.g., State Street Letter at 4, Interactive Brokers
Letter at 5, Nasdaq Letter I at 23, Nasdaq Letter II at 5-6, Letter
from Brett Kitt, Associate Vice President, Principal Associate
General Counsel, Nasdaq, Inc., dated Feb. 14, 2024 (``Nasdaq Letter
III'') at 5, Cboe Letter III at 8.
\518\ Nasdaq Letter III at 5.
\519\ See, e.g., Cboe Letter II at 8-9; Cboe Letter III at 8;
Nasdaq Letter I at 2; Nasdaq Letter I at 22.
\520\ See, e.g., CCMR Letter at 27, Interactive Brokers Letter
at 5.
\521\ Fidelity Letter at 14.
\522\ State Street Letter at 4. See also CCMR Letter at 27;
Interactive Brokers Letter at 5; Nasdaq Letter I at 23; Nasdaq
Letter II at 5-6; Nasdaq Letter III at 5.
\523\ Fidelity Letter at 14. See also e.g., Nasdaq Letter II at
6; State Street Letter at 4.
\524\ See IEX Letter IV at 22.
---------------------------------------------------------------------------
Certain exchanges also opposed any changes to the access fee caps,
stating that reducing the access fee caps would impede their ability to
offer competitive rebates and meaningful price differentiation,
hindering their ability to attract liquidity and compete with off-
exchange trading venues for order flow.\525\ One commenter stated that
``[c]ompressing the caps further . . . [would] introduce additional
concerns with implications for competition and market quality.'' \526\
In addition, this commenter stated that rebates, which are funded by
access fees, are ``innovative and critically important tools that
enhance market depth, promote tighter bid-ask spreads, and encourage
order flow to be routed to lit exchanges'' and any diminution in the
access fee caps would ``in fact disrupt current business practices and
competitive dynamics.'' \527\ This commenter also stated that reducing
the access fee caps could ``have significant revenue consequences,''
\528\ but provided no specifics. Another commenter stated that reducing
rebates ``discourages on-exchange market making,'' which could
deteriorate the NBBO as it ``would be drawn from a smaller and less
representative pool of displayed liquidity.'' \529\ According to this
commenter, although the proposal might make it cheaper for broker-
dealers to access liquidity, costs for liquidity providers and market
makers would increase, and spreads would widen, which in turn would
result in higher ``all-in'' costs for investors.\530\ Further, one
commenter stated that the proposal ``risks weakening the NBBO by
restricting exchanges' ability to offer meaningful rebates to encourage
more liquidity and tighter spreads that underpin the NBBO'' \531\ and
stated that the NBBO is ``comprised exclusively of trading interest
displayed on public exchanges. . . [and] limit[ing] exchanges' ability
to gather liquidity . . . would weaken the public reference price.''
\532\ According to this commenter, ``rebates are essential to market
quality as they encourage market participants to act as market makers
and provide two-sided quotes that make the equity markets function
soundly.'' \533\ This commenter further stated that rebates provide
``integral value to the operation of well-functioning, fair, and
orderly equity markets'' because they serve to cushion market makers
against the risks of adverse selection and price volatility, thereby
incenting them to continue to make markets, even in thinly-traded or
volatile securities, and to do so with tighter spreads than they would
otherwise.'' \534\ According to the same commenter, a reduction in
rebates would lead to greater market complexities because there would
be more ``speedbump or `quote protection' markets'' and a ``[g]reater
focus on segmentation.'' \535\ This commenter also stated that the
proposal would ``potentially undermine the competitive positions of the
exchanges and the market makers that quote on them by seeking to limit
their ability to charge fees and collect rebates for their respective
services.'' \536\ Another commenter stated that ``it is entirely
inappropriate to experiment with exchange pricing models for fear of
broker failings'' and ``exchange fees are extremely transparent . . .
and receive a significant amount of SEC review.'' \537\
---------------------------------------------------------------------------
\525\ See Cboe Letter II at 8-9; Cboe Letter III at 8; Nasdaq
Letter I at 2; Nasdaq Letter I at 22. See also e.g., Virtu Letter II
at 8; Citadel Letter I at 22.
\526\ See Cboe Letter II at 8. See also Cboe Letter IV at 2;
Nasdaq Letter I at 22-29; Nasdaq Letter IV at 8.
\527\ Cboe Letter III at 4. See also Cboe Letter IV at 3; Nasdaq
Letter I at 20.
\528\ Cboe Letter IV at 3.
\529\ Nasdaq Letter II at 7. See also Nasdaq Letter I at 19;
Nasdaq Letter III at 5-6. But see IEX Letter IV at 10-11 (stating
``[t]here is ample evidence that maintaining the access fee cap at
its current level has led to distortions the Commission sought to
avoid.'').
\530\ Nasdaq Letter I at 20; Nasdaq Letter II at 6-7.
\531\ Nasdaq Letter I at 2, 22.
\532\ Nasdaq Letter I at 22.
\533\ Nasdaq Letter I at 22
\534\ Nasdaq Letter I at 22.
\535\ Nasdaq Letter IV at 8.
\536\ Nasdaq Letter I at 21.
\537\ Cboe Letter III at 5-6.
---------------------------------------------------------------------------
Finally, one commenter ``question[ed] the Commission's authority to
reduce the fee cap beyond what is needed to accommodate the new,
smaller tick sizes, thereby with the implicit aim of limiting the
ability of exchanges to provide meaningful rebates to market
participants.'' \538\ According to this commenter, the Commission
``lacks the authority to enact radical changes to exchange access fees
without explicit congressional mandate'' and ``the Commission's charge
to establish a national market system evidences no express intent for
the Commission to impose price controls upon exchanges as a means of
promoting competition.'' \539\
---------------------------------------------------------------------------
\538\ Nasdaq Letter VI at 1.
\539\ Nasdaq Letter VI at 2.
---------------------------------------------------------------------------
Other commenters stated that the Commission had clear statutory
authority to adopt Rule 610 and to make subsequent adjustments to the
access fee caps.\540\ One commenter stated that ``a plethora of items
in the 34 Act, the SEC's prior 50 years of regulation of the National
Market System, and related constitutional precedent regarding the
private non-delegation doctrine specific to self-regulatory
organizations (SROs)--not only give the SEC sufficient delegation of
authority to adopt the pending NMS proposal, they compel the SEC to
exercise its authority that oversees a dynamically changing national
market system.'' \541\
---------------------------------------------------------------------------
\540\ See, e.g., IEX Letter IV at 3; Letter from J.W. Verret,
Associate Professor, George Mason University, dated Aug. 27, 2024
(``Verret Letter IV'').
\541\ Verret Letter IV at 2.
---------------------------------------------------------------------------
As discussed above,\542\ section 11A(c)(1)(B) of the Exchange Act
authorizes the Commission to adopt rules assuring the fairness and
usefulness of quotation information.\543\ Further, Congress explicitly
granted the Commission ``broad authority to oversee the implementation,
operation, and regulation of the national market system'' and the
``clear responsibility to assure that the system develops and operates
in accordance with Congressionally determined goals and objectives''
which requires balancing different, and often competing, interests and
components of the complex national market system.\544\ The access fee
caps in preexisting Rule 610(c) were adopted through rulemaking
pursuant to the Exchange Act, including section 11A, and recalibration
of the levels of the access fee caps falls squarely within the
Commission's statutory authority. The commenter that questioned the
Commission's authority to reduce the level of the access caps to 10
mils acknowledged the Commission's authority to reduce the access fee
caps ``to accommodate the new, smaller tick sizes'' \545\ and in an
earlier comment letter stated it ``supports adjusting the access fee
cap to accommodate new tick sizes.'' \546\ The Commission's authority
set forth in the Exchange Act is not circumscribed in the manner
suggested
[[Page 81656]]
by this commenter. Congress granted the Commission broad authority to
oversee the national market system and ensure that it is meeting the
investment needs of the public. The reductions in the access fee caps
adopted herein are designed to improve market quality for market
participants accessing protected quotations in all NMS Stocks, not just
those that will be assigned a new minimum pricing increment. As
discussed throughout, the adjusted level of the caps also will allow
trading centers to retain their net capture for transactions of
protected quotations priced $1.00 or more and therefore trading centers
who wish to use rebates to attract liquidity may continue to do so.
Accordingly, the Commission has considered and balanced policy
objectives in this complex area and reached an appropriate policy
decision.\547\
---------------------------------------------------------------------------
\542\ See supra section IV.A. See also sections I and I.B.
\543\ 15 U.S.C. 78k-1(c)(1)(B). Section 23 of the Exchange Act
also authorizes the Commission ``to make such rules and regulations
as may be necessary or appropriate to implement the provisions'' of
the Act. 15 U.S.C. 78w(a)(1).
\544\ See supra notes 2-3 and accompanying text.
\545\ Nasdaq Letter VI at 1 and Nasdaq Letter I at 2.
\546\ Nasdaq Letter I at 2.
\547\ See Regulation NMS Adopting Release, supra note 4, at
37498. As was the case when the Commission adopted the preexisting
fee caps, the rulemaking process has required the Commission to
``grapple with many difficult and contentious issues that have
lingered unresolved for many years'' and after examining these
issues and assessing the views of commenters, particularly those
that disagree with the proposal, ``decisions must be made and
contentious issues must be resolved so that the markets can move
forward with certainty.'' Id. While the Commission always seeks to
achieve a consensus, ``consensus can mean indefinite gridlock that
ultimately could damage the competitiveness of the U.S. equity
markets [ ]. [T]he time has come to make the difficult decisions
necessary to modernize and strengthen the national market system.''
Id.
---------------------------------------------------------------------------
Although some commenters stated that rebates are essential to
attract liquidity on exchanges,\548\ the access fee caps were not
established to support trading centers' ability to offer rebates or to
ensure a particular level of rebate payment; they were developed as a
means to help ensure fair, efficient, and ready access to protected
quotes, to preserve the integrity of displayed prices and to ensure
that the objectives of Rule 611 would not be undermined by trading
centers who might seek to charge exorbitant fees to those now required
to access their protected quotations.\549\ The Commission disagrees
that rebates are essential to attract liquidity on national securities
exchanges or the only means of attracting liquidity.\550\ The
Commission stated in 2005 that markets have ``significant incentives to
be near the top in order-routing priority'' \551\ and displaying the
best protected quotation will attract liquidity to a market. The
adopted amendments will continue to allow for trading centers to
develop different fee models while also preserving the objectives of
Rule 610(c). Further, market participants that post non-marketable
orders are able to price their orders to accommodate the risk of
adverse selection and rebates are not necessary for compensating this
risk.
---------------------------------------------------------------------------
\548\ See, e.g., Cboe Letter III at 6-7; World Federation of
Exchanges Letter at 4; Virtu Letter II at 8, 16-17; Citadel Letter I
at 22.
\549\ See Regulation NMS Adopting Release, supra note 4, at
37503.
\550\ See id. at 37596.
\551\ Id.
---------------------------------------------------------------------------
Reducing the access fee caps will help to alleviate the distortive
effects of the preexisting level of access fees and the rebates they
fund. The access fee caps were designed to protect limit orders and to
assure that orders could be routed to those markets that were
displaying the best-priced quotations.\552\ As the Commission stated
when it adopted the access fee caps, ``[a]ccess fees tend to be highest
when markets use them to fund substantial rebates to liquidity
providers, rather than merely to compensate for execution services. If
outlier markets are allowed to charge high fees and pass most of them
through as rebates, the published quotations of such markets would not
reliably indicate the true price that is actually available to
investors or that would be realized by liquidity providers.'' \553\ The
Commission also discussed the potential distortionary effect of high
fees and rebates on displayed quotes and sought to assure that
displayed prices were within a limited range of net prices.\554\
---------------------------------------------------------------------------
\552\ See Regulation NMS Adopting Release, supra note 4, at
37545.
\553\ See Regulation NMS Adopting Release, supra note 4, at
37545. See also IEX Letter V at 3 (stating ``there is an obvious and
direct connection between high access fees and the extent to which
displayed prices deviate from the true prices at which participants
are prepared to trade.'').
\554\ See Regulation NMS Adopting Release, supra note 4, at
37502.
---------------------------------------------------------------------------
In the current national market system, fees for access to protected
quotes are typically charged at the highest amount allowed under Rule
610(c) and the vast majority of the fees collected are paid out as
rebates.\555\ This practice results in displayed quotations prices that
are not reflective of underlying economics of liquidity supply and
demand, but rather displayed quotations prices that have been
calculated to account for the receipt of a rebate.\556\ The Commission
is concerned that this structure impairs the fairness and accuracy of
displayed quotations.
---------------------------------------------------------------------------
\555\ See infra section VII.C.2 and section VII.B.3 (stating
most exchanges charge the maximum fee (in the range of 30 mils) and
provide the maximum rebate (in the vicinity of 30 mils) and stating
that the primary reason that access fees remain near 30 mils on most
exchanges is to fund rebates) and infra section VII.C.2.c, table 4.
\556\ See infra section VII.B.3.
---------------------------------------------------------------------------
The access fee caps set an outer limit on the cost of accessing
protected quotations to ``assure[ ] order routers that displayed prices
[are] within a limited range, true prices.'' \557\ In setting the
maximum level for the access fees trading centers could charge market
participants to access a protected quotation in 2005, the Commission
specifically recognized that ``some markets might choose to charge
lower fees, thereby increasing their ranking in the preferences of
order routers. . . while [o]thers might charge the full $0.003 and
rebate a substantial proportion to liquidity providers.'' \558\ The
Commission left it to the markets and competition to determine what
strategies would be successful in attracting order flow, subject to the
maximum access fee cap.\559\ Without an access fee cap, the Commission
was concerned that certain markets would charge high fees and pass most
of them through as rebates, which would undermine price discovery and
price transparency.\560\ The Commission's concerns when it adopted
Regulation NMS, that access fees might gravitate to the highest level
permitted by Rule 610 and the impact on price transparency, have been
realized to the detriment of investors.\561\
---------------------------------------------------------------------------
\557\ Regulation NMS Adopting Release, supra note 4, at 37502.
See also IEX Letter IV at 4 and 15 (stating ``[t]he purpose [for
capping access fees] is not, and has never been, to allow exchanges
to maintain rebate payments at current high levels.'').
\558\ Regulation NMS Adopting Release, supra note 4, at 37545
(stating that establishing the $0.003 cap to ``limit the outlier
business model [and] plac[ing] all markets on a level playing field
in terms of the fees they can charge and the rebates they can pass
on to liquidity providers. Some markets might choose to charge lower
fees, thereby increasing their ranking in the preferences of order
routers. Others might charge the full $ 0.003 and rebate a
substantial proportion to liquidity providers. Competition will
determine which strategy is most successful.''). See also Proposing
Release, supra note 11 at 80348 (stating ``The Commission recognizes
that an access fee cap of 10 mils for stocks . . . would provide
exchanges with enough pricing freedom to continue to offer
economically meaningful rebate-tiering.''). One commenter stated
that ``diminished reliance on the maker-taker economics would
encourage a variety of alternative market models for providing
liquidity,'' which in this commenter's view would be consistent with
the outcome the Commission anticipated in 2005, but which has not
been realized. Decimus 2016 Letter at 11.
\559\ Regulation NMS Adopting Release, supra note 4, at 37545.
\560\ Id. See also Better Markets Letter II at 3 (lowering
access fees would ``ensure that the fees charged to access a
protected quotation do not distort the true price that is available
to investors.'').
\561\ See Proposing Release, supra note 11, at 80292 n.317 and
accompanying text and 80290 n.302. See also infra section VII.A.
---------------------------------------------------------------------------
As discussed below, the reduction in the access fees will improve
market quality.\562\ For those NMS stocks that
[[Page 81657]]
are not experiencing a constraint on the quoted spread due to the
minimum pricing increment, transaction costs will remain largely
unchanged under the amendments as spreads will adjust on average to
offset the reduction in access fees and rebates.\563\ For those NMS
stocks that do experience constraint on the quoted spread due to the
preexisting minimum pricing increment, transaction costs for liquidity
seekers will go down and the oversupply of liquidity will be reduced,
which will allow for shorter queues and higher fill rates.\564\ For
these NMS stocks, the access fee functions as a tax on liquidity
demand.\565\ Reducing the access fee in these constrained stocks will
result in savings for investors.\566\
---------------------------------------------------------------------------
\562\ See infra section VII.B.
\563\ See infra section VII.D.
\564\ See infra section VII.D.2.c.
\565\ See infra section VII.D.2.c.
\566\ See infra section VII.D.2.
---------------------------------------------------------------------------
Further, while some commenters stated that exchange volume would
decline and that OTC trading would increase if the access fee caps were
reduced, as discussed below,\567\ analysis indicates that liquidity
providers would not be deterred from quoting on exchange because they
will be able to widen their quote to reflect the reduced rebate,
thereby receiving the same economic profit as they received with the
rebate. Liquidity demanders would not be worse off because the
reduction in access fee would offset, or, in the case of stocks with an
economic spread of less than a tick, more than offset, the increase in
spread.\568\
---------------------------------------------------------------------------
\567\ See infra note 1469, and accompanying text.
\568\ See infra section VII.D.2.c.
---------------------------------------------------------------------------
Some commenters stated that the reduced access fee caps would help
to address potential conflicts of interest in routing decisions that
may harm execution quality of customer orders.\569\ One commenter
stated that ``lowering the access fee cap would lead to a reduction in
broker conflicts of interests.'' \570\ Another commenter stated that
the proposal ``will help to reduce the extent of the conflict of
interest in agency routing decisions.'' \571\ Another commenter stated
``[w]e have long supported the Commission addressing the conflict faced
by brokers related to incentives created by access fees and rebates in
the maker/taker model'' and that ``a simple reduction of access fees
across all venues to $0.001 would go a long way in mitigating order
routing conflicts.\572\
---------------------------------------------------------------------------
\569\ See, e.g., Proof Letter at 1; RBC Letter at 4; Ontario
Teachers et al. Letter at 2; NASAA Letter at 9; Vanguard Letter at
6; BlackRock Letter at 10; Capital Group Letter at 4; Themis Letter
at 7.
\570\ RBC Letter at 4.
\571\ Proof Letter at 1.
\572\ Capital Group Letter at 4.
---------------------------------------------------------------------------
Another commenter, however, stated that the Commission did not
provide any new data to support its position that access fees and
rebates are ``actually harmful to the market'' and further stated that
``the Commission's supposition that rebates present harmful conflicts-
of-interest to brokers is not supported with evidence, and it ignores
the countervailing benefits associated with rebates, which are
essential tools for gathering the displayed quotes that form the
NBBO.'' \573\ Another commenter stated that ``half of the rebates on
Cboe accrue to non-agency market-making activity--thus, there is no
real or perceived conflicts of interest'' and for agency order flow,
some of that flow is ```directed' meaning clients give specific
instructions for the order to be routed to a particular venue for
execution'' and thus there similarly is no conflict.\574\ This
commenter further stated, ``brokers have a duty of best execution
regardless of the pricing model used by the exchange.'' \575\
---------------------------------------------------------------------------
\573\ Nasdaq Letter I at 2 (stating ``It would be arbitrary and
capricious for the Commission to proceed with the Proposal in the
absence of evidence that the current fee cap is actually harmful to
the market and without meaningfully weighing the costs and benefits
of those reductions.''); Nasdaq Letter III at 6 (stating the
Commission ``did not cite any new research conducted subsequent to
the Transaction Fee Pilot to support the SEC's change of position
that access fees and rebates are actually harmful.'').
\574\ Cboe Letter III at 5-6.
\575\ Cboe Letter III at 5-6.
---------------------------------------------------------------------------
The Commission disagrees with those commenters that questioned the
existence of potential conflicts of interest. The Commission has
received comments from market participants that have stated that
potential conflicts of interest are a concern because of the fee/rebate
models. Some commenters stated that fees and rebates that are currently
benchmarked against the 30 mil cap have a negative impact on routing
practices \576\ and one commenter offered evidence that ``exchanges
that pay the highest rebates often provide worse execution quality.''
\577\ Another commenter provided data it said demonstrates that ``poor
execution quality is directly linked to high access fees.'' \578\
Moreover, the Commission has heard similar concerns about potential
conflicts of interest created by the fee and rebate schedules and their
impact on market quality for many years.\579\ The 10 mils access fee
cap is appropriate because it will mitigate the potential conflicts of
interest associated with the current fee and rebate models, while still
allowing exchanges to use rebates to attract liquidity.
---------------------------------------------------------------------------
\576\ See, e.g., IEX Letter IV at 2; Better Markets Letter I at
16.
\577\ Better Markets Letter I at 16.
\578\ IEX Letter VI at 7.
\579\ For example, in 2018, one market participant stated that
the preexisting level of the access fee cap may ``create misaligned
incentives and potential conflicts of interest for broker dealers'
routing and execution decisions . . . because broker dealers may
elect to post non-marketable limit orders on market venues offering
the highest rebate and bypass those venues where there is greater
likelihood of execution, but a higher fee.'' Goldman 2018 Letter at
3-4. This market participant went on to state that ``[b]y
maintaining the Fee Cap at the level adopted in 2005 as spreads have
narrowed and commissions have decreased over the past 13 years,
these misaligned incentives and potential conflicts of interest have
grown.'' Id. at 4. According to this market participant, adjusting
the access fee cap to 10 mils ``would reduce the effect of these
misaligned incentives and the potential conflicts of interest.'' Id.
Further, a decade ago, one commenter stated ``[a] reduction in the
cap [to 10 mils] . . . would naturally move more executions back to
exchanges.'' Citigroup 2014 Letter at 7. Another commenter stated in
2015 that a reduction in the access fee cap to 5 mils (half of the
amended level adopted) would ``still allow room for exchanges to
provide rebates to market participants in order to incentivize
liquidity, while at the same time significantly reducing the market
distortions and unnecessary complexity that access fees have
caused.'' Letter from Theodore R. Lazo, Managing Director &
Associate General Counsel, SIFMA, to Mary Jo White, Chair,
Commission, dated May 24, 2015, at 2-3 (stating its support for
BATS' 2015 Petition for Rulemaking to, among other things, reduce
the baseline access fee cap to 5 mils).
---------------------------------------------------------------------------
c. Agency Market Business Model
Several commenters stated that the Commission should not consider
whether the proposed lowered access fee caps would unduly impair
current agency market business models \580\ as a factor in its
analysis.\581\ According to one commenter, ``setting access fee caps,
or designing any aspect of market structure, specifically to preserve
or protect existing exchange fee models is an inappropriate policy
rationale.'' \582\ Similarly, another commenter stated that it ``is not
the Commission's role to ensure that trading centers `maintain their
current net capture rate.' '' \583\ Further, one commenter stated that
while, in its opinion, it would be appropriate for the Commission to
``assess the impact of the proposed access fee cap on market
participants' varying business models,'' it must ``then account for the
impact of the proposed access fee cap on all market participants and
attempt to create the most
[[Page 81658]]
competitive and effective environment on an overall basis, rather than
doing so exclusively for exchanges.'' \584\
---------------------------------------------------------------------------
\580\ See Proposing Release, supra note 11, at 80270 n.35
(stating ``[a]gency market trading centers are those that bring
together buyers and sellers and typically charge a fee for their
execution services.''). See also Regulation NMS Adopting Release,
supra note 4, at 37545.
\581\ See, e.g., Citadel Letter I at 22-23; Schwab Letter II at
36; SIFMA Letter II at 39-40.
\582\ SIFMA Letter II at 39-40.
\583\ Schwab Letter II at 36. See also Virtu Letter II at 17-18.
\584\ Virtu Letter II at 18.
---------------------------------------------------------------------------
In considering whether to adjust the level of the access fee caps,
and if so, by what amount, the Commission has considered the impact of
such modifications on market participants to help ensure that all
investors will continue to have fair and non-discriminatory access to
protected quotations and the Commission has not prioritized exchange
revenues over other considerations. When the Commission adopted the
access fee caps in Regulation NMS, it considered the impact of the caps
on the agency market business model, as it has done in this release as
well.\585\ Agency market trading centers have historically charged
transaction fees for their agency services in bringing together buyers
and sellers to execute transactions. The Commission is not prohibiting
agency market trading centers from continuing to assess fees for
providing execution services to access protected quotations. As
discussed above, it was appropriate and consistent with its
responsibilities under the Exchange Act for the Commission to consider
the impact of the original access fee caps on the ongoing viability of
different trading centers. Because of the important role agency market
trading centers continue to play in the national market system, it is
similarly consistent with the Exchange Act for the Commission to
undertake a similar analysis today in adjusting the level of the caps.
---------------------------------------------------------------------------
\585\ See Regulation NMS Adopting Release, supra note 4, at
37545 (``stating ``the adopted [30 mils] fee limitation will not
impair the agency market business model.'').
---------------------------------------------------------------------------
The Proposing Release estimated the effect on exchange net capture
because exchanges are the only trading centers that impose fees for
access to protected quotations at this time and, therefore, are subject
to the access fee caps.\586\ Further, the Commission's analysis in the
Proposing Release appropriately considered the impact of proposed
changes to Rule 610(c) on entities employing an agency market business
model because Rule 610(c) applies to those entities and the Commission
was cognizant of not compressing the access fee caps so far as to
effectively eliminate such business models.\587\ As discussed above,
the Commission stated that if markets are allowed to charge high access
fees and pass most of them through as rebates, the published quotations
of such markets will not reliably indicate the true price that is
available and investors may be overcharged for taking liquidity.\588\
As discussed below, the Commission estimated the current net capture of
the exchanges at approximately 2 to 6 mils and anticipates that will
remain the same under amended Rule 610(c).\589\
---------------------------------------------------------------------------
\586\ The Commission used these same estimates to determine the
changes in the amount that liquidity demanders would pay and the
amounts that liquidity providers would receive. See Proposing
Release, supra note 11, at section V.D.3 (discussing impact of the
proposed lower access fee caps on exchanges' net capture).
\587\ See Proposing Release, supra note 11, at 80290-91
(proposing new caps designed to ``allow current business practices
to continue while adjusting access fee levels to align with the
proposed lower minimum pricing increments as well as reflect market
innovations and technological efficiencies that have driven
transaction costs down since rule 610(c) was adopted.'').
\588\ See Regulation NMS Adopting Release, supra note 4, at
37584.
\589\ See infra sections VII.C.2 and VII.D.2.b and notes 1101-
1103 and accompanying text.
---------------------------------------------------------------------------
As was the case with the original access fee caps, the amended fee
caps will preserve the agency business model because trading centers
will continue to be able to assess fees for transaction services at a
level that will result in the same net capture as they earn today if
they so choose. In this manner, the amended fee cap for protected
stocks priced $1.00 and above has been ``drafted to have minimal impact
on competition and individual business models while furthering the
objectives of the Exchange Act by preserving the fairness and
usefulness of quotations.'' \590\ In determining the new levels of the
access fee caps, the Commission has considered many factors, including
allowing for a diversity of business models.\591\ The amendments to
Rule 610(c) will continue to allow the exchanges to provide execution
services using their current business models, innovate and compete for
order flow, while also reducing the costs to investors who must access
protected quotations because access fees are being reduced to amounts
above the exchanges' net capture rates.
---------------------------------------------------------------------------
\590\ See Regulation NMS Adopting Release, supra note 4, at
37545. For the reasons discussed, the new access fee caps will
continue to ``provide the necessary support for the proper
functioning of the Order Protection Rule, and private linkages,
while leaving trading centers otherwise free to set fees subject
only to other applicable standards (e.g., prohibiting unfair
discrimination.''). Id.
\591\ See infra section VII.D.2.b. As discussed below, the fees
charged by ATSs for execution services are often in the range of 10
mils. See infra section VII.C.2, note 1118 accompanying text.
---------------------------------------------------------------------------
Exchanges are the only trading centers that currently display
protected quotes in the national market system, and they play an
important role in bringing together multiple buyers and sellers of
securities.\592\ Exchanges are also responsible for certain important
processes in the national market system, including openings, re-
openings, and closings on the primary listing market; trading halts;
initial public offerings and exclusively listed securities. The
exchanges, along with FINRA, are also responsible for producing data
for the consolidated market data feeds as well as the operation of the
exclusive SIPs. In addition, the Commission has recognized these
functions as ``critical'' to the operation of the securities markets
for purposes of imposing requirements under Regulation SCI, which
established a regulatory framework for oversight of the core technology
of the U.S. securities markets.\593\ Therefore, it continues to be
appropriate to consider the impact on this business model, i.e., the
agency market business model, when considering amendments to the access
fee caps.
---------------------------------------------------------------------------
\592\ Rule 610(c) imposes the access fee caps on trading
centers, which are defined to include other types of entities that
can display protected quotes, including ATSs, OTC market makers and
any broker or dealer that executes orders internally. 17 CFR
242.600(b)(106).
\593\ Securities Exchange Act Release No. 73639 (Nov. 19, 2014),
79 FR 72252 (Dec. 5, 2014) at 72277 (Final Rule ``Regulation Systems
Compliance and Integrity'').
---------------------------------------------------------------------------
d. Comments on the Proposed 10 Mils Access Fee Cap
There was divergence of opinion around the appropriate level of the
access fee cap for protected quotations priced at $1.00 or
greater.\594\ A number of commenters viewed 10 mils as the appropriate
level.\595\ Commenters stated that a reduction in the level of the
access fee cap to 10 mils is warranted because, among other reasons, it
will result in lower costs to investors to access protected quotes;
\596\ align access fees with other elements of investor transaction
costs (all of which have decreased); \597\ recalibrate the access fee
cap levels to reflect increased efficiencies, technological
[[Page 81659]]
advancements and structural changes in the markets since Rule 610(c)
was adopted; \598\ continue to allow for competitive business models
and innovation; \599\ and align on-exchange pricing more closely with
off-exchange venues such as ATSs.\600\ One commenter stated a 10 mil
cap ``would have the added benefit of aligning exchange fees with
prevailing ATS fees and creating a more equitable competitive landscape
across trading venues'' \601\ and another stated that ``[t]he economic
difference to a broker between routing to an . . . ATS [ ] versus an
exchange would be much smaller than it is today, if a $.0005-$.0010
access fee cap replaces the current $.0030 mil cap.'' \602\ According
to one commenter, lowering the cap to 10 mils should ``(1) lead to an
increase in investor interaction with displayed quotes, (2) provide an
economic reason for all participants to submit displayed quotes to an
exchange, and (3) end the corrosive and discriminatory nature of the
current exchange fee and rebate system.'' \603\ Another commenter
stated that reducing the level of the cap from 30 mils to 10 mils would
``remove barriers to entry for new market participants,'' especially
smaller trading firms and retail investors.\604\ Further, commenters
stated that the proposed reduction of the access fee caps would be
beneficial to retail investors, as well as institutional investors and
long-term investors.\605\
---------------------------------------------------------------------------
\594\ See, e.g., supra notes 460, 463, 466 (recommending a
reduction in fee cap to $0.0015).
\595\ See, e.g., ASA Letter at 5 (strongly supporting 10 mils
access fee cap for all NMS stocks); Better Markets Letter I at 16;
BlackRock Letter at 10-11; BMO Letter at 3-4; Brandes Letter at 3;
Boston Partners, Calamos Advisors, Glenmede Investment, and Janus
Henderson Letter; Budish Letter; Capital Group Letter at 4; Council
of Institutional Investors Letter at 3; IEX Letter IV at 1; Healthy
Markets Letter I at 24; Ontario Teachers et al. Letter at 1-2;
Themis Letter at 7-8; Vanguard Letter at 2 & 6; Invesco Letter at 2
and 4; JPMorgan Letter at 6; NASAA Letter at 9; Pragma Letter at 7;
XTX Letter at 5; and Verret Letter II. See also supra note 422.
\596\ See, e.g., Ontario Teachers et al. Letter at 2; ASA Letter
at 5; Council of Institutional Investors Letter at 3; Better Markets
Letter I at 16.
\597\ IEX Letter VI at 1-2.
\598\ See, e.g., Ontario Teachers et al. Letter at 1-2; Brandes
Letter at 3; Invesco Letter at 4; Vanguard Letter at 6; Verret
Letter I at 5; Verret Letter III at 4-5; Letter from John Ramsey,
Chief Market Policy Officer, IEX, dated Feb. 23, 2024 (``IEX Letter
V'') at 2-3; IEX Letter VI at 4.
\599\ See, e.g., Proof Letter at 1; BMO Letter at 4; Verret
Letter I at 9; Ontario Teachers et al. Letter at 1-2; Verret Letter
III at 22.
\600\ See, e.g., RBC Letter at 4; BlackRock Letter at 11; Verret
Letter I at 7; Verret Letter III at 4-5; IEX Letter V at 5; IEX
Letter VI at 5. But see Letter from Kevin R. Edgar, Partner, Baker &
Hostetler LLP, dated Feb. 7, 2024, (``Equity Markets Association
Letter'') at 2 (stating ``alleg[ations] . . . that access fees are
excessive, both in an absolute sense and relative to ATSes'' are
improper and further stating there is ``no basis for such
conclusions other than by making bald assumptions about exchanges'
costs . . . .[N]et transaction fees are far lower on exchanges than
they are on ATSes.''); Nasdaq Letter III at 3 (stating commenters'
analysis of ATS-Ns revealed ``large variations among ATS fees and
some of them are similar or higher than exchange fees . . .
including fees as high as $0.06); Nasdaq Letter IV at 7-8 (providing
data regarding range of ATSs minimum/maximum fees and stating ATS
fees are highly variable and 10 mils is not representative of
transaction fees on- or off-exchanges).
\601\ BlackRock Letter at 11.
\602\ RBC Letter at 4.
\603\ ASA Letter at 5.
\604\ See Verret Letter I at 9; Verret Letter III at 22. But see
Letter from Barbara Comstock, Executive Director, American Consumer
and Investor Institute, dated June 1, 2023 (``ACII Letter I'') at 6
(commenting generally that fundamental changes to existing market
structure could roll back innovations that have ``opened up today's
markets to millions of new and diverse investors.''); NASP Letter at
2 (commenting generally that proposed NMS changes could harm retail
investors by ``making the process of buying and selling stock more
difficult and potentially reinstating barriers to entry''); Nasdaq
Letter II at 2-3 (stating ``the cost of access fees has actually
fallen since 2005 by one-third'' and ``the burden of access fees
relative to the all-in trading costs of participants has not grown
over time; instead, it has remained relatively flat.'').
\605\ See, e.g., Council of Institutional Investors Letter at 3;
Ontario Teachers et al. Letter at 2; Themis Letter at 8; IEX Letter
IV at 13, 23; Better Markets Letter I at 16.
---------------------------------------------------------------------------
Several commenters stated their support for reducing the amount of
the access fee caps, but cautioned that further analysis is necessary
to determine the appropriate amount and parameters of any reduction to
avoid unintended consequences.\606\ The Commission disagrees for a
number of reasons. Further delay is not warranted because the access
fee caps have been extensively considered for many years.\607\ In
addition, the Commission has weighed several factors in determining to
reduce the access fee caps to 10 mils and, as discussed further below
in the Economic Analysis, concludes that this reduced level
appropriately accommodates various competing interests.\608\ The
adopted level of 10 mils for access to protected quotes priced $1.00 or
more reflects the views of many commenters to the Proposing Release
\609\ and has been suggested by market participants in other
contexts.\610\ Further, as discussed above, a 10 mil cap strikes an
appropriate balance between reducing the cap to help to address
distortions in the market associated with the preexisting fee caps,
while also preserving the ability of the national securities exchanges
to continue to operate with their current net capture rates.\611\
Finally, the Commission has reviewed the fees charged by trading
centers that do not have protected quotes so do not have an incentive
to charge excessive fees to market participants required to access
protected quotes and 10 mils is consistent with the range of rates
assessed by such trading centers.\612\
---------------------------------------------------------------------------
\606\ See, e.g., STA Letter at 7-8; ICI Letter I at 16; Nasdaq
Letter I at 2 and 19.
\607\ See supra notes 362 and 364.
\608\ See infra section VII.D.2.
\609\ See supra note 595.
\610\ See Goldman 2018 Letter at 1-2 (stating commenter's
support for reducing the access fee cap to $0.0010 because a 10 mil
cap would be calibrated with then-present-day [2018] trading and
execution costs, would better ensure displayed prices reflect the
actual economic costs of an execution, and would allow exchanges to
continue maintain their current net capture rates, while also
choosing to offer rebates to incentivize liquidity provision if they
chose to do so). Further, the EMSAC also considered, among other
things, whether the access fee cap should be modified. See supra
note 4.
\611\ See infra section VII.D.2.
\612\ See infra note 1118. The Commission acknowledges
variability within the rates assessed by ATSs, with some
transactions subject to fees above 10 mils and some below 10 mils
based on attaining certain levels of volume as well as other
variability within the fee schedules. Commenters have stated that
the fees they experience are often in the range of 10 mils, which is
informative in considering an appropriate level of the access fee
caps because such statements reflect the current market rate paid
for execution services as reported by participants. See infra notes
658-659 and accompanying text.
---------------------------------------------------------------------------
Other commenters recommended applying different access fee caps
depending on the liquidity profile of a particular security.\613\
Further, some commenters suggested specific alternative models and/or
levels of access fee caps.\614\ A few commenters stated that the access
fee caps should be expanded to cover full depth-of-book quotations
\615\ or auctions.\616\ However, one commenter disagreed with these
concerns and stated that ``the fee cap has been equally applied to all
stocks regardless of price, spread, or trading volume since it was
enacted'' and further stated that ``[e]xchange processing costs are
exactly the same'' regardless of these varying characteristics.\617\
---------------------------------------------------------------------------
\613\ See, e.g., BlackRock Letter at 10-11; T. Rowe Price at 4-
5; Citigroup Letter at 5-6.
\614\ See, e.g., MEMX Letter at 3 and 24-28; Nasdaq Letter I at
19; William O'Brien, Former CEO, Direct Edge, dated Apr. 13, 2023
(``O'Brien Letter'') at 5; Optiver Letter at 3.
\615\ See, e.g., Citadel Letter I at 25; FIA PTG Letter II at 3.
\616\ MEMX Letter at 3, 24-28.
\617\ IEX Letter VI at 3. This commenter further stated that the
``benefits that exchanges have received from technological advances
and increased efficiencies in determining their own costs to process
orders . . . apply exactly in the same way for trading in all
classes of securities'' and therefore retaining a uniform, lower
``fee cap across all stocks (priced greater than $1.00 per share)
avoids further complexity to trading decisions from fees that can
vary for the same stock based on changes in the applicable tick
size.'' Id.
---------------------------------------------------------------------------
As discussed above, the access fee caps under Rule 610 establish
the upper limit for fees that trading centers can charge for access to
protected quotations. The access fee caps do not apply to depth-of-book
quotations or auctions because these are not protected quotations. As
discussed throughout, the access fee caps are designed to preserve fair
and efficient access to protected quotations, regardless of the
liquidity profiles of NMS stocks. Trading centers are able to develop
different fee structures within this construct and in a manner that is
consistent with the Exchange Act. The Commission is not setting the
access fee caps to a specific percentage of the
[[Page 81660]]
minimum pricing increments in part because they address different
regulatory objectives. An important objective of an access fee cap (to
preserve access to protected quotes) is distinct from an objective of
tick size (e.g., to prevent stepping ahead of displayed orders).\618\
However, as discussed above, because the Commission is reducing the
minimum pricing increment under Rule 612, it is also reducing the
levels of the access fee caps to prevent the distortions that would
occur if an access fee is more than one half of the tick.\619\
---------------------------------------------------------------------------
\618\ See supra section III.A and section III.C.1.
\619\ See infra section VII.D.2.a and notes 1419 and 1425.
---------------------------------------------------------------------------
Further, amending Rule 610 to adopt variable caps to reflect
different liquidity profiles of different stocks would expand and
change the objective of the rule, which is to ensure the fairness and
accuracy of protected quotations by establishing an outer limit on the
cost of accessing such quotations.\620\ The access fee caps were not
designed to establish fees for executions, they were designed to limit
the amount of fees that can be charged for access to the best priced
quotes in the national market system. Trading centers may adopt fees
(and rebates) to incentivize trading in NMS stocks with different
liquidity profiles in a manner consistent with the Exchange Act,
including the limits imposed by the access fee caps.
---------------------------------------------------------------------------
\620\ See also note 351.
---------------------------------------------------------------------------
Another commenter stated that there is no valid basis to support a
claim that the current fee cap is excessive.\621\ This commenter stated
that the Commission did not substantiate reduced costs as a
justification for lowering the access fee caps.\622\ This commenter
also stated that, ``the Commission present[ed] no cost-based
methodology for arriving at the levels of access fee caps it proposes''
\623\ and therefore the proposed caps are ``arbitrary and capricious''
because the caps do not ``bear a reasonable relationship to the actual
costs of executing trades on the exchanges.'' \624\ The commenter
further stated that ``[t]echnology costs, and improvements thereto, are
not significant determinants of access fee levels.'' \625\ Further,
this commenter also stated that ``exchange platform costs'' (i.e., the
constellation of related services of which transaction services are
only one part) to market participants have ``remained competitive over
time.'' \626\ Finally, according to this commenter, ``access fees and
rebates represent more than the simple economic costs to an exchange of
effecting a trade; they also reflect the value of the information that
quotes provide to the market, and the value to participants of having
access to those quotes.'' \627\ Another commenter stated that the
current cap was ``rather arbitrarily selected'' and in its view has
``resulted in continued industry disagreement.'' \628\
---------------------------------------------------------------------------
\621\ Nasdaq Letter III at 2.
\622\ See Nasdaq Letter I at 21; Nasdaq Letter II at 4. This
commenter stated that ``determining such costs and setting
appropriate rates based upon those costs are inherently difficult''
and further stated that ``a government agency like the Commission is
ill-suited to tackle [such a task]'' and should refrain from doing
so. Nasdaq Letter I at 22. See also Cboe Letter III at 5; Nasdaq
Letter III; Equity Markets Association Letter at 2.
\623\ Nasdaq Letter II at 4-5.
\624\ Nasdaq Letter II at 4. See also Nasdaq Letter III at 2;
Nasdaq Letter I at 22.
\625\ Nasdaq Letter II at 4. See also Nasdaq Letter III at 2.
\626\ Nasdaq Letter II at 4-5.
\627\ Nasdaq Letter I at 21; Nasdaq Letter II at 4.
\628\ Cboe Letter II at 8.
---------------------------------------------------------------------------
However, another commenter disagreed, stating ``the current access
fees are unreasonably high when taking into consideration the lower
exchange costs stemming from increased efficiencies and technology
advancements that have occurred since 2005.'' \629\ Although other
trading costs have decreased, access fees have not and, according to
this commenter, such fees ``now represent an outsized portion of
transaction costs.'' \630\ The commenter further stated it was
appropriate for the Commission to rely on reduced costs to justify the
reduction in the access fee cap.\631\ The commenter stated because
``the 30-mil cap exceeds the typical cost to trade on non-protected
venues, it encourages investors to seek alternatives to accessing
displayed quotes'' which drives order flow to off-exchange venues.\632\
---------------------------------------------------------------------------
\629\ IEX Letter IV at 8. See also Better Markets Letter I at 16
(``There is certainly no economic justification in terms of
defraying the exchanges' costs of processing and matching trades, as
those costs have dropped with the advent of advances in
technology.''); Verret Letter III at 13 (``Access fees charged to
broker-dealers and other market participants simply to access
liquidity on certain exchanges often greatly exceed the actual costs
associated with providing that liquidity access.'').
\630\ IEX Letter IV at 8. See also Goldman Sachs 2018 Letter at
1-2.
\631\ See IEX Letter IV at 6.
\632\ IEX Letter IV at 8.
---------------------------------------------------------------------------
As discussed throughout this release,\633\ market participants have
stated that the access fee caps are outdated and no longer reflect the
current market structure. One commenter stated that the Commission, by
identifying that the markets have changed due to market innovations and
technological efficiencies and that transaction and trading costs had
been reduced, and providing statements of market participants to
support this statement,\634\ was suggesting that the access fee cap
``no longer bears a reasonable relationship to the actual costs of a
trade.'' \635\ This misconstrues the Commission's statement recognizing
that the markets are different than they were in 2005. Under the
preexisting access fee caps, fee and rebate structures have developed
such that access fees are predominantly used to pay rebates to
liquidity providers and these structures have resulted in distortions
in the market.
---------------------------------------------------------------------------
\633\ See, e.g., supra section IV.B.2.
\634\ See Proposing Release, supra note 11, at 80290 n.293.
\635\ Nasdaq Letter I at 21. See also Nasdaq Letter II.
---------------------------------------------------------------------------
The Commission considered many factors and the views of commenters,
and balanced competing factors when it adopted the original fee caps in
2005.\636\ Likewise, as discussed above, the Commission again has
considered and balanced many factors,\637\ including the effects on
liquidity and trading costs for market participants \638\ in coming to
the determination that the 10 mils access fee cap is appropriate for
all protected quotations priced $1.00 or more.\639\ As discussed
throughout this release, the Commission has reduced the caps to a level
that is sufficient to mitigate the market distortions associated with
the fee schedules that have been developed under preexisting access fee
caps and to accommodate the new minimum pricing increments under
amended Rule 612, while also preserving the viability of the agency
market business model.\640\
---------------------------------------------------------------------------
\636\ See generally, Regulation NMS Adopting Release, supra note
4.
\637\ See supra note 608 and surrounding text.
\638\ See infra section VII.D.2.c (analyzing the effects of
rebates for providing liquidity).
\639\ Several commenters stated that access fees under the
preexisting caps have become a larger portion of overall transaction
costs because such costs have decreased significantly since the
access fee cap levels were established almost two decades ago. See,
e.g., infra notes 1438-1441 and accompanying text and supra 364 and
365 and supra notes 597-598 (describing reasons why costs have
decreased). The Commission has considered costs, and specifically
commenters' concerns relating to costs, as one of several factors in
its analysis and determination that a 10 mils access fee cap is
appropriate. As the Commission stated in 2005, reaching appropriate
policy decisions in a complex area such as fees for access to the
best quotations displayed in the national market system requires
balancing policy objectives that sometimes may not point in
precisely the same direction. See Regulation NMS Adopting Release,
supra note 4, at 37498.
\640\ See supra section IV.D.1.c. and infra section VII.D.2.b.
---------------------------------------------------------------------------
Most exchanges provide access to protected quotations and retain an
estimated net capture of 2 mils.\641\
[[Page 81661]]
However, as discussed below,\642\ a net capture of 2 mils is not
uniform across all exchanges and some have an estimated net capture
that is higher than 2 mils.\643\ This suggests that the preexisting
levels of the access fee caps are higher than necessary to preserve the
viability of the agency market business models. The adopted level of 10
mils for access to protected quotes priced $1.00 or more is appropriate
because it will allow trading centers to continue to provide access to
protected quotations and retain a net capture to fund their transaction
services. Recalibrating the level of the cap with a consideration of
current market rates to provide execution services is appropriate and
consistent with how the Commission set the preexisting rates.\644\
---------------------------------------------------------------------------
\641\ See supra section IV.D.1.c. and infra sections VII.C.2 and
VII.D.2.b and notes 1101--1103 and accompanying text. As discussed
below in the Economic Analysis, the Commission estimates for
purposes of this release that exchange net capture is 2 mils, while
also recognizing that net capture can range from approximately 2 to
6 mils. See infra note 1103.
\642\ See id.
\643\ See id.
\644\ See infra note 357 and discussion below.
---------------------------------------------------------------------------
Finally, as stated above, the access fee caps were not developed as
a means to enable the payment of rebates. However, under the
preexisting access fee caps, access fees are predominantly used to fund
the rebates paid to liquidity providers. As also stated above and
discussed further below, liquidity providers are able to post bid and
offer prices that account for the risk of displaying protected
quotations without needing the payment of a rebate.\645\
---------------------------------------------------------------------------
\645\ See supra section IV.B.2.b. See also infra section
VII.D.2.c and note 1458 and accompanying text.
---------------------------------------------------------------------------
In deciding to adopt a single 10 mil fee cap for all protected
quotes in NMS stocks priced $1.00 or more, the Commission has also
considered the rates charged by other agency markets for access to non-
protected quotation liquidity because such trading centers are not
subject to the preexisting 30 mils access fee cap and therefore the
rates for execution services established by such markets are subject to
competitive market forces that are not capped.\646\
---------------------------------------------------------------------------
\646\ See infra notes 1116-1118 and accompanying text.
---------------------------------------------------------------------------
One commenter stated that ATS fees are not a good benchmark to
determine the appropriate level of exchange access fees because, in
their view, exchange access fees should be higher than off-exchange
venues' access fees.\647\ This commenter stated that ``[e]xchange
access fees compensate for the risk associated with posting lit quotes
as well as the value associated with accessing immediate liquidity.''
\648\ In addition, according to this commenter, exchange pricing is
``designed to attract quotes, whereas ATS pricing is designed only for
trades'' and ATSs ``leverage lit quotes'' produced by exchanges to
determine ATS's transaction pricing.\649\ Finally, this commenter
stated that the rates charged by off-exchange venues ``vary
significantly in structure, functionality, and fees'' to the extent
they are actually known publicly and disagreed with the conclusion that
``10 mils is a representative fee for accessing liquidity off
exchange.'' \650\
---------------------------------------------------------------------------
\647\ Nasdaq Letter IV at 8.
\648\ Nasdaq Letter IV at 8.
\649\ Id.
\650\ Nasdaq Letter IV at 7. This commenter further stated
``nothing beyond anecdotal reports suggests that 10 mils is a
representative fee for accessing liquidity off exchange.'' Id.
---------------------------------------------------------------------------
Other commenters disagreed.\651\ According to one commenter,
certain ATSs provide specialty services such as block trading and the
ability to use conditional order types to achieve certain trading
strategies, and typically charge higher than 10 mils for such
specialized services.\652\ However, according to this commenter, ATSs
that operate a continuous book market are similar to exchanges that
provide similar services and those ATSs charge ``a maximum rate of 10
mils'' for such services and such venues collectively represent
approximately 42% of all ATS volume during 2023.\653\ According to this
commenter, ``this data is strong evidence that the standard comparative
rate for immediate access to liquidity in NMS stocks on ATSs that offer
this [continuous book] service is, in fact, 10 mils per share.'' \654\
This commenter further stated that exchanges are ``able to charge
higher prices than other markets, precisely because of the `protected
quote' status'' which is ``what the SEC sought to prevent in 2005, in
furtherance of the statutory goal of fair distribution of quotation
information.'' \655\
---------------------------------------------------------------------------
\651\ See, e.g., IEX Letter VI at 5 (stating ``ATSs that accept
and process orders for NMS stocks in the same way as exchanges do
characteristically charge in the range of 10 mils per share.'');
BlackRock Letter at 11 (stating that 10 mil cap ``would have the
added benefit of aligning exchange fees with prevailing ATS fees and
creating a more equitable competitive landscape across trading
venues''); Verret Letter I at 7 (lowering access fee cap from 30
mils to 10 mils would be ``more in line with the fees charged by
most ATS platforms,''); IEX Letter V at 5 (stating because ATS fees
``are affected by market forces and not pegged by regulation, they
are highly relevant to the question of where to set an updated fee
cap.''). See also Letter from Stacey Cunningham, President, NYSE, to
Brent Fields, Secretary, Commission, dated Oct. 2, 2018 (commenting
on File No. S7-05-18 ``Transaction Fee Pilot for NMS Stocks'')
(stating reducing the access fee cap to 10 mils will bring the
access fees exchanges charge to remove liquidity in line with the
rates charged by ATSs).
\652\ IEX Letter VI at 5-6; IEX Letter V at 5-6.
\653\ IEX Letter VI at 5 (referencing public data showing ATS
access fees in the range of 10 mils and below). See also IEX Letter
V at 5-6.
\654\ IEX Letter VI at 5 (stating 10 mils is the relevant
comparative rate charged by ATSs).
\655\ IEX Letter IV at 6. But see Nasdaq Letter V at 2 (stating
``ATSes enjoy advantages [including the ability to segment order
flow] that would persist, and likely increase, with a lower cap on
access fees.'').
---------------------------------------------------------------------------
The fees charged by many ATSs that provide execution services
similar to exchanges are often reflected in a range and sometimes are
based on volume transacted, and ATSs typically do not pay rebates.\656\
As stated above, several commenters stated that a 10 mils access fee
cap would be consistent with the access fees charged by ATSs.\657\
These statements are informative in considering an appropriate level of
the access fee caps because they reflect the current market rate paid
for execution services as reported by market participants.\658\ This,
in concert with the net capture rates discussed above, suggests that
the current access fee caps may not be consistent with current market
rates for providing execution services.\659\
---------------------------------------------------------------------------
\656\ See infra sections VII.C.2.b and VII.D.2, note 1118 and
accompanying text. See also Proposing Release, supra note 11, at
80314 (stating a review of form ATS-Ns on which ATSs provide a range
of the fees charged shows such fees are often in the range of 10
mils).
\657\ See supra note 651.
\658\ See, e.g., IEX Letter I at 22-23; IEX Letter IV at 14-15;
IEX Letter V at 5-6; BlackRock Letter at 11; Verret Letter II at 4.
See also Letter from Stacey Cunningham, President, NYSE, to Brent
Fields, Secretary, Commission, dated Oct. 2, 2018 (commenting on
File No. S7-05-18 ``Transaction Fee Pilot for NMS Stocks'') (stating
reducing the access fee cap to 10 mils will bring the access fees
exchanges charge to remove liquidity in line with the rates charged
by ATSs). According to one commenter, ``the rates charged by ATSs to
access liquidity allow comparison to market-based prices that are
not affected by prices imposed by exchanges to access protected
quotes . . . [and] an informal survey of ATS operators indicates
that the standard access fee charged by most ATSs is approximately
10 mil.''). See also supra note 651 and accompanying text.
\659\ See Letter from Theodore R. Lazo, Managing Director &
Associate General Counsel, SIFMA, to Brent J. Fields, Secretary,
Commission, dated Mar. 9, 2017, at 8 (stating ``a significant
portion of access fees are used to subsidize rebates with the
exchanges' net capture reflecting today's market norms for accessing
liquidity, which is approximately 3-5 cents per 100 shares traded .
. . or 3-5 mils.).''
---------------------------------------------------------------------------
Considering the rates charged by trading centers is consistent with
the analysis the Commission conducted to determine the appropriate
level of the preexisting access fee caps when it adopted them.
Specifically, the Commission considered the access fees charged by ECNs
and other types of trading centers, including self-regulatory
organizations (``SROs''), when it adopted the preexisting 30 mil
[[Page 81662]]
access fee cap to gain insight into the then current market rates for
execution services.\660\ At the time of adoption in 2005, the $0.0030
fee limitation was based on the then-prevailing market rates for
execution services and general business practices, as very few trading
centers charged fees in excess of that amount.\661\ As it did in 2005
in establishing the preexisting fee caps, to determine the appropriate
level of the amended access fee caps, the Commission has similarly
considered the current market rate for execution services as measured
by the rates charged by other trading centers as a factor in
considering the level of the adopted access fee caps.\662\ This factor
is useful in calculating the level of the access fee caps, but it is
not the only factor. The Commission is balancing the need to set a
level of the access fee caps to allow for fair and efficient access
while also seeking to ensure that trading centers are not impaired in
their ability to provide execution services.
---------------------------------------------------------------------------
\660\ See Regulation NMS Adopting Release, supra note 4, at
37545.
\661\ See Regulation NMS Adopting Release, supra note 4, at
37545 (stating that the $0.0030 per share cap largely codified the
then-prevailing fee level set through competition among the various
trading centers).
\662\ See supra note 357 and infra section VII.D.2.
---------------------------------------------------------------------------
e. Protected Quotes Priced Under $1.00
With respect to the access fee cap for protected quotations priced
under $1.00, one commenter stated its view that ``the negative impact
of the proposed access fee caps is much more pronounced for securities
priced less than $1.00.'' \663\ The commenter stated that for protected
quotations priced less than $1.00, the ``estimated revenue impact to
exchanges providing rebates in these securities [those priced below $1]
is not insignificant'' and that ``the access fee cap must remain
unchanged to support competition, differentiation, and liquidity
provision.'' \664\ This commenter also stated that ``[s]implistic
proportionality is not a sufficient justification for this reduction''
and instead ``[a]nalysis of whether there will be proportionate
outcomes is necessary to overcome the arbitrary and capricious nature
of this reduction.'' \665\ Finally, this commenter stated that the
reduction would ``likely impact exchanges' ability to differentiate, as
well as materially limit the transaction revenue that exchanges apply
towards developing innovative solutions that contribute to the
robustness of the U.S. marketplace.'' \666\
---------------------------------------------------------------------------
\663\ Cboe Letter II at 9. The Commission conducted similar
analysis when it adopted preexisting Rule 610(c), and as discussed
below, having different access fee caps apply to a bid that is
priced under $1.00 and an offer that is priced over $1.00 in the
same NMS stock would create pricing distortions. See infra section
VII.D.2.c.
\664\ Cboe Letter II at 9.
\665\ Cboe Letter II at 9.
\666\ Cboe Letter II at 9.
---------------------------------------------------------------------------
One commenter stated that reducing the access fee cap generally
would reduce the disincentive to trade on exchanges because costs would
be lower.\667\ Another commenter stated that the estimated loss in net
capture due to the reduction in the access fees for protected quotation
priced below $1.00 ``will not harm the major exchanges.'' \668\
---------------------------------------------------------------------------
\667\ Better Markets Letter I at 16.
\668\ Themis Letter at 7.
---------------------------------------------------------------------------
The Commission is adopting a modified access fee cap of 0.1% of the
share price for protected quotations priced under $1.00. The Commission
proposed a lower access fee cap of 0.05% of the quotation price per
share in light of the proposed 5 mils access fee cap. Since the 5 mils
access fee cap is not being adopted, the Commission has modified the
access fee cap for protected quotes priced under $1.00 so that it is
consistent with the 10 mils access fee cap for protected quotes priced
$1.00 or more.
The adopted 0.1% access fee cap will align this cap with the 10
mils access fee cap that will apply to protected quotations in NMS
stocks priced $1.00 or greater.\669\ This alignment is consistent with
the access fee caps that apply under the preexisting rule, which are 30
mils and 0.3% respectively. Alignment of the access fee caps for
protected quotations in NMS stocks priced below $1.00 and those priced
$1.00 and above is necessary to preserve continuity at the $1.00 cutoff
to ensure that cost to access a protected quote for an NMS stock that
is priced below $1.00 is not more compared to the cost to access a
protected quote for the same NMS stock that is priced $1.00 or
more.\670\ For example, an NMS stock could have a protected bid that is
priced below $1.00 and a protected offer that is priced above $1.00. If
the access fee cap for protected quotes priced below $1.00 remained at
the preexisting level, the access fee for the protected bid would be
almost three times higher than the access fee for the protected
offer.\671\ In such an instance, it would cost more to trade against
the bid than to trade against the offer and negatively impact
incentives for accurate price formation.\672\
---------------------------------------------------------------------------
\669\ See infra section VII.D.2.c. As the price of a stock moves
across the $1.00 cutoff, its access fee would not experience a
discontinuous jump because 0.1% of $1.00 is 0.1 cents, i.e., 10
mils. Such alignment prevents the anomalous result that could occur
if the NBB was priced under $1.00 and the NBO was priced over $1.00
and each protected quote would be subject to a different access fee.
\670\ See also infra section VII.D.2.c.
\671\ See also infra section VII.E.2.b.
\672\ See also infra section VII.D.2.c.
---------------------------------------------------------------------------
The Commission understands that reducing the access fee cap for
stocks priced below $1.00 could reduce exchange revenue.\673\ This is
because exchanges typically charge the maximum fee of .3% for accessing
protected quotes priced less than $1.00 and do not offer rebates, or
offer rebates in small amounts.\674\ Accordingly, exchanges typically
retain the full amount of the access fee charged. However, in order to
prevent the distortions that would occur if a higher access fee cap
were applied to protected quotes priced less than $1.00, the Commission
is adopting the percentage cap that is aligned with the access fee cap
that is applicable to protected quotes priced $1.00 or more to preserve
continuity at the $1.00 cutoff.
---------------------------------------------------------------------------
\673\ See table 14, infra section VII.D.2.b.
\674\ See infra note 1433 and accompanying text.
---------------------------------------------------------------------------
f. Comments on Implementation
Some commenters stated that the Commission should build in an
evaluation process to assess the benefits and any potential
degradations to market quality resulting from changes to the access fee
caps.\675\ One commenter stated that the rule should ``include a
mechanism in the rule to periodically re-evaluate the access fee caps
set in the proposal to ensure that access fee levels continue to have
the anticipated benefits.'' \676\ Others recommended application of the
new access fee caps to a smaller subset of NMS stocks before rolling it
out to all NMS stocks.\677\ One commenter stated that changes to access
fees should be adopted as part of a pilot to allow for a study of the
effects on market quality.\678\ Another commenter stated the Commission
should first collect more data and industry input and conduct further
analysis to determine the optimal access fee cap levels before
proceeding.\679\ One commenter, however, disagreed that any delay to
collect further data or conduct additional analysis was warranted.\680\
[[Page 81663]]
According to this commenter, there is a ``mountain of evidence
supporting a reduction in the access fee cap from current levels'' and
``general consensus in favor of reducing the cap.'' \681\
---------------------------------------------------------------------------
\675\ See, e.g., Citigroup Letter at 6; NASAA Letter at 9; MEMX
Letter at 41; Nasdaq Letter I at 2; and Nasdaq Letter IV at 12.
\676\ NASAA Letter at 9.
\677\ See, e.g., STA Letter at 8; State Street Letter at 5; CCMR
Letter at 27; GTS Letter at 6-7; Nasdaq Letter I at 30.
\678\ CCMR Letter at 27.
\679\ See, e.g., T. Rowe Price Letter at 3-5. See also SIFMA
Letter II at 39-40; Chamber of Commerce Letter at 1; Cboe, State
Street, et al. Letter at 3; Citadel Letter I at 24-25.
\680\ See IEX Letter III at 4 (``There is clear evidence that
the 30-mil `limit' has acted to keep access fees artificially high,
leading to price distortions and increasing costs to institutional
investors in particular.'').
\681\ IEX Letter III at 4. See also Verret Letter II at 4; IEX
Letter III.
---------------------------------------------------------------------------
The Commission disagrees with commenters' suggestions that further
study be conducted before adopting the amendment or that the amendment
should be incrementally rolled out. A delayed or incremental approach
to reducing the preexisting access fee caps would delay the benefits to
investors of the reduced caps. The Commission has extensively
considered the adopted amendments, reviewed all comments letters and
conducted extensive economic analysis in deciding to adopt this
amendment. The amended access fee caps will provide savings for
investors and should be implemented.
E. Final Rule 610(d) Requiring That All Exchange Fees and Rebates Be
Determinable at the Time of an Execution
Many exchange fees and rebates are calculated at the end of the
month, which impedes the ability of market participants, including
investors, to understand at the time of execution the full cost of
their transaction. For example, the exchanges have developed complex
fee and rebate schedules, some of which include tiers or other
incentives based on a market participant's relative monthly trading
volume or relative volume compared to the consolidated trading volume
in the current month, with higher volume tiers receiving a higher
(lower) per unit rebate (fee). This means that the exact fee or rebate
amount for an order cannot be determined until the end of the month,
after an execution occurs, and is not known to the parties to the trade
at the time of execution. Further, uncertainty regarding the fee amount
at the time of execution can hinder the ability of market participants
to conduct best execution analyses and can affect order routing
decisions.
To provide further transparency regarding transaction pricing, the
Commission proposed to amend Rule 610 to add a new subsection (d)
``Transparency of Fees,'' which would prohibit a national securities
exchange from imposing, or permitting to be imposed, any fee or fees,
or providing, or permitting to be provided, any rebate or other
remuneration (e.g., discounted fees, other credits, or forms of linked
pricing) for the execution of an order in an NMS stock unless such fee,
rebate or other remuneration can be determined by the market
participant at the time of execution. As the Commission explained in
the Proposing Release, under proposed Rule 610(d), any national
securities exchange that imposes a fee or provides a rebate that is
based on a certain volume threshold, or establishes tier requirements
or tiered rates based on minimum volume thresholds, would be required
to set such volume thresholds or tiers using volume achieved during a
stated period prior to the assessment of the fee or rebate so that
market participants are able to determine what fee or rebate level will
be applied to any submitted order at the time of execution.\682\ For
example, if an exchange proposed a lower fee for members that reach a
certain level of trading volume in a month, the required level of
trading volume would have to be achieved based on a month prior to the
imposition of the fee or payment of the rebate.\683\
---------------------------------------------------------------------------
\682\ National securities exchanges establish and amend their
fee schedules by filing proposed fee rule changes, pursuant to
section 19(b) of the Exchange Act and rule 19b-4 thereunder, for
Commission review. National securities exchange fee schedules are
posted on their websites. See Rule 19b-4(l). Some national
securities exchanges currently use volume calculated on a monthly
basis to determine the applicable threshold or tier rate. See, e.g.,
fee schedules of Nasdaq PSX available at https://www.nasdaqtrader.com/Trader.aspx?id=PSX_pricing (as of Mar. 2024)
(calculating fees based on ``average daily volume during the
month'') and Cboe EDGA available at https://www.cboe.com/us/equities/membership/fee_schedule/edga/ (as of Mar. 2024)
(calculating fees based on ``average daily volume'' and ``total
consolidated volume'' on a monthly basis).
\683\ This amendment to Rule 610 does not alter an exchange's
ability to determine the measurement period during which volume is
calculated (e.g., a week prior, two weeks prior, or prior monthly),
rather the rule will instead require the measurement period to be
prior to the date of execution so that market participants can
determine the amount of the fee at the time of execution.
---------------------------------------------------------------------------
The Commission has considered commenters' views (as discussed
below) and is adopting Rule 610(d) as proposed. Investors can use this
information to assess their broker-dealer's routing decisions and such
information will help to inform market participants' best execution
analysis.
1. General Comments
The Commission received comments from a broad range of commenters
who stated that proposed Rule 610(d) would provide enhanced
transparency surrounding transaction fees and rebates \684\ and
alleviate concerns related to potential conflicts of interest.\685\ One
commenter stated that proposed Rule 610(d) would ``shed greater
transparency on the use of fee and rebate tiers and their impact on
individual trades'' and ``help to address concerns related to conflicts
of interest[ ] because . . . investors will be in a better position to
identify and seek the recovery of rebates that accrue specifically to
their orders . . .'' \686\ Further, one commenter stated that such a
change is ``a great step forward and long overdue,'' \687\ and another
commenter stated that it would be ``a positive outcome for the industry
and investors and w[ould] reduce market complexity and increase
transparency.'' \688\ One commenter stated that Rule 610(d) ``has the
potential to facilitate broker-dealers in passing-through access fees
and rebates to their customers, and in doing so, it could alleviate
concerns [ ] about perceived conflicts-of-interest associated with the
maker-taker model and the provision of exchange rebates to broker-
dealers.'' \689\
---------------------------------------------------------------------------
\684\ See, e.g., Ontario Teachers et al. Letter at 2; ASA Letter
at 6; Angel Letter at 8; Letter from Kelvin To, Founder and
President, Data Boiler Technologies, LLC, dated Apr. 12, 2023
(``Data Boiler Letter II'') at 3 (agreeing with Angel Letter);
Citigroup Letter at 6; BMO Letter at 4; Council of Institutional
Investors at 4. See also Comment Letter Type H, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
\685\ See, e.g., BMO Letter at 4; NASAA Letter at 9.
\686\ IEX Letter I at 28-29. See also Letter from Stanislav
Dolgopolov, Chief Regulatory Officer, Decimus Capital Markets, LLC,
dated Mar. 31, 2023, at 3 (``Decimus 2023 Letter'').
\687\ Angel Letter at 8. See also BMO Letter at 4; Healthy
Markets Letter I at 25.
\688\ Fidelity Letter at 15. See also ICI Letter I at 17.
\689\ Nasdaq Letter I at 32. See also Letter from Tyler
Gellasch, President & CEO, Healthy Markets Association, dated Aug.
1, 2023 (``Healthy Markets Letter II'') at 11.
---------------------------------------------------------------------------
Other commenters' support for Rule 610(d) was more measured because
they stated that the proposal did not go far enough to address market
distortions resulting from fee and rebate tiers.\690\ One such
commenter stated that although it was ``encouraged about eliminating
the retroactive attributes of exchange volume tiers,'' it felt a ``more
optimal solution [ ] would be to remove them entirely'' \691\ because
``exchange
[[Page 81664]]
volume tiers create barriers to entry that only benefit the largest,
most active trading firms at the expense of smaller competitors.''
\692\ One commenter stated its support for Rule 610(d), but also stated
that the Commission should ``take additional steps . . . to prohibit or
restrict the use of CADV-based tiers,'' which in this commenter's view
are ``by their nature [ ] highly anti-competitive and discriminatory.''
\693\ Further, one commenter ``encouraged the Commission to review and
address the issue of `bespoke' pricing tiers prevalent in today's
volume tiered pricing models.'' \694\ Another commenter, while agreeing
with the objective of fee transparency, was skeptical that Rule 610(d)
``would `materially reduce' uncertainty regarding the fee amount at the
time of execution'' and stated that the proposal would provide ``little
practical transparency for most Market Participants.'' \695\
---------------------------------------------------------------------------
\690\ See, e.g., RBC Letter at 5; ASA Letter at 5; IEX Letter I
at 29; Letters from Kelvin To, Founder and President, Data Boiler
Technologies, LLC, dated Mar. 31, 2023 (``Data Boiler Letter I'') at
5; Themis Letter at 1 (expressing support for the proposal, but
expressing disappointment that the Commission did not go further and
calling for the elimination of rebates); Healthy Markets Letter I at
24 (``If two different brokers send the exact same order to an
exchange, they should get the same pricing for that order. Pricing
should be based on the order being sent, not the other business or
trading by the party sending it.'').
\691\ RBC Letter at 5. See also Proof Letter at 1-2 (supporting
requiring exchange pricing to be computable at the time of the trade
but questioning what, if any, impact this will have on complexity of
existing pricing tiers and expressing preference that the Commission
adopt a ``more drastic policy change.''); IEX Letter I at 6.
\692\ RBC Letter at 5. See also Citigroup Letter at 6; Proof
Letter at 1-2.
\693\ IEX Letter I at 29. See also Citigroup Letter at 6; John
Ramsay, Chief Market Policy Officer, Investors Exchange LLC, dated
Sept. 20, 2023 (``IEX Letter II'') at 4; Healthy Markets Letter II
at 3-6. However, not all commenters agree that volume-based fee/
rebate tiers are anticompetitive. See, e.g., Cboe Letter III at 6-7
(arguing ``volume-based tiers do not restrain trade or represent a
burden on competition . . . By contrast, limiting volume-based
rebate tiers would in fact harm competition and disadvantage the
very small and mid-sized brokers who support this myth.'').
\694\ BMO Letter at 4. See also Fidelity Letter at 15; Proof
Letter at 1-2.
\695\ Pragma Letter at 7-8.
---------------------------------------------------------------------------
Rule 610(d) will provide additional certainty, transparency and
clarity to exchange fee structures, which will assist investors and
other market participants in assessing their order placement. Further,
certainty about the cost of a transaction at the time of the trade will
help broker-dealers make more informed order routing decisions,
particularly benefitting customers that are sensitive to transaction
costs at the execution venue, because broker-dealers and their
customers will know with more certainty the cost of an exchange
transaction at the time of the trade. Investors will be able to obtain
or request at the time of execution details about the exchange fees and
rebates assessed on their orders without having to wait weeks until
that pricing is determined and invoiced.
In addition, because the rule will allow market participants to
know the amount of fees and rebates that are applicable to their
transactions at the time of the trade, the rule will facilitate the
ability of broker-dealers to pass back to their customers, if the
customer requests and the customer and the broker-dealer both are able
to accommodate the pass-through of fees, rebates, and other forms of
remuneration in a more timely fashion.\696\ Today, lower fees or higher
rebates based on volume achieved in a current trading month can lead to
routing for purposes of achieving a certain level of volume or
attaining a possible tier level rather than routing solely to achieve
best execution. While tiers that are based on volume from a previous
time-period may still incentivize routing by a broker-dealer to try to
secure a higher rebate/lower fee tier in the following month, certainty
regarding what tier applies at the time of trade will facilitate the
ability of a broker-dealer to pass those fees and rebates through to
their customers, if they so decide,\697\ on a more timely basis because
they will be known at the time of the trade.\698\ Requiring certainty
regarding the amount of the fee/rebate is an incremental step toward
addressing commenters' concerns regarding the ability of exchanges and
brokers to pass back the actual fee or report on each transaction.
---------------------------------------------------------------------------
\696\ However, a broker-dealer may choose not to offer pass-
through of fees, rebates and other forms of remuneration to its
customers or may choose not to pass through the entirety of the
incentive it receives, or the customer may not want or be able to
accommodate such pass-throughs. In these cases, a conflict of
interest would continue to exist between the broker-dealer and its
customer when the broker-dealer routes the customer's order for
execution based on the broker-dealer's economic benefit from its
routing decision. Notwithstanding, even if pass-through of fees,
rebates and other forms of remuneration to customers does not
happen, Rule 610(d) will provide certainty regarding the applicable
fee/rebate at the time of execution, which will facilitate a
customer's ability to evaluate their broker's routing decisions and
could improve broker-dealer accountability, provide greater
transparency regarding executions and lead to improved order
execution for customers. See infra section VII.D.3.
\697\ See id.
\698\ See infra section VII.D.3.
---------------------------------------------------------------------------
If market participants pass through in their entirety the exchange
fees/rebates to their customers, an ancillary benefit of the new rule
will be that the potential inducement to broker-dealers to route orders
based on garnering the highest rebate/paying the lowest fee will be
reduced since a broker-dealer would no longer retain for itself the
transaction pricing benefit from its routing decision. The new rule
also will facilitate a customer's ability to obtain more timely
information about what exchange transaction pricing the broker-dealer
receives, which may increase accountability of the broker-dealer to the
customer in ways that could lead to better order execution and more
transparency regarding the fees/rebates applicable to a particular
order.
Other commenters voiced support for the Commission's objectives in
proposing Rule 610(d), but suggested certain modifications. One
commenter stated that it ``did not object in principle to . . .
requir[ing] exchanges to set volume-based access fees and rebates as of
the time of execution'' provided other market centers would be held to
``the same standards of transparency.'' \699\ As discussed in the
Proposing Release, exchange fees and the fees of non-exchange trading
centers are treated differently under the Federal securities laws.\700\
Specifically, non-exchange fees are not subject to the requirements
applicable to exchange fees under sections 6(b) and 19(b) of the
Exchange Act \701\ and rule 19b-4 thereunder.\702\ Exchange fees are
subject to the requirements of the Exchange Act and the rules
thereunder, which requires, among other things, that every exchange
post and maintain a current and complete version of the entirety of
each and every fee, due, and charge assessed by an exchange,\703\ and
that such exchange rules applicable to fees must provide for the
equitable allocation of reasonable dues, fees, and other charges among
its members and issuers and other persons using its facilities and not
be designed to permit unfair discrimination between customers, issuers,
brokers or dealers.\704\ Rule 610(d) is consistent with that statutory
framework as it provides members and their customers with more
certainty and transparency at the time of trade when exchange
transaction pricing may be relevant to, and impactful on, the broker-
dealer's order routing decision. New Rule 610(d) is narrowly tailored
to improve certainty and transparency regarding exchange fees and
rebates within the current regulatory framework.
---------------------------------------------------------------------------
\699\ Nasdaq Letter I at 2, 31-32.
\700\ See Proposing Release, supra note 11, at 80287. If an ATS
or OTC market maker displayed a protected quotation, its fees would
be subject to the access fee caps under rule 610(c). However,
exchange fees and the fees of non-exchange trading centers are
treated very differently under the Federal securities laws. For
example, one of the distinguishing features of registered national
securities exchanges is that--unlike non-exchange trading centers--
their fees are subject to the principles-based standards set forth
in the Exchange Act, as well as the rule filing requirements
thereunder.
\701\ 15 U.S.C. 78(f)(b) and (s)(b).
\702\ 17 CFR 240.19b-4.
\703\ 17 CFR 240.19b-4(m)(1).
\704\ 15 U.S.C. 78f(b)(4) and (5). See also Proposing Release,
supra note 11, at 80287.
---------------------------------------------------------------------------
One commenter stated that the proposal might make it more difficult
[[Page 81665]]
for smaller broker-dealers to compete against established firms because
``participation in the exchanges' growth programs'' might become more
expensive in the initial month of participation and ``limit exchanges'
ability to incent market makers and other participants to quote at the
NBBO and to do so in a large number of securities, including thinly-
traded securities.'' \705\ The commenter did not provide detail as to
how these impacts would arise, but requested the Commission to ``exempt
growth programs and special pricing programs that reward market makers
and other participants for quoting at the NBBO and providing market
quality'' from the requirements of Rule 610(d).\706\
---------------------------------------------------------------------------
\705\ Nasdaq Letter I at 33.
\706\ Id. at 33.
---------------------------------------------------------------------------
Finally, some commenters did not support proposed Rule 610(d)
because they stated it would negatively impact the ability to
incentivize liquidity provision, ``disrupt[ ] existing economic
incentives without justification,'' \707\ add ``an unnecessary layer of
complexity,'' \708\ and inappropriately ``wade into the business of
telling private companies how to charge their customers.'' \709\ One
commenter stated that ``existing fee constructs such as volume-based
pricing tiers are important tools that allow exchanges to compete with
one another and with non-exchanges.'' \710\ This commenter further
stated that volume-based tiers are entirely consistent with the
Exchange Act and vital tools exchanges use to ``incentivize greater
participation and improve liquidity and market quality.'' \711\
Finally, one commenter stated that the Commission ``include[d] no data
[ ] showing this change would cure any harm, nor does it claim any
anticipated benefits that might flow from this change.'' \712\
---------------------------------------------------------------------------
\707\ Cboe Letter II at 9-10 (stating that removing incentives
provided by exchange rebate tiers would drive liquidity off-exchange
and negatively impact exchange liquidity provision). See also Data
Boiler Letter I at 28-29 (stating that determinable requirement will
lead to increased costs that will be passed along to customers as
higher commissions or reduced services and could lead to higher
barriers to entry because the requirement may cause a ``direct hit
to broker-dealers' bottom line'' which will force them to ``find
alternative ways to squeeze, exploit, or rent seek to cover their
losses'' and urging adoption of ``Copyright Licensing mechanism''
instead.).
\708\ Virtu Letter II at 9-10 (stating the ``effort required to
understand the volume fee system, forecast volume fees for an
upcoming period, and confirm that fees are indeed being calculated
appropriately will especially disadvantage smaller brokers, who
typically have less resources . . . for needless work such as
this.'').
\709\ Virtu Letter II at 9.
\710\ Cboe Letter III at 2.
\711\ Cboe Letter III at 6-7.
\712\ Virtu Letter III at 9-10.
---------------------------------------------------------------------------
The new rule does not prohibit exchange liquidity provision
incentives nor add complexity as suggested by some commenters. It
instead shifts the time of calculating fees or rebates so that
investors and other market participants are informed, when placing an
order, of the amount of the fee or rebate that will be assessed. Rule
610(d) does not alter an exchange's ability to offer incentive programs
based on volume tiers or any another metric; rather it will provide
prospective certainty regarding what fee/rebate the market participant
will incur/earn by achieving the requisite benchmark. Therefore,
exceptions for liquidity provision incentives are not appropriate or
necessary. Further, Rule 610(d) does not require exchange fees, rebates
or other remuneration to be based on activity from a specific
measurement period provided the metric used can be achieved prior to
the time of execution, nor does it impose any obligations or additional
costs on market participants to perform any new calculations, make new
projections or forecasts, or undertake any new responsibilities. The
Commission is not requiring market participants to undertake any
obligations regarding the calculation of the applicable fee/rebate.
Instead, Rule 610(d) will facilitate a market participant's ability to
know the amount of the fee/rebate based on historical (rather than
future) volume, so that it can understand how a volume-based fee/rebate
will apply at the time of execution.\713\ Rule 610(d) will allow market
participants to calculate the amount of the fee/rebate using data
available at the time of execution rather than have to forecast or
estimate the cost of their transaction. As discussed below, this
certainty and transparency regarding the fee and rebate that will apply
to a particular transaction will benefit market participants and
improve market quality.\714\
---------------------------------------------------------------------------
\713\ Implicit in the use of historical rather than future
volume is that market participants will know the amount of the fee/
rebate that will apply at the time of the execution. As discussed
above, exchange fees are subject to the requirements of the Exchange
Act and the rules thereunder, which require, among other things,
that every exchange post and maintain a current and complete version
of the entirety of each and every fee, due, and charge assessed by
an exchange, Because information necessary to calculate the amount
of the fee or rebate will be knowable at the time of execution, Rule
610(d) will provide market participants with the ability to
determine the fee or rebate due at the time of execution.
\714\ See infra section VII.D.3.
---------------------------------------------------------------------------
Further, one commenter requested the Commission withdraw proposed
Rule 610(d) in light of its subsequent proposed rule addressing volume-
based transaction pricing because the proposals are ``inextricably
linked'' and ``so contradictory and indeterminate that the public has
not had a reasonable opportunity to comment on what the Commission is
actually proposing.'' \715\
---------------------------------------------------------------------------
\715\ Citadel Letter II at 6 (stating that comments on Rule
610(d) were provided ``on the basis that volume-based fee tiers were
explicitly not being prohibited'' and requesting the Commission
``propose and publish an analysis assessing the cumulative effect of
the two proposals that allows commenters to consider the broader
implications (and the Commission's analysis of those implications)
of prohibiting volume-based transaction pricing for certain
orders.'').
---------------------------------------------------------------------------
The Commission disagrees. The Fee Tiers Proposal remains a
proposal. As explained throughout this section, the increased certainty
and transparency Rule 610(d) will require will provide benefits to
investors and other market participants on its own. Further, the two
proposals are not contradictory because Rule 610(d) applies to all
exchange fees and rebates not just those that are volume-based, whereas
the Fee Tiers Proposal specifically concerns volume-based pricing and
agency-related orders as well as a disclosure requirement that would be
fully compatible with Rule 610(d). Accordingly, both proposals are
compatible in their different scopes, objectives, and application and
so are not in conflict. In addition, Rule 610(d) is not indeterminate
but rather straightforward; the public has had the opportunity to
comment and many have, in fact, so commented.
Another commenter stated the Commission should revise proposed Rule
610(d) to require fees and rebates to be known ``before the time of
execution'' and require ``all affected trading venues to publish their
fees in machine-readable format'' to allow market participants to more
readily consume a trading venue's fee schedule and update participant
systems.\716\ Implicit in the requirement that fees/rebates be
determinable at the time of execution is use of historical, rather than
future, volume to benchmark any qualifying criteria for a particular
fee/rebate and thus the fee/rebate would be calculatable or
determinable before the time of execution. For a fee to be determinable
at the time of execution, it must be ascertainable before or
contemporaneously with the execution. Otherwise, the purpose of Rule
610(d)--to provide certainty and transparency regarding what fee/rebate
will apply at the time of execution--would be undermined. No change or
clarification of Rule 610(d) is necessary because implicit in the
requirement that fees/
[[Page 81666]]
rebates be determinable at the time of execution is the ability to
calculate the fee/rebate prior to execution.
---------------------------------------------------------------------------
\716\ FIA PTG Letter II at 4. See also Citadel Letter I at 25;
Healthy Markets Letter I at 26.
---------------------------------------------------------------------------
Finally, new Rule 610(d) enhances transparency regarding exchange
fees and rebates but does not require a specific format for publication
of the information. Exchange fees are required to be posted on exchange
websites \717\ and, as is true today, if market participants need a
specific format, they can make such requests to the exchanges.
---------------------------------------------------------------------------
\717\ See rule 19b-4(m)(1). 17 CFR 240.19b-4(m)(1).
---------------------------------------------------------------------------
V. Final Rule--Transparency of Better Priced Orders
The Commission, among other things, adopted new definitions of
round lot \718\ and odd-lot information \719\ under the MDI Rules to
enhance the transparency for investors and other market participants of
quotes and orders in NMS stocks that have better prices than what has
been provided in SIP data.\720\ In the Proposing Release, the
Commission proposed: (1) to accelerate the implementation of these two
definitions adopted under the MDI Rules, (2) an amendment to the
definition of ``regulatory data'' in Rule 600(b)(78)(iv),\721\ (3) to
require each exclusive SIP to represent quotation sizes in consolidated
information in terms of the number of shares, rounded down to the
nearest multiple of a round lot, and (4) to amend the definition of
odd-lot information to include a best odd-lot order. As discussed in
detail below, the Commission is: (1) adopting an accelerated
implementation schedule for the round lot and odd-lot information
definitions, with modifications to the proposed implementation
schedule; (2) adopting the amendment to the definition of ``regulatory
data'', as proposed; \722\ (3) requiring each exclusive SIP to
represent quotation sizes in consolidated information in terms of the
number of shares, rounded down to the nearest multiple of a round lot,
as proposed; and (4) modifying the Commission's approach in the
Proposing Release by adopting amendments to the round lot definition
that will require less frequent round lot adjustments--i.e.,
semiannually, rather than monthly--by defining a round lot ``Evaluation
Period'' and by specifying an operative period.\723\ In addition, the
Commission is adopting the best odd-lot order data element, as
proposed.
---------------------------------------------------------------------------
\718\ The MDI Rules adopted the definition of round lot in rule
600(b)(82). This provision was subsequently renumbered to Rule
600(b)(93) by the Rule 605 Amendments. 17 CFR 242.600(b)(93); Rule
605 Amendments, supra note 10.
\719\ The MDI Rules adopted the definition of odd-lot
information in rule 600(b)(59). This provision was subsequently
renumbered to Rule 600(b)(69) by the Rule 605 Amendments. 17 CFR
242.600(b)(69); Rule 605 Amendments, supra note 10.
\720\ See MDI Adopting Release, supra note 10.
\721\ See supra note 320.
\722\ The Commission is adopting this amendment to the
definition of regulatory data in Rule 600(b)(89)(iv). See supra note
320.
\723\ The Commission is not changing the calculation used to
assign round lots or the round lot tiers in the round lot definition
adopted in the MDI Rules.
---------------------------------------------------------------------------
A. Background
The MDI Rules expanded NMS information and established a
decentralized consolidation model, pursuant to which competing
consolidators will eventually replace the exclusive SIPs for the
collection, consolidation, and dissemination of NMS information.\724\
The Commission adopted a phased transition plan for the MDI Rules,\725\
which has been delayed.\726\ Accordingly, NMS information is currently
collected, consolidated and disseminated within the national market
system by the exclusive SIPs as SIP data.\727\
---------------------------------------------------------------------------
\724\ See MDI Adopting Release, supra note 10.
\725\ See Proposing Release, supra note 11, at 80295 (describing
the phased transition plan for the MDI Rules).
\726\ See supra notes 74-78 and accompanying text.
\727\ See supra notes 63-65 and accompanying text for a
description of SIP data.
---------------------------------------------------------------------------
Because the MDI Rules are not yet implemented, NMS stock quotation
information that is included in SIP data is provided in round lots, as
defined in exchange rules,\728\ and for most NMS stocks a round lot is
defined as 100 shares.\729\ Under Rule 600(b)(93), as adopted by the
MDI Rules,\730\ round lot sizes are assigned to each NMS stock based on
its average closing price and those NMS stocks that have an average
closing price in the prior month greater than $250.00 will be assigned
a round lot in a size that is less than 100 shares.
---------------------------------------------------------------------------
\728\ See supra note 66.
\729\ See supra note 67.
\730\ See supra note 718.
---------------------------------------------------------------------------
Moreover, because the MDI Rules are not yet implemented,
information about orders in NMS stocks that have a size less than a
round lot, i.e., odd-lot orders, is available on individual exchange
proprietary data feeds, and market participants interested in quotation
information for individual odd-lot orders must purchase these
proprietary feeds.\731\ SIP data includes odd-lot transaction
information but does not include odd-lot quotation information, except
to the extent that odd-lot orders are aggregated into round lots
pursuant to exchange rules.\732\
---------------------------------------------------------------------------
\731\ See supra notes 67-68 and accompanying text.
\732\ See MDI Adopting Release, supra note 10, at 18727.
---------------------------------------------------------------------------
The MDI Rules were designed to increase transparency into, among
other things, the best priced quotations available in the market.\733\
Under the MDI Rules' phased transition plan, the round lot and odd-lot
information definitions were scheduled to be implemented during later
phases in order to avoid imposing costs on the exclusive SIPs, which
will be retired upon full implementation of the MDI Rules.\734\ Due to
the delays in the MDI Rules' implementation, as discussed in the
Proposing Release,\735\ the Commission is adopting an accelerated
implementation schedule, with some modifications from the proposal, so
that market participants, including investors, will be provided with
the enhanced transparency benefits earlier than anticipated in the MDI
Rules.
---------------------------------------------------------------------------
\733\ MDI Adopting Release, supra note 10, at 18601-02, 18617;
see also 17 CFR 242.600(b)(93).
\734\ See MDI Adopting Release, supra note 10, at 18700-01; see
also Proposing Release, supra note 11, at 80295, 80298-99.
\735\ See Proposing Release, supra note 11, at 80295.
---------------------------------------------------------------------------
B. Final Rule--Round Lots
The Commission is amending the implementation schedule for the
round lot definition that was adopted in the MDI Rules. The round lot
definition will be implemented on the first business day of November
2025. This adopted compliance date is modified from the proposal, which
required compliance with the round lot definition 90 days from Federal
Register publication of any Commission adoption of an earlier
implementation of the round lot definition.\736\ The Commission has
provided more time than what was proposed so that market participants
can update and modify their systems.\737\ However, the adopted
compliance date still accelerates the time by which the definition will
be implemented as compared to the preexisting schedule adopted in the
MDI Rules.\738\
---------------------------------------------------------------------------
\736\ See Proposing Release, supra note 11, at 80300-01.
\737\ See infra section VI.C.
\738\ See Proposing Release, supra note 11, at 80295; see also
supra notes 74-78 and accompanying text.
---------------------------------------------------------------------------
In addition, the Commission is amending the round lot definition to
include a new definition for an ``Evaluation Period'' and a provision
specifying the operative dates for round lot assignments. These
amendments will align the dates for assigning round lots
[[Page 81667]]
to the dates for assigning minimum pricing increments under Rule
612.\739\
---------------------------------------------------------------------------
\739\ See supra section III.C.8.
---------------------------------------------------------------------------
Finally, the Commission is also adopting, as proposed, the
amendment to the ``regulatory data'' definition in Rule
600(b)(89)(iv).\740\
---------------------------------------------------------------------------
\740\ See supra note 320.
---------------------------------------------------------------------------
1. Round Lot Definition
Rule 600(b)(93), as adopted by the MDI Rules,\741\ defines a round
lot for NMS stocks that have an average closing price on the primary
listing exchange during the prior calendar month of: (1) $250.00 or
less per share as 100 shares; (2) $250.01 to $1,000.00 per share as 40
shares; (3) $1,000.01 to $10,000.00 per share as 10 shares; and (4)
$10,000.01 or more per share as 1 share.\742\ For any new NMS stock for
which the prior calendar month's average closing price is not
available, a round lot is 100 shares. As a result of the MDI Rules'
round lot definition, each exchange's BBO and the NBBO for an NMS stock
could be based upon smaller, potentially better priced orders,\743\
which would improve transparency regarding the better priced quotations
available in the market and the ability of market participants to
access these quotations.\744\
---------------------------------------------------------------------------
\741\ See supra note 718.
\742\ 17 CFR 242.600(b)(93). The definition of regulatory data
adopted in the MDI Rules also requires that a round lot indicator be
included in NMS information so that market participants will know
the size of a round lot for each NMS stock. The primary listing
exchange must provide, among other things, an ``indicator of the
applicable round lot size'' to competing consolidators and self-
aggregators. 17 CFR 242.600(b)(89); MDI Adopting Release, supra note
10, at 18634. In addition, the MDI Rules require competing
consolidators to represent quotation sizes for certain core data
elements in terms of the number of shares, rounded down to the
nearest multiple of a round lot. 17 CFR 242.600(b)(26)(iii); MDI
Adopting Release, supra note 10, at 18615.
\743\ Orders currently defined as odd-lots often reflect
superior pricing. See MDI Adopting Release, supra note 10, at 18616
n.241 (describing analysis of data from May 2020 that found that
``approximately 45% of all trades executed on exchange and
approximately 10% of all volume executed on exchange in corporate
stocks and ETFs occurred in odd-lot sizes (i.e., less than 100
shares), and 40% of those odd-lot transactions (representing
approximately 35% of all odd-lot volume) occurred at a price better
than the NBBO''). More recent data and updated analyses confirm that
these pricing patterns in odd-lot trading have continued. See
Proposing Release, supra note 11, at 80296.
\744\ See MDI Adopting Release, supra note 10, at 18601, 18615,
18742, 18744-45. In the MDI Proposing Release, the Commission
explained the importance of increasing transparency into odd-lot
quotation information by demonstrating that odd-lot transactions
make up a significant proportion of transaction volume in NMS stocks
(including ETPs), through provision of the daily exchange odd-lot
rate (i.e., the number of exchange odd-lot trades as a proportion of
the number of exchange trades) for corporate stocks and ETPs in 2018
and in June 2019. See Securities Exchange Act Release No. 88216
(Feb. 14, 2020), 85 FR 16726, 16739 (Mar. 24, 2020) (``MDI Proposing
Release''). For this release, Commission staff repeated this
analysis to determine the daily exchange odd-lot rate for 2023.
Based on data from the Commission's MIDAS analytics tool, the daily
exchange odd-lot rate for all corporate stocks ranged from
approximately 61% to 70% of trades and the daily exchange odd-lot
rate for all ETPs ranged from 31% to 42% of trades in 2023.
Accordingly, accelerating the implementation of the round lot and
odd-lot information definitions will increase the pre-trade
transparency of better priced orders that are prevalent in the
national market system.
---------------------------------------------------------------------------
In the MDI Adopting Release, the Commission analyzed data from May
2020 on the portion of all corporate stock and ETF volume executed on
an exchange, transacted in a quantity less than 100 shares, at a price
better than the prevailing NBBO, occurring in a quantity that would be
defined as a round lot under the MDI Rules.\745\ The Proposing Release
repeated this analysis using data for the dates March 25-31, 2022.\746\
Both analyses demonstrated that the round lot definition adopted in the
MDI Rules will capture significant percentages of better priced odd-lot
orders for NMS stocks with an average closing price greater than
$250.00.
---------------------------------------------------------------------------
\745\ See MDI Adopting Release, supra note 10, at 18612 (table
1).
\746\ See Proposing Release, supra note 11, at 80296-97 (tables
1 and 2).
---------------------------------------------------------------------------
The Commission has updated this analysis from the Proposing Release
with data from October 2023. As discussed below, upon further
consideration and evaluation of comments, the Commission is adopting
modifications to the round lot definition to require less frequent
round lot adjustments so that they occur on a semiannual basis, rather
than on a monthly basis and the calculation of the average closing
price on the primary listing exchange will be based on a one-month
``Evaluation Period.'' \747\ The updated analysis accounts for the
modifications to the round lot definition and is based on data from
October 23-27, 2023, using a March 2023 Evaluation Period to be
consistent with the adopted rule.\748\ The updated analysis
demonstrates that the round lot definition, as amended, will capture
significant percentages of better priced odd-lot orders.
---------------------------------------------------------------------------
\747\ See infra section V.B.3.b.iv. Amended Rule 600(b)(93)(iii)
defines the Evaluation Period as (A) all trading days in Mar. for
the round lot assigned on the first business day in May and (B) all
trading days in Sept. for the round lot assigned on the first
business day of Nov. during which the average closing price of an
NMS stock on the primary listing exchange shall be measured by the
primary listing exchange to determine the round lot for each NMS
stock.
\748\ See infra section V.B.3.b.iv. The analysis used the
average closing price of the NMS stocks on their primary listing
exchange for all trading days in Mar. 2023 and used those average
prices to determine the size of the round lot for each stock in the
universe. Those round lots were then applied to the analysis of the
stocks' trading data for Oct. 23-27, 2023.
---------------------------------------------------------------------------
Tables 1 and 2 examine the portion of all corporate stock and ETP
share volume and trades executed on an exchange, transacted in a
quantity less than 100 shares, at a price better than the prevailing
NBBO, occurring in a quantity defined as a round lot under the MDI
Rules, as amended by the Commission.
Table 1
------------------------------------------------------------------------
Round lot size
Round lot tier (shares) Percent \1\
------------------------------------------------------------------------
$0-$250.00.............................. 100 0.00
$250.01-$1,000.00....................... 40 52.89
$1,000.01-$10,000.00.................... 10 76.89
$10,000.01 or more...................... 1 100.00
------------------------------------------------------------------------
\1\ Portion of all corporate stock and ETP share volume executed on an
exchange, transacted in a quantity less than 100 shares, at a price
better than the prevailing NBBO, occurring in a quantity that would be
defined as a round lot under the MDI Rules as amended, for Oct. 23-27,
2023.
Source: Equity consolidated data feeds (CTS and UTDF), as collected by
MIDAS; NYSE Daily TAQ.
[[Page 81668]]
Table 2
------------------------------------------------------------------------
Round lot size
Round lot tier (shares) Percent \1\
------------------------------------------------------------------------
$0-$250.00.............................. 100 0.00
$250.01-$1,000.00....................... 40 13.79
$1,000.01-$10,000.00.................... 10 26.63
------------------------------------------------------------------------
$10,000.01 or more...................... 1 100.00
------------------------------------------------------------------------
\1\ Portion of all corporate stock and ETP trades executed on an
exchange, transacted in a quantity less than 100 shares, at a price
better than the prevailing NBBO, occurring in a quantity that would be
defined as a round lot under the MDI Rules as amended, for Oct. 23-27,
2023.
Source: Equity consolidated data feeds (CTS and UTDF), as collected by
MIDAS; NYSE Daily TAQ.
The Proposing Release also included the results of a simulation
conducted by the Commission, using exchange direct feed data from MIDAS
for every trading day in March 2022, to create a mockup competing
consolidator feed that included quotation information for a sample of
NMS stocks priced at or over $250.01 using the priced-based round lot
sizes adopted in the MDI Rules' round lot definition as opposed to the
round lot sizes that are defined in current exchange rules (typically
100 shares).\749\ Snapshots of this simulated feed were compared
against snapshots of the exclusive SIP feed for the sample of NMS
stocks at the same point in time. For two of the three round lot price
tiers above $250.01, the simulated competing consolidator feed showed
better prices, on average, than the exclusive SIP feed.\750\
---------------------------------------------------------------------------
\749\ See supra note 66.
\750\ See Proposing Release, supra note 11, at 80297 (stating,
``[f]or stocks priced between $250.01 and $1,000.00 per share, which
will have a round lot size of 40 under the round lot definition, the
price reflected in the simulated competing consolidator feed was
better than the exclusive SIP feed 21.47% of the time and worse less
than .1% of the time. For stocks priced between $1,000.01 and
$10,000.00 per share, which will have a round lot size of 10 under
the round lot definition, the price reflected in the simulated
competing consolidator feed was better than the exclusive SIP feed
64.67% of the time and worse less than .1% of the time.''). For the
third round lot price tier above $250.01, for stocks priced
$10,000.01 or more, the Proposing Release stated that there was one
stock that was priced over $10,000 per share and was already quoted
in one-share round lots on the exclusive SIP feed; therefore, the
simulated feed and exclusive SIP feed showed the same prices for the
stock. Id. at 80297 n.369.
---------------------------------------------------------------------------
The Commission has updated this analysis using exchange direct feed
data from MIDAS for every trading day in November 2023 and to account
for the modifications to the round lot definition.\751\ Like the prior
analysis, the Commission conducted a simulation of a competing
consolidator feed that provides quotation information for a sample of
NMS stocks priced at or over $250.01 using the priced-based round lot
sizes adopted in the MDI Rules' round lot definition as opposed to the
round lot sizes that are defined in current exchange rules (typically
100 shares).\752\ Snapshots of this simulated feed were compared
against snapshots of the exclusive SIP feed for that NMS stock at the
same point in time. For two of the three price tiers and corresponding
round lot sizes, the simulated feed showed better prices, on average,
than the exclusive SIP feed.\753\ For stocks priced between $250.01 and
$1,000.00 per share, which will have a round lot size of 40 under the
round lot definition, the price reflected in the simulated competing
consolidator feed was better than the exclusive SIP feed 31.93% of the
time and worse less than .1% of the time. For stocks priced between
$1,000.01 and $10,000.00 per share, which will have a round lot size of
10 under the round lot definition, the price reflected in the simulated
competing consolidator feed was better than the exclusive SIP feed
80.77% of the time and worse less than .2% of the time. The updated
analysis continues to demonstrate that the simulated competing
consolidator feed, which reflects the round lot sizes adopted in the
MDI Rules' round lot definition, provides better prices than the
exclusive SIP feeds, which reflect the prior round lot size, in NMS
stocks priced over $250.01, even with the modifications to the round
lot definition.\754\
---------------------------------------------------------------------------
\751\ The Commission assigned a sample of NMS stocks to a round
lot tier based upon their average closing prices on the primary
listing exchange during Sept. 2023 to account for the adopted
definition of Evaluation Period under the round lot definition,
which states that the Evaluation Period for the round lot assigned
on the first business day of Nov. would be all trading days in Sept.
See amended Rule 600(b)(93)(iii).
\752\ See supra note 66.
\753\ In the third price tier, which defines a round lot for
stocks priced $10,000.01 or more per share as an order for the
purchase or sale of an NMS stock of one share, only one stock, which
is already quoted in one share round lot on the exclusive SIP feed,
was priced over $10,000 per share, so the simulated feed and
exclusive SIP feed showed the same prices for this stock.
\754\ See supra note 750.
---------------------------------------------------------------------------
2. Proposed Acceleration of Round Lot Definition
The Commission proposed to accelerate the implementation of the
round lot definition set forth in Rule 600(b)(93).\755\ Specifically,
the Commission proposed to require compliance with the round lot
definition 90 days from Federal Register publication of any Commission
adoption of an earlier implementation of the round lot definition.\756\
---------------------------------------------------------------------------
\755\ See supra note 718.
\756\ See Proposing Release, supra note 11, at 80300.
---------------------------------------------------------------------------
In the MDI Adopting Release, the Commission stated that
``sequencing [round lot implementation] after the parallel operation
period is important to avoid either: (1) potential confusion and market
disruption that could result from two different round lot structures
operating at the same time; or (2) imposing reprogramming costs on the
exclusive SIPs for a limited time period prior to their retirement.''
\757\ However, because full implementation of the MDI Rules as adopted
pursuant to the phased transition plan \758\ likely will not occur
until at least two years after new proposals to amend the effective
national market system plan(s) are developed, filed and approved by the
Commission,\759\ the Commission
[[Page 81669]]
proposed to amend the phased transition schedule of the MDI Rules to
allow the benefits of the round lot definition to be made available to
investors sooner.\760\ The benefits identified in the MDI Adopting
Release justify the costs of accelerating the implementation of the
round lot definition in this rulemaking.\761\
---------------------------------------------------------------------------
\757\ MDI Adopting Release, supra note 10, at 18701. The
Commission stated that ``the consolidated market data products
offered by competing consolidators during the initial parallel
operation period would be based on the current definition of round
lot.'' Id. at 18700. However, because the Commission is accelerating
the implementation of the round lot definition, the exclusive SIPs
will be providing SIP data that reflects the new round lot sizes
during the initial parallel operation period. Further, the
acceleration of the implementation of the round lot definition will
result in its use during the parallel operation period by both the
exclusive SIPs and competing consolidators. See supra note 73 for a
discussion of the parallel operation period; infra section VI.C for
a discussion of the modified compliance deadline.
\758\ See supra notes 73-78 and accompanying text.
\759\ See Proposing Release, supra note 11, at 80295. See also
MDI Adopting Release, supra note 10, at 18699-701. The two-year
estimated timeframe includes the implementation of the round lot
definition, which was scheduled to occur at the end of the
transition plan.
\760\ See Proposing Release, supra note 11, at 80300-01.
\761\ See infra section VII.D.5.
---------------------------------------------------------------------------
Further, as part of accelerating the implementation of the round
lot definition, the Commission proposed to amend the definition of
``regulatory data'' in Rule 600(b)(78) to require the indicator of the
applicable round lot size to be provided to the exclusive SIPs for
collection and dissemination.\762\ The preexisting definition of
``regulatory data'' required the primary listing exchange for an NMS
stock to provide to competing consolidators and self-aggregators an
indicator of applicable round lot size.\763\ The Commission proposed to
add new paragraph (iv) to the definition of ``regulatory data'' to
require the primary listing exchanges to also make the indicator
available to the exclusive SIPs.\764\ Referencing the round lot
indicator adopted with regard to competing consolidators in the MDI
Rules, the Commission stated that such an indicator will ``help market
participants ascertain the applicable round lot size for each NMS stock
on an ongoing basis'' \765\ and ``reduce confusion as market
participants adjust to the new round lot sizes.'' \766\ For these same
reasons, the Commission proposed to require this indicator to be
provided to the exclusive SIPs for collection and dissemination.\767\
---------------------------------------------------------------------------
\762\ Under the MDI Rules, the definition of regulatory data
requires the primary listing exchange to make an indicator of the
applicable round lot size to competing consolidators and self-
aggregators. See Rule 600(b)(89)(i)(E), 17 CFR 242.600(b)(89)(i)(E).
See also supra note 320.
\763\ 17 CFR 242.600(b)(89)(i)(E).
\764\ See Proposing Release, supra note 11, at 80299.
\765\ See Proposing Release, supra note 11, at 80299. For more
details, see MDI Proposing Release, supra note 744, at 16762.
\766\ See Proposing Release, supra note 11, at 80299; MDI
Adopting Release, supra note 10, at 18619.
\767\ See Proposing Release, supra note 11, at 80299. As
discussed below, since the MDI Rules already require the primary
listing exchanges to provide an indicator of the applicable round
lot size to competing consolidators and self-aggregators, the
incremental cost of providing this indicator to the two exclusive
SIPs should be low. See infra section VIII.G.
---------------------------------------------------------------------------
3. Comments and Response
a. Comments Supporting the Proposed Change
The Commission received comments in support of the proposed
acceleration of the implementation of the round lot definition from
individuals,\768\ firms,\769\ exchanges,\770\ and associations.\771\
Several commenters supported the proposed acceleration of the
implementation of the round lot definition because they said that it
would improve transparency \772\ and enhance price discovery.\773\
Several commenters stated that the proposed change would result in more
accurate prices,\774\ allow investors to make more informed trading
decisions,\775\ improve execution quality,\776\ and reduce transaction
costs and inefficiencies.\777\
---------------------------------------------------------------------------
\768\ See Comment Letters Type E, F, G, H, I, J, K, available at
https://www.sec.gov/comments/s7-30-22/s73022.htm; see, e.g., Letters
from Aron Tastensen (Feb. 23, 2023); Aswin Joy (Mar. 7, 2023);
Abraham (Mar. 14, 2023); Andrew A. (Mar. 19, 2023).
\769\ See, e.g., XTX Letter at 5; BMO Letter at 2; Schwab Letter
II at 36; Citigroup Letter at 3; Hudson River Letter at 2; Fidelity
Letter at 9, 16; BlackRock Letter at 11-12; Vanguard Letter at 2.
\770\ See, e.g., NYSE Letter I at 7; Nasdaq Letter I at 3; MEMX
Letter at 2, 4, 5-6; Cboe Letter II at 2, 10; IEX Letter I at 6, 30;
Cboe Letter III at 10 and n.18. The Commission also received comment
letters submitted by both exchanges and firms that supported the
accelerated implementation of the round lot definition. See Cboe,
State Street, et al. Letter at 2; NYSE, Schwab, and Citadel Letter
at 2.
\771\ See, e.g., MFA Letter at 3, 13-14; Better Markets Letter I
at 16-17; SIFMA AMG Letter I at 9; CCMR Letter at 22; FIA PTG Letter
II at 4-5; ICI Letter I at 6-7; SIFMA Letter II at 34, 44; STA
Letter at 8. See also AIMA Letter at 3 (stating that the Commission
should prioritize the implementation of the round lot definition and
the Rule 605 Proposal).
\772\ See Comment Letter Type E, I, available athttps://www.sec.gov/comments/s7-30-22/s73022.htm; see, e.g., Letters from
Bill Gilbert (Mar. 7, 2023); Richard Pasquali (Mar. 16, 2023); IEX
Letter I at 6; MFA Letter at 13-14; XTX Letter at 5; Better Markets
Letter I at 16-17; BlackRock Letter at 11-12 (stating that
accelerated implementation of the round lot and the odd-lot
information definitions would increase pre-trade transparency for
investors); FIA PTG Letter II at 4; Hudson River Letter at 2; ICI
Letter I at 6.
\773\ See, e.g., XTX Letter at 5 (also referring to the proposed
publication on the exclusive SIPs of odd-lots priced better than the
NBBO); Cboe Letter II at 10, Cboe Letter I at 10 (stating that the
proposed acceleration of the round lot definition and the display of
odd-lot orders would improve price discovery and reduce spreads);
ICI Letter I at 6.
\774\ See, e.g., IEX Letter I at 6 (referring to the proposed
acceleration of the implementation of the round lot definition as
well as the display of odd-lot orders).
\775\ See, e.g., MFA Letter at 13-14; Better Markets Letter I at
16-17; FIA PTG Letter II at 4.
\776\ See, e.g., MFA Letter at 13-14; Better Markets Letter I at
16-17; BlackRock Letter at 11 (stating that accelerated
implementation of the round lot and the odd-lot information
definitions would result in enhanced execution quality for
investors).
\777\ See FIA PTG Letter II at 4.
---------------------------------------------------------------------------
One commenter supported the proposed acceleration of the
implementation of the round lot definition because the commenter said
it would narrow the NBBO spread by incorporating current odd-lot
interest and ``mak[ing] the notional size associated with the NBBO more
uniform across stock price levels.'' \778\ Another commenter supported
the proposal because the round lot definition ``would enhance the
accuracy of the NBBO for high-priced stocks.'' \779\ Some commenters
supported the proposed acceleration of the implementation of the round
lot definition because they stated that this change would restore
public trust,\780\ or because this change and the dissemination of odd-
lot information by the exclusive SIPs would enhance reporting
efficiency and reduce delays.\781\ Commenters also supported the
proposed acceleration of the implementation of the round lot definition
because they stated that the round lot definition would result in lot
sizes that would better suit the needs of investors.\782\
---------------------------------------------------------------------------
\778\ See Hudson River Letter at 2.
\779\ See CCMR Letter at 22.
\780\ See, e.g., Letters Type E, G, J, available at https://www.sec.gov/comments/s7-30-22/s73022.htm; see also e.g., Letters
from Christopher Nieto (Mar. 31, 2023); Michael Montalban (Mar. 31,
2023).
\781\ See, e.g., Letter Type K, available at https://www.sec.gov/comments/s7-30-22/s73022.htm.
\782\ See, e.g., IEX Letter I at 30; BlackRock Letter at 11
(stating that the proposed acceleration of the implementation of
both the round lot and odd-lot definitions would increase the
usefulness of the exclusive SIPs because of the prevalence of
current odd-lot sizes); MEMX Letter at 4; STA Letter at 8.
---------------------------------------------------------------------------
In the MDI Adopting Release, as well as the Proposing Release, the
Commission described the benefits of the adopted round lot
definition.\783\ The Commission stated that the new round lot
definition will ``narrow NBBO spreads for most stocks with prices
greater than $250,'' \784\ improve transparency and ``the
comprehensiveness of and usability of core data, facilitate the best
execution of customer orders, and reduce information asymmetries.''
\785\ The Commission also stated that the reduced round lot size for
high priced NMS stocks would ``better ensure the display and
accessibility of significant liquidity for high-priced stocks.'' \786\
Some commenters supported the proposed acceleration of the
implementation of the round lot definition but stated that more time
than proposed was needed
[[Page 81670]]
for compliance.\787\ As discussed later in this release, the Commission
is providing more time to implement the round lot definition than the
90-days that was proposed for implementation.\788\
---------------------------------------------------------------------------
\783\ See Proposing Release, supra note 11, at 80296.
\784\ Id.
\785\ Id.
\786\ Id.
\787\ See, e.g., NYSE Letter I at 7; Nasdaq Letter I at 3;
Fidelity Letter at 16; Cboe Letter III at 10 n.18.
\788\ See infra section VI.C.
---------------------------------------------------------------------------
One commenter supported the proposed acceleration of the
implementation of the round lot definition subject to ``regulatory and
industry-wide education to investors on the changes.'' \789\ As with
many regulatory changes, investor education and notification may be
useful so that investors better understand the implications of the size
of their orders.
---------------------------------------------------------------------------
\789\ See Fidelity Letter at 16.
---------------------------------------------------------------------------
b. Comments Objecting to the Proposed Change
The Commission also received comments that raised objections to
specific aspects of the proposed acceleration of the implementation of
the round lot definition. These comments are addressed below.
i. Comments on the Interaction Between the Round Lot Definition and the
Proposed Minimum Pricing Increments
The Commission received comments that expressed concern about the
potential impact of both the implementation of the round lot definition
and the proposed changes to the minimum pricing increments.\790\
Specifically, some commenters raised concerns about the potential
impact of both proposed changes on liquidity and NBBO depth.\791\ One
commenter stated that smaller round lot sizes would make the NBBO
``less robust, as a smaller amount of liquidity would now establish the
NBBO benchmark,'' compounded by the proposed reduction in quoting tick
sizes that would require liquidity to be dispersed in finer pricing
increments.\792\ One commenter stated that the proposed minimum pricing
increments and ``the round lot reforms'' would be duplicative because
they would both result in narrow spreads.\793\ In addition, one
commenter stated that the round lot definition and proposed minimum
pricing increments ``would significantly reduce transparency on the SIP
and force more participants to purchase costly direct feeds to maintain
the same level of transparency of liquidity.'' \794\
---------------------------------------------------------------------------
\790\ See, e.g., Virtu Letter II at 10; Citadel Letter I at 26;
ASA Letter at 6; Tastytrade Letter at 22; SIFMA Letter II at 34;
Morgan Stanley Letter at 3. See also Nasdaq Letter I at 34 (stating
that ``an effective tick reform proposal may alleviate the need to
speed implementation of the round and odd lot proposals.'').
\791\ See, e.g., Virtu Letter II at 10; Citadel Letter I at 26.
\792\ See Virtu Letter II at 10.
\793\ See ASA Letter at 6.
\794\ See Virtu Letter II at 10.
---------------------------------------------------------------------------
Although the Commission agrees that both the round lot definition
adopted in the MDI Adopting Release and the amended minimum pricing
increments will impact the NBBO and will result in a narrower spread
for impacted NMS stocks, the NMS stocks that would be subject to both
the round lot definition and the amended minimum pricing increments are
likely to be very small in number and also extremely liquid, which
could counteract any potential harm to liquidity resulting from the
interaction of both changes. The round lot definition adopted in the
MDI Adopting Release and the amended minimum pricing increments each
will impact the NBBO and each will result in a narrower spread for
those NMS stocks that are assigned a smaller round lot \795\ or a
smaller minimum pricing increment.\796\
---------------------------------------------------------------------------
\795\ See infra section VII.D.4.a.
\796\ See infra section VII.B.2; section VII.D.1.
---------------------------------------------------------------------------
The round lot definition will narrow the spread for NMS stocks that
have an average closing price over $250 per share by showing better
prices for these stocks. The amended minimum pricing increments will
reduce the spread for those NMS stocks that have a narrow TWAQS and
will allow these stocks to be priced more competitively in smaller
increments, which will more accurately reflect supply and demand.
Accordingly, the smaller round lot and the smaller minimum pricing
increment narrow spreads in different ways. In response to the comment
stating that the round lot definition and the proposed tick size
changes are duplicative because they would both result in narrow
spreads,\797\ both requirements will narrow spreads, but they are not
duplicative.
---------------------------------------------------------------------------
\797\ See ASA Letter at 6.
---------------------------------------------------------------------------
Although there may be NMS stocks that are assigned both a smaller
round lot and a smaller minimum pricing increment, Commission analysis
of data discussed below shows that this overlapping universe of NMS
stocks is very small. In other words, most NMS stocks will not be
assigned both a round lot that is less than 100 shares and a smaller
$0.005 minimum pricing increment and therefore will not experience a
combined impact on the NBBO spread or depth. As explained in the
analysis, since only a few NMS stocks are expected to be subject to
both a smaller round lot and a smaller tick size, the potential
combined impact of the amendments to the minimum pricing increments and
round lots should be limited.\798\ Specifically, in response to
comments expressing concerns about the combined impact of the proposed
smaller minimum pricing increments and the implementation of the round
lot definition,\799\ the Commission conducted the analysis to determine
the magnitude of NMS stocks that would be impacted by both changes.
According to the Commission's analysis, as of November 30, 2023, only
163 NMS stocks were priced above $250.00 per share and would have been
potentially eligible to be assigned to a round lot size smaller than
100 shares, out of a universe of 11,200 NMS stocks on that date.\800\
Further, based on Commission review of the 163 NMS stocks that would
have been assigned to a round lot less than 100 shares, as of November
30, 2023, only two out of the 163 had an average quoted spread over the
previous thirty trading days of $0.015 or less and therefore may have
been potentially eligible to have been assigned to the smaller $0.005
minimum pricing increment.\801\ The two NMS stocks--SPY and QQQ--are
among the most liquid exchange-traded products.\802\ For the month of
November 2023, SPY had the highest average daily traded value, while
QQQ was ranked second in average daily traded value.\803\
---------------------------------------------------------------------------
\798\ See infra section VII.D.4.a. (discussing the impact of the
acceleration of the implementation of the round lot definition and
reduced tick sizes and stating that the number of stocks trading
over $250 with spreads narrower than $0.015 is ``likely very
small'').
\799\ See, e.g., Virtu Letter II at 10; Citadel Letter I at 26;
ASA Letter at 6.
\800\ The Proposing Release stated that, based on average
closing prices on the primary listing exchange in Mar. 2022, there
were 181 NMS stocks priced over $250. See Proposing Release, supra
note 11, at 80300 n.407.
\801\ These NMS stocks were ETFs: SPY and QQQ. As of Nov. 30,
2023, the last sale price for SPY was $454.30 and its average bid-
ask spread over the previous 30 trading days was $0.0105. For QQQ,
as of Nov. 30, 2023, the last sale price was $386.70 and the average
bid-ask spread over the previous 30 trading days was $0.0116. This
analysis was conducted using Bloomberg data. See also supra section
III for a discussion of the amended minimum pricing increments. The
calculation of a TWAQS for NMS stocks will occur during an
Evaluation Period for purposes of assigning minimum pricing
increments. See Rule 612(a)(1).
\802\ SPY and QQQ are also among the most liquid NMS stocks.
Using Bloomberg data, for the period Jan. 22, 2024-Feb. 16, 2024,
SPY had the highest average daily traded value of all NMS stocks,
while QQQ was ranked fourth. Specifically, for this period, the
average daily traded value per day for SPY was $36,581,363,712, or
5.9% of total value traded of all U.S. equity trading, and for QQQ
the average daily traded value per day was $18,632,960,000, or 3.5%
of total value traded of all U.S. equity trading.
\803\ Based on daily average traded value for Nov. 1, 2023-Nov.
30, 2023, using Bloomberg data. For SPY, the daily average traded
value for Nov. 2023 was $31,652,396,337. For QQQ, it was
$17,527,314,000.
---------------------------------------------------------------------------
[[Page 81671]]
Based on this information, while the Commission recognizes that the
interaction of the minimum pricing increment and the round lot
definition may result in some reduction of depth for the very few NMS
stocks that may be subject to both a smaller tick size and a round lot
size of less than 100 shares,\804\ the impact of the reduction of NBBO
depth should not be of such a level as to impede trading in the
affected NMS stocks because these stocks are highly liquid, which
should greatly mitigate the impact of reduced depth at the NBBO.\805\
Furthermore, these changes will benefit investors and other market
participants trading these stocks through more accurate pricing and a
reduction in spreads. While the extent of any reduced depth at the NBBO
is not known at this time, due to the volume traded in these two NMS
stocks, any potential reduction will not impair market participants'
ability to trade these stocks because these stocks would be among the
most liquid and therefore easily traded.\806\
---------------------------------------------------------------------------
\804\ See infra section VII.D.4.a.
\805\ See infra section VII.D.4.a. (stating, ``[t]he exceptional
liquidity of the affected stocks will likely protect their NBBO from
material deterioration.'').
\806\ Id.
---------------------------------------------------------------------------
Additionally, some liquidity that is consolidated at the
preexisting round lot \807\ and $0.01 minimum pricing increment may be
reflected in the adopted round lot sizes and smaller minimum pricing
increment. Specifically, some odd-lot orders are currently aggregated
into the 100-share round lot.\808\ Upon implementation of the round lot
definition, some orders that were considered odd-lots may be of round
lot size as defined. Further, interest that is displayed at the
previously required $0.01 minimum pricing increment may be reflected in
orders that are entered in the smaller tick size. Once these amendments
are implemented, the NBBO will reflect better prices, both because of
the smaller round lot size for some NMS stocks and new $0.005 increment
for some other NMS stocks. As smaller sized orders in higher priced
stocks are often priced better than orders that are currently in round
lots, the smaller round lot sizes will allow potentially better priced
orders to be the basis of the NBBO.\809\ The new $0.005 increment will
also result in the NBBO reflecting better prices because the smaller
increment will allow orders to be priced in a manner that is more
reflective of the supply and demand of liquidity for the stock.\810\
Accordingly, each of these amendments will result in narrower NBBO
spreads and better prices.\811\ Further, those market participants that
may need to trade in large sizes may be able to see liquidity outside
of the NBBO by considering the new odd-lot information that will be
available in SIP data as well as depth of book data that is available
via exchange proprietary data feeds.\812\
---------------------------------------------------------------------------
\807\ See supra note 66.
\808\ See MDI Proposing Release, supra note 744, at 16738-39
(describing exchange rules on aggregating odd-lot across multiple
prices and providing them to the exclusive SIPs at the least
aggressive price if the combined odd-lot interest is equal to or
greater than a round lot).
\809\ See Proposing Release, supra note 11, at 80294; MDI
Adopting Release, supra note 10, at 18616.
\810\ See infra section VII.D.1; section VII.D.1.b.i; section
VII.D.1.b.ii; section VII.E.3.
\811\ See infra section VII.D.1 and note 1145 and accompanying
text.
\812\ Once implemented, the MDI Rules will add depth of book
information to consolidated market data, and this information will
provide information about depth outside of the NBBO for those market
participants that would find this information useful. See infra
section VII.D.4.a.; MDI Adopting Release, supra note 10, at 18728,
18730 (explaining that SIP data currently only includes top-of-book
quotes). 17 CFR 242.600(b)(26)(I) (defining ``core data'' to include
depth of book data) and 17 CFR 242.600(b)(24)(i) (defining
``consolidated market data'' to include core data).
---------------------------------------------------------------------------
In response to the comment that raised concerns about the potential
for the proposed variable minimum pricing increments and the new round
lots to reduce transparency on the exclusive SIPs,\813\ for the reasons
discussed above,\814\ the combined impact of the adopted minimum
pricing increments and round lot definition should not reduce
transparency for most NMS stocks and the exclusive SIPs will provide a
more accurate NBBO once these amendments are implemented. While the
round lot will be smaller for certain NMS stocks, as described above,
the NBBO based on the new round lots will in many cases reflect better
prices.\815\ Therefore, while the actual number of shares will be
smaller for certain NMS stocks, the disseminated prices will likely be
better.\816\ In addition, as discussed above, the new minimum pricing
increment required under Rule 612 for certain NMS stocks will allow the
NBBO that is disseminated by the exclusive SIPs to reflect more
competitive pricing. These amendments will enhance the NBBO that is
calculated and disseminated by the exclusive SIPs by reflecting more
competitive and better available prices.\817\
---------------------------------------------------------------------------
\813\ See Virtu Letter II at 10.
\814\ See supra notes 807-812 and accompanying text.
\815\ See supra section V.B.1. (table 1 and table 2).
\816\ See MDI Adopting Release, supra note 10, at 18742, 18743.
\817\ See supra section III.C. See also MDI Adopting Release,
supra note 10, at 18744, 18745. See also infra section VII.D.4.a.
(stating that the round lot definition would shrink the NBBO for
stocks priced greater than $250, would increase transparency and
would result in better order execution).
---------------------------------------------------------------------------
Commenters also expressed concern that the implementation of the
round lot definition and the proposed changes to the minimum pricing
increments would confuse investors.\818\ One commenter warned that
changes to round lots and tick sizes would confuse retail investors and
reduce trust in the market.\819\
---------------------------------------------------------------------------
\818\ See, e.g., Tastytrade Letter at 22; SIFMA Letter II at 34;
Morgan Stanley Letter at 3.
\819\ See, e.g., Tastytrade Letter at 22.
---------------------------------------------------------------------------
Because there are expected to be only a small number of NMS stocks
that could be subject to both a change in a minimum pricing increment
and a change to the round lot size, the risk of investor confusion is
limited. Market participants may choose to educate investors about the
new round lot and amended minimum pricing increments, as they sometimes
choose to educate investors regarding other regulatory changes that
impact how investors enter orders. Investors are already familiar with
three round lot sizes \820\ and two minimum pricing increments,\821\ so
the addition of only one round lot size and one minimum pricing
increment is unlikely cause investor confusion.\822\ The Commission is
also adopting new indicators for dissemination on the exclusive SIPs of
the assigned round lots and minimum pricing increments to alert market
participants, including investors, of the relevant round lot and
minimum pricing increment for each NMS stock. These indicators will
also help to mitigate concerns about any potential for investor
confusion.\823\
---------------------------------------------------------------------------
\820\ Under exchange rules, there are three different round lot
sizes. See supra note 67. The MDI Rules' round lot definition adds
one more round lot size, i.e., 40 shares. Consistent with its views
stated here, the Commission previously considered potential investor
confusion with the additional round lot size and did not believe it
will be confusing to investors. See MDI Adopting Release, supra note
10, at 18618.
\821\ Preexisting Rule 612 included two minimum pricing
increments based on the price of a quote or order--$0.01 and
$0.0001.
\822\ See supra section III. See also infra notes 1589-1593 and
accompanying text.
\823\ See infra note 1385; MDI Adopting Release, supra note 10,
at 18619.
---------------------------------------------------------------------------
ii. Comments on the Round Lot Indicator
The Commission proposed to amend Rule 600(b)(78) to add a
requirement to make the indicator of the applicable round lot size
available to the exclusive
[[Page 81672]]
SIPs in Rule 600(b)(78)(iv).\824\ The Commission is adopting this
requirement as proposed in Rule 600(b)(89)(iv).\825\ The Commission did
not receive comments specifically supporting or objecting to the
proposed amendment. However, two commenters cited this requirement as
support for their arguments that the proposed 90-day compliance
deadline for the round lot and odd-lot information definitions would
provide an insufficient amount of time.\826\ As discussed below,\827\
the Commission is providing more time for compliance with the round lot
definition, which is a substantially longer period for compliance than
the 90 days that was proposed.\828\ The Commission is providing a
longer compliance period than proposed after considering the
information provided by commenters that requested more time to comply
with the implementation of the round lot definition, including
implementation of the round lot indicator.\829\ The additional time
will provide market participants with time to make the changes
necessary to implement the round lot indicator.
---------------------------------------------------------------------------
\824\ See Proposing Release, supra note 11, at 80299.
\825\ See supra note 320.
\826\ See NYSE Letter I at 7-8; Letter from Robert Books, Chair
of the Operating Committee, Operating Committees of the CTA Plan, CQ
Plan and UTP Plan, dated Mar. 28, 2023 (``CTA, CQ, UTP Plans
Operating Committees Letter'') at 3. See also infra section VI.C.
\827\ See infra section VI.C.
\828\ See Proposing Release, supra note 11, at 80300-01.
\829\ See supra note 826; infra section VI.C.
---------------------------------------------------------------------------
iii. Comments on the Round Lot Definition
The Commission received comments on the round lot definition that
was adopted in the MDI Rules.\830\ Commenters raised concerns about the
defined round lot sizes,\831\ the determination of round lot size based
on price,\832\ the impact of smaller round lot sizes on the relevance
of the NBBO,\833\ and the impact of the round lot definition as well as
the odd-lot information definition on bandwidth.\834\ The Commission
considered and addressed issues related to adopting the round lot
definition and the odd-lot information definition in the MDI Adopting
Release.\835\
---------------------------------------------------------------------------
\830\ See, e.g., RBC Letter at 5; Tastytrade Letter at 21; T.
Rowe Price Letter at 4; Pragma Letter at 8-9; Data Boiler Letter I
at 8 and Data Boiler Letter II at 2.
\831\ See, e.g., Pragma Letter at 1, 8, 9; T. Rowe Price Letter
at 4.
\832\ See, e.g., Tastytrade Letter at 21.
\833\ See, e.g., RBC Letter at 5.
\834\ See Data Boiler Letter I at 8; see also Data Boiler Letter
II at 2. The commenter raised concerns about the round lot and odd-
lot information definitions.
\835\ See MDI Adopting Release, supra note 10, at 18615-22.
---------------------------------------------------------------------------
One commenter stated that investors trading in options may be
confused by the round lot definition, stating that retail investors who
trade options know one options contract represents 100 shares or a
``round lot.'' \836\ The Commission considered and addressed the
interaction of the new round lot definition and options trading in the
MDI Adopting Release.\837\ Further, it is unlikely that the
acceleration of the round lot definition could confuse retail investors
trading in options. Specifically, the round lot size will not change
the size of the options contract and precedent exists for standard
options contracts on stocks with a round lot size less than 100
shares.\838\ Furthermore, corporate actions, such as rights offerings,
stock dividends, and mergers can result in adjusted contracts
representing stock in amounts other than 100 shares, so investors have
some familiarity already with options on underlying NMS stocks that
have a ``round lot'' that is less than 100 shares.\839\
---------------------------------------------------------------------------
\836\ See Tastytrade Letter at 22.
\837\ See MDI Adopting Release, supra note 10, at 18619, 18747.
\838\ For example, as of Oct. 26, 2023, one NMS stock has a
round lot size of 10 shares while also possessing an option contract
size of 100 shares.
\839\ See MDI Adopting Release, supra note 10, at 18619. See
also infra section VII.D.4.a.
---------------------------------------------------------------------------
Several commenters also suggested eliminating the concept of round
lots altogether.\840\ The Commission is not eliminating the concept of
round lots. As the Commission stated in the MDI Adopting Release, round
lot orders continue to play an important role in the national market
system by delineating orders of meaningful size and focusing regulatory
requirements and protections--such as those set forth in rules 602, 604
and 611 of Regulation NMS--on such orders.\841\ Further, as the
Commission stated in the MDI Adopting Release, eliminating the concept
of a round lot could also cause investor confusion and other unintended
consequences.\842\
---------------------------------------------------------------------------
\840\ See, e.g., Nasdaq Letter I at 34-35; XTX Letter at 5;
Angel Letter at 2-3; Anonymous Letter (Feb. 12, 2023) (stating that
order sizes should be treated the same, regardless of status as a
round lot or an odd-lot).
\841\ See MDI Adopting Release, supra note 10, at 18618 n.274.
\842\ See MDI Adopting Release, supra note 10, at 18618 n.274.
---------------------------------------------------------------------------
iv. Modified Round Lot Assignment Frequency and Evaluation Period for
Round Lots
Under the MDI Rules, each NMS stock was assigned a round lot size
every month based on its average closing price for the prior calendar
month on its primary listing exchange.\843\ Commenters raised concerns
about potential confusion and operational risks arising due to the fact
that round lots and the proposed minimum pricing increments would be
changed at different times,\844\ and several commenters suggested
aligning the assignment of round lots and minimum pricing increments,
either on a quarterly or on a semiannual basis.\845\ One commenter
stated ``having two to four adjustments per year strikes the
appropriate balance between having the optimal round lot and minimum
pricing increment with reducing the time that market participants are
adjusting to the changes.'' \846\ Another commenter warned that having
differing assignment schedules for round lot sizes and minimum pricing
increments ``will materially elevate systemic risk since it only
requires a single large market participant to create widespread
disruption by failing to properly modify their systems.'' \847\ The
commenter suggested reducing the frequency of changes to quarterly or
semiannually and synchronizing the round lot and minimum pricing
increment changes.\848\
---------------------------------------------------------------------------
\843\ MDI Adopting Release, supra note 10, at 18617.
\844\ See, e.g., FIF Letter at 13; Hudson River Letter at 2, 4;
BlackRock Letter at 9, 10. See also SIFMA Letter II at 34, 35
(stating that ``The Tick Size Proposal would make dynamic three
components of trading that are static today: (i) tick sizes; (ii)
access fees; and (iii) round lots. Exacerbating this complication,
tick sizes would adjust quarterly, while round lots would change
monthly. Tick sizes would be based on average quoted spread, while
round lots are based on a stock's price.''); Morgan Stanley Letter
at 3 (stating that the proposed amendments would require ``market
participants to make frequent changes to their systems . . .'' that
``. . . the risk of technology failure created by monthly and/or
quarterly changes to systems by the buy-side, sell-side, exchanges
and vendors (including security information processors) may
introduce new market and operational risks . . . '' and that ``. . .
this dynamic aspect of the proposal could create investor confusion
and potentially drive trading inefficiencies.'').
\845\ Two commenters suggested aligning the assignment of round
lots and minimum pricing increments on a quarterly or semiannual
basis. See Hudson River Letter at 2, 4; BlackRock Letter at 10. One
commenter suggested aligning both assignments on a quarterly basis.
See FIF Letter at 13. See also infra notes 849-853 and accompanying
text.
\846\ Hudson River Letter at 4.
\847\ BlackRock Letter at 10.
\848\ See id.
---------------------------------------------------------------------------
After further consideration, the Commission agrees with the
concerns
[[Page 81673]]
raised by the commenters. Frequent systems changes and updates can
introduce risks for market participants and the market, and frequent
changes to the terms of how an order is entered for an NMS stock can
potentially cause investor confusion. Therefore, the Commission is
amending the round lot definition to make the timing for assigning
round lots consistent with the timing for assigning minimum pricing
increments. Such alignment on a semiannual basis will help to
facilitate an orderly market and help reduce operational risks and
investor confusion.
As described above, the Commission is adopting a definition of
Evaluation Period under Rule 612 that will result in the assignment of
minimum pricing increments on a semiannual basis instead of on a
quarterly basis, as proposed.\849\ The Commission has therefore decided
to amend the round lot definition to change the frequency of round lot
changes from a monthly basis to a semiannual basis (the round lot sizes
and price tiers for assigning round lots adopted in the MDI Adopting
Release have not changed). Specifically, the Commission is amending
Rule 600(b)(93) to require round lots to be assigned on a semiannual
basis instead of on a monthly basis, which will match the minimum
pricing increment assignment frequency of amended Rule 612.\850\
Amended Rule 600(b)(93)(i) will assign each NMS stock to a round lot
size based on the NMS stock's average closing price on the primary
listing exchange during a one month Evaluation Period. Amended Rule
600(b)(93)(iii) will define the Evaluation Period as (A) all trading
days in March for the round lot assigned on the first business day in
May and (B) all trading days in September for the round lot assigned on
the first business day of November during which the average closing
price of an NMS stock on the primary listing exchange shall be measured
by the primary listing exchange to determine the round lot for each NMS
stock.\851\
---------------------------------------------------------------------------
\849\ While the Commission proposed in Rule 612(a) to define a
quarterly tick evaluation period, it is adopting a semiannual tick
Evaluation Period in amended Rule 612(a)(1). See supra section
III.C.7.a.
\850\ The Commission's amendments to Rule 600(b)(93) will also
renumber the sub-provisions within rule 600(b)(93). The round lot
tiers in preexisting Rule 600(b)(93)(i)-(iv) will be renumbered as
Rule 600(b)(93)(i)(A)-(D). Preexisting Rule 600(b)(93)(v) will be
amended and renumbered as Rule 600(b)(93)(ii). Preexisting Rule
600(b)(93)(iii)-(iv) will contain provisions related to a new
``Evaluation Period.''
\851\ Amended Rule 600(b)(93) and amended Rule 612(a)(1) each
define ``Evaluation Period'' differently.
---------------------------------------------------------------------------
Further, amended Rule 600(b)(93)(iv) will provide time for market
participants to implement any reassignments of round lots and provides
that the assigned round lots will be operative until the next
semiannual date for a new round lot change. Specifically, round lots
assigned under Rule 600(b)(93) shall be operative on (A) the first
business day of May for the March Evaluation Period and continue
through the last business day of October of the calendar year, and (B)
the first business day of November for the September Evaluation Period
and continue through the last business day of April of the next
calendar year. For both round lots and minimum pricing increments, the
adopted semiannual assignment dates will be the first business day in
May and the first business day of November.
Like amended Rule 612, in amended Rule 600(b)(93)(iv) the
Commission is adopting a one-month time period between the conclusion
of each Evaluation Period and each operative date (i.e., the date on
which the round lot assignment becomes effective) to provide market
participants with time to implement any new round lot assignments.
These changes address concerns about operational risks and will help to
ensure the orderly implementation of the systems changes necessary to
implement the new round lots and minimum pricing increments. Further,
market participants will be able to use the one-month implementation
period to communicate with investors about any upcoming changes, which
will help to minimize potential investor confusion.
These amendments are responsive to commenters who suggested
aligning the assignment of round lots and minimum pricing
increments.\852\ By aligning the timing for assigning round lot sizes
to the timing for assigning minimum pricing increments, there will be
fewer modifications to market participants' systems and they will have
more time to implement such systems changes. These amendments to the
round lot definition address commenter concerns about systemic risk and
the risk of market disruptions because market participants will only
have to make systems changes two times per year for round lot
assignments and tick reassignments rather than twelve and four times
per year, respectively.\853\ These amendments to make the assignment of
round lots uniform with the assignment of minimum pricing increments
will reduce potential confusion for investors because they will only
have to understand two round lot assignments per year instead of
twelve.
---------------------------------------------------------------------------
\852\ See Hudson River Letter at 2, 4; BlackRock Letter at 10.
\853\ Id.
---------------------------------------------------------------------------
The Commission is not changing the round lot sizes, pricing tiers
or the calculation used to assign round lots. As originally adopted,
round lots were assigned based on the average closing price of the
prior month on the primary listing exchange. Under Rule
600(b)(93)(iii), round lots will be assigned based on the average
closing price during a specified month, e.g., March or September on the
primary listing exchange.
In the MDI Adopting Release, the Commission explained that
assigning a round lot size based on the NMS stock's average closing
price on the primary listing exchange for the prior calendar month
would strike ``an appropriate balance between using accurate, up-to-
date pricing information and avoiding the cost and complexity of over-
frequent computation and potential round lot reassignment.'' \854\ The
Commission also stated that market participants are accustomed to
monthly updates, so monitoring for round lot size changes and
implementing systems changes to account for the monthly calculation
``would not be overly burdensome or costly.'' \855\
---------------------------------------------------------------------------
\854\ See MDI Adopting Release, supra note 10, at 18619.
\855\ Id.
---------------------------------------------------------------------------
In light of the concerns raised by commenters about potential
confusion and potential operational risks due to the fact that round
lots and minimum pricing increments would be changed at different
times, the Commission reviewed data that compared how often round lot
sizes would change if subject to monthly evaluations, as the MDI Rules
previously required, to how often they would change if subject to a
semiannual evaluation based on one month of prices.\856\ The Commission
examined the average closing prices of NMS stocks from January 2019
through September 2023 (6,052 stocks) using
[[Page 81674]]
Bloomberg data. During this time, only 323 NMS stocks moved above or
below the round lot tiers of $250.01 per share, $1,000.01 per share,
and $10,000.01 per share. If round lot sizes were updated on a monthly
basis, there would have been 1,012 total changes over the past five
years, for an average of 17 changes per month. If round lot sizes were
updated every six months, there would have been 454 total changes over
the past five years, for an average of 50 changes every six
months.\857\ The data suggests that lengthening the time between
assigning round lots will reduce the number of re-assignments. Some of
the re-assignments identified using the monthly reassignment were the
result of some NMS stocks shifting between round lots sizes from month
to month. The shifting of round lot size tiers from month to month may
increase the potential for investor confusion. This is similar to the
concerns expressed by commenters about having asynchronous round lot
and minimum pricing increments changes.
---------------------------------------------------------------------------
\856\ See infra section VII.D.1.d for a discussion of the
semiannual evaluation of minimum pricing increments. As discussed
above, the Commission is aligning the assignment of round lots and
minimum pricing increments. A longer round lot assignment frequency,
such as annual assignments, would more likely result in round lot
sizes calculated based on prices not reflective of current trading.
Limiting round lot reassignments to a frequency of once every six
months was determined to be sufficient to achieve the goals stated
in the MDI Adopting Release, while reducing costs and complexity for
market participants. See also infra notes 1594-1595 and accompanying
text (discussing the impact of the semiannual evaluation period and
the lag between evaluation and implementation on the accuracy of
round lot assignments).
\857\ The Commission's analysis revealed that certain NMS stocks
shifted above and below the $250 per share threshold, which resulted
in the difference in number of changes between the monthly round lot
updates and the semiannual round lot updates. For example, one NMS
stock would have changed round lot size 24 times over five years if
round lot sizes were adjusted monthly, as compared to four times if
round lot sizes were adjusted semiannually.
---------------------------------------------------------------------------
Finally, the Commission is amending Rule 600(b)(93)(v), which
previously stated that a round lot for an NMS stock for which the prior
calendar month's average closing price is not available is an order for
the purchase or sale of 100 shares. This preexisting provision assigned
new NMS stocks to a 100-share round lot because such NMS stocks that
started trading intra-month did not have an average closing price from
the prior calendar month upon which to make a round lot
assignment.\858\ As amended, preexisting section (v) will be renumbered
as Rule 600(b)(93)(ii) and will be amended to state instead that any
security that becomes an NMS stock during an operative period as
described under new paragraph (iv) shall be assigned a round lot of 100
shares. This provision is consistent with the preexisting provision.
New NMS stocks that begin trading during an operative period will not
be able to have an average closing price calculated during an
Evaluation Period. Further, this new language will make the round lot
definition similar to Rule 612 in identifying those NMS stocks that
become NMS stocks during an operative period and have not yet been an
NMS stock during an Evaluation Period.
---------------------------------------------------------------------------
\858\ See MDI Adopting Release, supra note 10, at 18619.
---------------------------------------------------------------------------
C. Final Rule--Odd-Lot Information
The Commission is adopting an accelerated implementation schedule
for odd-lot information definition that is modified from the proposal
in order to provide a longer time for market participants to update and
modify their systems.\859\ Further, the Commission is adopting
amendments to Rule 603(b) under Regulation NMS, as proposed, to require
the exclusive SIPs to collect, consolidate and disseminate odd-lot
information. Finally, the Commission is adopting amendments to the
definition of odd-lot information to include the best odd-lot order, as
proposed.
---------------------------------------------------------------------------
\859\ See infra section VI.C.
---------------------------------------------------------------------------
1. Proposed Acceleration of Odd-Lot Information Definition
The Commission proposed to accelerate the implementation of the
odd-lot information definition under Rule 600(b)(59) \860\ by requiring
SROs to provide the data necessary to generate odd-lot information to
the exclusive SIPs and to require the exclusive SIPs to collect,
consolidate, and disseminate odd-lot information.\861\ Specifically,
the Commission proposed to amend Rule 603(b) under Regulation NMS to
require the national securities exchanges and national securities
associations to make all data necessary to generate odd-lot information
available to the exclusive SIPs and to require the exclusive SIPs to
collect, consolidate, and disseminate odd-lot information.\862\
---------------------------------------------------------------------------
\860\ The MDI Rules adopted the definition of odd-lot
information in rule 600(b)(59). This provision was subsequently
renumbered to Rule 600(b)(69) by the Rule 605 Amendments. 17 CFR
242.600(b)(69); Rule 605 Amendments, supra note 10.
\861\ Pursuant to the implementation period for the MDI Rules,
odd-lot information will be collected, consolidated, and
disseminated by competing consolidators, beginning during the
parallel operation period. See Proposing Release, supra note 11, at
80298.
\862\ See proposed Rule 603(b)(3). While the MDI Rules do not
require competing consolidators to disseminate all consolidated
market data elements, such as odd-lot information, in consolidated
market data products, the Commission proposed to require the
exclusive SIPs to collect, consolidate, and disseminate odd-lot
information. Under the decentralized consolidation model, competing
consolidators will be permitted to design consolidated market data
products with different elements of consolidated market data for
their subscribers and subscribers will be able to choose competing
consolidators and consolidated market data products that meet their
needs. See MDI Adopting Release, supra note 10, at 18659. Under the
existing exclusive SIP model, the exclusive SIPs are the only source
of consolidated NMS information and--while proprietary data products
offer some of the same data content, including odd-lot quotations--
subscribers would have no alternative providers of consolidated NMS
information if such data were not required to be collected,
consolidated, and disseminated by the exclusive SIPs. Therefore, the
Commission proposed that the exclusive SIPs be required to
disseminate odd-lot information.
---------------------------------------------------------------------------
The Commission proposed to divide Rule 603(b) into three new
subsections to reflect the requirements under Rule 603(b) until the MDI
Rules are implemented. As proposed, Rule 603(b)(1) governs the
applicability of Rules 603(b)(2) and (b)(3) by describing the
compliance dates set forth in the MDI Rules. Proposed Rule 603(b)(2)
governs the provision of consolidated market data by competing
consolidators and self-aggregators pursuant to the decentralized
consolidation model set forth in the MDI Rules. Proposed Rule 603(b)(3)
governs the provision of NMS information by the exclusive SIPs,
including the new requirements regarding the collection, consolidation,
and dissemination of odd-lot information.
Therefore, proposed Rule 603(b)(1)(i) states that compliance with
Rule 603(b)(3) is required until the date indicated by the Commission
in any order approving amendments to the effective national market
system plan(s) to effectuate a cessation of the operations of the plan
processors that disseminate consolidated information regarding NMS
stocks. Proposed Rule 603(b)(1)(ii) states that compliance with
proposed Rule 603(b)(2) is required 180 calendar days from the date of
the Commission's approval of the amendments to the effective national
market system plan(s) required under rule 614(e).\863\
---------------------------------------------------------------------------
\863\ 17 CFR 242.614(e). See also MDI Adopting Release, supra
note 10, at 18700 n.1355.
---------------------------------------------------------------------------
Preexisting Rule 603(b), which imposes requirements on the
dissemination of consolidated market data by national securities
exchanges and national securities associations, was proposed to be
renumbered as Rule 603(b)(2). Proposed Rule 603(b)(3) requires every
national securities exchange on which an NMS stock is traded and
national securities association to act jointly pursuant to one or more
effective NMS plans to disseminate consolidated information, including
a national best bid and national best offer and odd-lot information, on
quotations for and transactions in NMS stocks, and the effective plan
or plans must provide for the dissemination of all consolidated
information for an individual NMS stock through a single plan
processor. The single plan processor must represent quotation sizes in
such consolidated information in terms of the number of shares, rounded
down to the nearest multiple of a round lot.
[[Page 81675]]
Additionally, every national securities exchange on which an NMS stock
is traded and national securities association shall make available to a
plan processor all data necessary to generate odd-lot information.
The Commission did not receive any comments on proposed Rule
603(b)(1), which added the compliance dates already adopted in the MDI
Rules, or the renumbering of current Rule 603(b) as Rule 603(b)(2), and
is adopting these changes, as proposed. The Commission discusses
proposed Rule 603(b)(3) \864\ herein, which the Commission is adopting
as proposed.
---------------------------------------------------------------------------
\864\ See infra section V.C.1.a; section V.D.
---------------------------------------------------------------------------
a. General Comments and Response
The Commission received comments in support of the proposed
acceleration of the implementation of the odd-lot information
consistent with the MDI Rules from individuals, firms, exchanges, and
associations.\865\ Generally, individual commenters supported the
proposed acceleration of the implementation of the odd-lot information
definition because it would increase transparency and ``because odd-
lots represent the majority of trades.'' \866\ Certain other market
participants also supported the proposed acceleration of the odd-lot
information definition for similar reasons, stating greater
transparency would enhance price discovery, improve decision-making
with respect to order routing, and reduce spreads.\867\ Commenters
further supported the acceleration of odd-lot information requirements
because it would improve the quality of SIP data and ``make more data
accessible to investors at lower prices by introducing competition into
an otherwise monopolistic data market.'' \868\
---------------------------------------------------------------------------
\865\ The Commission also received comment on the timing
proposed to implement the odd-lot information and round lot
definitions. These comments are discussed below in section VI.C.
\866\ See, e.g., Form Letter Type I, of which 22 comments were
received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm. See also e.g., Form Letter Type D, of which 255 comments
were received, Form Letter Type J, of which 15 comments were
received, and Form Letter Type K, of which 22 comments were
received, available at https://www.sec.gov/comments/s7-30-22/s73022.htm; Letters from Aric Ott (Mar. 6, 2023); Austin Peck (Mar.
31, 2023); Colin Clarry (Mar. 6, 2023); Aron Tastensen (Feb. 23,
2023); Dave and Paula Wager (Mar. 6, 2023); Mark Rogers (Mar. 30,
2023); Erik Jansen (Mar. 31, 2023).
\867\ See, e.g., Cboe Letter II at 10 (stating that odd-lot
transactions represent a majority of trades, odd-lot quotations
represent significant price improvement on Cboe's exchanges and
stating ``the inclusion of odd-lot quotations on the SIPs is long
overdue''); IEX Letter I at 6; Nasdaq Letter I at 3, 10; MFA Letter
at 13-14; Better Markets Letter I at 16-17; NYSE Letter I at 7; Cboe
Letter II at 2; SIFMA AMG Letter I at 9; JPMorgan Letter at 2;
Letter from Tom Davin, Senior Vice President, Software & Information
Industry Association, Managing Director, Financial Information
Services Division, Financial Information Services Division of the
Software & Information Industry Association, dated Mar. 29, 2023
(``FISD Letter'') at 1, 3.
\868\ See, e.g., Robinhood Letter at 5; Proof Letter at 1.
---------------------------------------------------------------------------
Certain market participants stated that accelerating the
implementation of odd-lot information is not necessary and overly
burdensome given the other components of the proposal.\869\ One
commenter supported the acceleration of the MDI Rules with respect to
odd-lots but stated that ``[o]ver the long run, eliminating round lots
altogether . . . may be a better resolution.'' \870\ In contrast, two
commenters opposed accelerating the implementation of the odd-lot
information definition, stating that it would increase the amount of
development work required of market participants and therefore ``delay
the additional transparency that could be afforded by solely modifying
the round lot definition.'' \871\ Additional commenters supported
acceleration of the revised round lot definition but not the odd-lot
information definition, without providing a specific reason for the
distinction, and generally urged the Commission to revisit comments on
odd-lot dissemination.\872\
---------------------------------------------------------------------------
\869\ See, e.g., FIA PTG Letter II at 4; Hudson River Letter at
2; NYSE, Schwab, and Citadel Letter at 2; STA Letter at 8; Schwab
Letter II at 6; BlackRock Letter at 12; MEMX Letter at 7.
\870\ Nasdaq Letter I at 3, 10.
\871\ See FIA PTG Letter II at 4; Hudson River Letter at 2.
\872\ See, e.g., NYSE, Schwab, and Citadel Letter at 2; STA
Letter at 8; Schwab Letter II at 6.
---------------------------------------------------------------------------
The Commission is adopting proposed Rule 603(b)(3) with respect to
the provision of odd-lot information, as proposed. The provision of
odd-lot information within the national market system will provide
significant benefits to investors by increasing transparency about
better priced orders that are available in the market. The round lot
definition will not provide transparency about those orders that remain
odd-lots, i.e., odd-lot quotation information. As discussed above, only
those NMS stocks that are priced greater than $250 will be assigned a
smaller round lot size. These NMS stocks may still have odd-lots
available at prices better than the round lot price. Further, the odd-
lot information definition will provide transparency about better
priced odd-lot orders for all NMS stocks.
While some commenters suggested that the Commission consider
alternative sequencing of the proposal and expressed concern regarding
the acceleration of the implementation of the odd-lot information
definition: (1) ahead of other elements of MDI Rules, (2)
simultaneously with minimum pricing increments, and (3) simultaneously
with the changes to the round lot definition,\873\ investors and market
participants should be provided with the benefits of odd-lot
information sooner than the originally adopted implementation schedule
in the MDI Adopting Release,\874\ and the adoption of the acceleration
of the odd-lot information requirements should occur contemporaneously
with adoption of the other requirements outlined in the Proposing
Release. Timelier implementation of the odd-lot information definition
allows investors to benefit from greater transparency and accessibility
of better priced orders and improved execution quality; waiting to
implement the definition would delay these benefits for market
participants.\875\ Further, the implementation of the MDI Rules
continues, although on a delayed basis as compared to the adopted
implementation schedule. The implementation of odd-lot information on
an accelerated schedule will not impede the further implementation of
the remaining MDI Rules.\876\
---------------------------------------------------------------------------
\873\ See, e.g., ICI Letter I at 2, 7; FIA PTG Letter II at 4-5;
Robinhood Letter at 5, 44. See also supra section I.D (discussing
overarching comments on the proposal in general).
\874\ See Proposing Release, supra note 11, at 80295.
\875\ As discussed below, the Commission is adopting an
accelerated implementation schedule for the odd-lot information
definition that is modified from the proposal in order to provide a
longer period of time for market participants to update and modify
their systems. See infra section VI.C.
\876\ See infra section V.E.
---------------------------------------------------------------------------
Apart from comments regarding sequencing, certain industry
participants expressed concern that adding odd-lot information,
including the BOLO, to the exclusive SIPs will increase message traffic
and therefore increase costs.\877\ One commenter stated that the
proposal would add odd-lot information to exclusive SIP data without
disclosing how much the SROs would charge retail investors and broker-
dealers for the new data fields.\878\ The commenter, while recommending
that the Commission proceed with implementation of the MDI Rules and
governance changes, stated that exclusive SIP data fees are ``complex
and often opaque'' and that while SIP data costs are charged to retail
customers on a per investor basis, the cost to produce SIP data does
not scale on a per investor basis.\879\
---------------------------------------------------------------------------
\877\ See, e.g., FIF Letter at 13; FIA PTG Letter II at 5;
Fidelity Letter at 17; FISD Letter at 3.
\878\ See Fidelity Letter at 17.
\879\ See id.
---------------------------------------------------------------------------
[[Page 81676]]
While the addition of odd-lot quotation information to the
exclusive SIPs will increase the number of messages that the exclusive
SIPs will have to collect and consolidate and the number of messages
that will be made available to market participants, the exclusive SIPs
and market participants can handle such increased message traffic. As
discussed in the Proposing Release, the systems used by exchanges and
other market participants can handle many levels of data messages at
extreme low latency and should be able to adjust to the addition of
odd-lot quotation information.\880\ Further, the exclusive SIPs have
been discussing the addition of odd-lot quotation information to SIP
data for several years \881\ and should be able to make the necessary
adjustments to their processors in the adopted timeframe.\882\ To the
extent that increased message traffic increases costs for the exclusive
SIPs, the Commission estimated those costs in the Proposing Release,
which are discussed below.\883\
---------------------------------------------------------------------------
\880\ See Proposing Release, supra note 11, at 80279 (discussing
the potential increased system traffic for the proposed minimum
pricing increments).
\881\ See Proposing Release, supra note 11, at 80297 (discussing
a 2019 proposal by the CTA/CQ and UTP Plans to add odd-lot
information to the exclusive SIPs).
\882\ See infra section VI.C (revising the compliance timeframe
from 90 days as proposed to 18 months from the effective date of the
Adopting Release).
\883\ See infra sections VII.D.5 and VIII; see also Proposing
Release, supra note 11, at 80299, 80301.
---------------------------------------------------------------------------
The Commission discussed the potential for new fees related to
consolidated market data, which includes odd-lot information, in the
MDI Adopting Release.\884\ Fees imposed by the exclusive SIPs are
subject to the Exchange Act and the rules thereunder. The Commission
discussed the statutory standards for any potential fees for
consolidated market data, which includes odd-lot information, in the
MDI Adopting Release.\885\ Specifically, the statutory standards that
apply to fees proposed by the effective market system plan(s) include
section 11A(c)(1)(C)-(D) of the Exchange Act and rule 603(a) under
Regulation NMS. Proposed fees must be fair and reasonable and not
unreasonably discriminatory. As discussed in the MDI Adopting Release,
the Commission has historically assessed fees for data, such as the
data content underlying consolidated market data of which odd-lot
information is a part, using a reasonably related to cost
standard.\886\ To the extent that the exclusive SIPs propose to
increase SIP data fees because of the addition of odd-lot information,
any such new proposed fees must be filed with the Commission pursuant
to rule 608, published for public comment and approved by the
Commission before they can take effect.\887\ Further, as discussed in
the Proposing Release, expediting the inclusion of odd-lot information
to the exclusive SIPs can provide additional competition to the segment
of the market that subscribes to proprietary data with odd-lot
information for use in visual display settings.\888\
---------------------------------------------------------------------------
\884\ See MDI Adopting Release, supra note 10, at 18684.
\885\ See MDI Adopting Release, supra note 10, at 18650, 18684.
\886\ See MDI Adopting Release, supra note 10, at 18684 n.1158.
\887\ See 17 CFR 242.608(b). See also Rescission of Effective-
Upon-Filing Procedure for NMS Plan Fee Amendments and Modified
Procedures for Proposed NMS Plans and Plan Amendments, Securities
Exchange Act Release No. 89618 (Aug. 19, 2020), 85 FR 65470 (Oct.
15, 2020).
\888\ See Proposing Release, supra note 11, at 80338 (discussing
how expediting the inclusion of odd-lot data into the exclusive SIPs
would impact competition among data providers).
---------------------------------------------------------------------------
One commenter requested confirmation that exchange proprietary data
feeds could be used to provide odd-lot information to the exclusive
SIPs consistent with statements in the MDI Adopting Release that odd-
lot information could be made available to competing consolidators and
self-aggregators (under the decentralized consolidation model) using
``existing proprietary data feeds, a combination of proprietary data
feeds, or a newly developed consolidated market data feed.'' \889\ As
previously stated by the Commission, the use of proprietary data feeds
for delivering odd-lot information is consistent with the MDI Rules in
the context of the decentralized consolidation model.\890\ The use of
proprietary data feeds for purposes of providing data to the exclusive
SIPs may require consideration by the exclusive SIPs and the Operating
Committees of the technical specifications that may be necessary for
purposes of collecting and distributing such information to the
exclusive SIPs.\891\
---------------------------------------------------------------------------
\889\ MEMX Letter at 7.
\890\ See MDI Adopting Release, supra note 10, at 18653.
\891\ Under rule 603(a), an SRO is prohibited from making its
core data available to vendors on a more timely basis than it makes
such data available to the exclusive SIPs. In the MDI Adopting
Release, the Commission stated that rule 603(a) prohibits an SRO
from making its NMS information available to any person on a more
timely basis (i.e., by any time increment that could be measured by
the SRO) than it makes such data available to the exclusive SIPs.
See MDI Adopting Release, supra note 10, at 18656.
---------------------------------------------------------------------------
One commenter stated that ``disseminating odd lot quotes on the SIP
could lead investors to expect prices that are not available.'' \892\
Odd-lot quotation information will reflect actual prices of actual
orders that have been submitted by market participants. This
information will provide investors with valuable information about the
prices at which other market participants are willing to trade.
However, as with any change, market participants may have to educate
investors as to the existence of odd-lot quotation information on the
exclusive SIPs.
---------------------------------------------------------------------------
\892\ See Schwab Letter II at 36.
---------------------------------------------------------------------------
2. Proposed Amendment to Odd-Lot Information Definition for Best Odd-
Lot Orders
The odd-lot information definition includes (1) odd-lot
transactions,\893\ and (2) odd-lots at a price greater than or equal to
the national best bid and less than or equal to the national best
offer, aggregated at each price level at each national securities
exchange and national securities association.\894\ Accordingly, once
implemented, information on odd-lot orders priced better than the NBBO
will be included in the NMS information that is made available to
market participants within the national market system.
---------------------------------------------------------------------------
\893\ Odd-lot transaction information is currently collected,
consolidated, and disseminated by the exclusive SIPs. See Securities
Exchange Act Release Nos. 70793 (Oct. 31, 2013), 78 FR 66788 (Nov.
6, 2013) (order approving Amendment No. 30 to the UTP Plan to
require odd-lot transactions to be reported to the consolidated
tape); 70794 (Oct. 31, 2013), 78 FR 66789 (Nov. 6, 2013) (order
approving Eighteenth Substantive Amendment to the Second Restatement
of the CTA Plan to require odd-lot transactions to be reported to
consolidated tape).
\894\ 17 CFR 242.600(b)(69); MDI Adopting Release, supra note
10, at 18613.
---------------------------------------------------------------------------
The Commission proposed to amend the definition of odd-lot
information to include a BOLO as new Rule 600(b)(59)(iii).
Specifically, for each NMS stock, the best odd-lot order to buy would
mean the highest priced odd-lot order to buy that is priced higher than
the national best bid, and the best odd-lot order to sell would mean
the lowest priced odd-lot order to sell that is priced lower than the
national best offer. Similar to the definition of the NBBO, in the
event that two or more national securities exchanges or associations
provide odd-lot orders at the same price, the exclusive SIPs, competing
consolidators and self-aggregators would be required to determine the
best odd-lot order by ranking all such identical odd-lot buy orders or
odd-lot sell orders (as the case may be) first by size (giving the
highest ranking to the odd-lot buy order or odd-lot sell order
associated with the largest size), and then by time (giving the highest
ranking to the odd-lot buy order or odd-lot sell order received first
in time).\895\
---------------------------------------------------------------------------
\895\ See 17 CFR 242.600(b)(60) (defining NBBO and setting forth
the manner in which the NBBO is determined ``in the event two or
more market centers transmit to the plan processor, a competing
consolidator or a self-aggregator identical bids or offers for an
NMS security'').
---------------------------------------------------------------------------
[[Page 81677]]
a. General Comments and Response
The Commission received comments supporting the requirement to
identify the BOLO from individuals \896\ and market participants.\897\
As stated above, many individual commenters supported publishing odd-
lot information, including the BOLO, in the exclusive SIPs in order to
further transparency and aid investors in making more informed trading
decisions.\898\ Similarly, certain other commenters viewed the
additional information as a useful measure for all investors and their
agents to better evaluate the best prices in NMS stocks and would
enhance the ability to trade and route orders effectively as well as
facilitate best execution.\899\ Certain market participants that
support the publication of the BOLO cautioned against requiring broker-
dealers to use the metric as a benchmark against execution
quality.\900\ One of the commenters requested guidance regarding
whether ``market participants would be expected to clear the best odd-
lot orders as part of ISO routing, notwithstanding that the odd-lot
orders are not protected quotations.'' \901\ Finally, a number of
individual commenters that expressed support for including odd-lot
information in the exclusive SIPs urged the Commission to include
``odd-lot transactions'' in the NBBO, citing the fact that odd-lot
transactions are now a majority of the market and most prevalent among
retail investors.\902\
---------------------------------------------------------------------------
\896\ See, e.g., Comment Letter Type D; Letters from Aron
Tastensen (Feb. 23, 2023); Bill Goerger (Mar. 19, 2023); Carson
Bruenderman (Mar. 7, 2023).
\897\ See, e.g., IEX Letter I at 6; Cboe, State Street, et al.
Letter at 2; ASA Letter at 5; SIFMA Letter II at 4, 32; BlackRock
Letter at 12. The Commission also received a comment that said that
the Commission should not amend rule 603(c) to require the display
of odd-lot information, but did not discuss the costs (or benefits)
of such a requirement. See FIF Letter at 13. The Commission is not
amending rule 603(c) in this release.
\898\ See, e.g., Comment Letter Type D, I, and K, available at
https://www.sec.gov/comments/s7-30-22/s73022.htm; see supra notes
866 and 896.
\899\ See, e.g., IEX Letter I at 30-31; ICI Letter I at 6; Cboe
Letter II at 10; BlackRock Letter at 12.
\900\ See, e.g., ASA Letter at 5; Fidelity Letter at 16; SIFMA
Letter II at 4, 32.
\901\ SIFMA Letter II at 34, 43.
\902\ See, e.g., Comment Letter Type D, I and K; Letters from
Chris Eastvedt (Mar. 6, 2023); M B (Mar. 6, 2023); Prakash Tamang
(Mar. 6, 2023); Adam Aiello (Mar. 7, 2023); Shayne Gallagher (Mar.
7, 2023); Aswin Joy (Mar. 7, 2023); Daryll Fogal (Mar. 15, 2023);
Bill Goerger (Mar. 19, 2023); Allie Birge (Mar. 31, 2023); Eileen
Loh (Mar. 19, 2023).
---------------------------------------------------------------------------
The Commission also received comments opposing the requirement to
identify the BOLO, stating the information would create ambiguity or
investor confusion.\903\ Some of these commenters stated that confusion
would arise from the fact that a customer would expect to receive the
BOLO price even though it is not a protected quote.\904\ One commenter
stated that the round lot and odd-lot requirements outlined in the MDI
Adopting Release sufficiently provide increased transparency while
minimizing confusion.\905\ One commenter stated that the transparency
of odd-lot orders may be ``gameable'' such that a limit order to buy
one share could change all execution quality benchmarks for
brokers.\906\
---------------------------------------------------------------------------
\903\ See, e.g., Data Boiler Letter II at 3; JPMorgan Letter at
2, 7.
\904\ See JPMorgan Letter at 7 (stating further if the protected
price is lower than the BOLO then ``Rule 605 could show misleading
negative price improvement while ignoring the order's size'');
Fidelity Letter at 16; SIFMA Letter II at 43.
\905\ JPMorgan Letter at 7.
\906\ Fidelity Letter at 16 (supporting ``adding better-priced
odd lots to the SIP when this information provides actionable
information to the marketplace, such as helping broker-dealers meet
their best execution obligations'' but urging the Commission to
``balance the advantages and disadvantages of odd-lot
transparency,'' such as the ability to influence execution quality
statistics).
---------------------------------------------------------------------------
As discussed below, the Commission is adopting the amendment to
odd-lot information to include a BOLO, as proposed.\907\ While
initially market participants may need to explain to their customers
about the existence of the BOLO, this new data element is not expected
to confuse investors. Investors are already able to see odd-lot
transaction information and, upon implementation, the BOLO will provide
them with information about the best odd-lot quotations. The BOLO is an
informative, useful piece of information for investors to use when
considering prices related to NMS stocks. Among other uses, the BOLO
may serve as the benchmark execution price for execution quality
statistics in rule 605 reports that measure price improvement relative
to the best available displayed price.\908\ However, in rule 605
reports, the price improvement statistics relative to the best
available displayed price will be a supplement to, rather than a
replacement for, price improvement statistics relative to the
NBBO.\909\ Further, rule 605 reports present information, including
price improvement, in order size categories based on notional order
size and whether the order is for a fractional share, odd-lot, or round
lot.\910\ These provisions provide more context for price improvement
statistics that consider the best available odd-lot price and thus
mitigate concerns about ``gaming'' execution quality reports. Further,
rule 605 reports represent monthly, aggregated execution quality
statistics and thereby dilute the effect of an odd-lot price at one
specific point in time.
---------------------------------------------------------------------------
\907\ The Commission is adopting this amendment to the
definition of odd-lot information in Rule 600(b)(69)(iii). See supra
note 860.
\908\ See 17 CFR 242.600(b)(14) (defining best available
displayed price) and 17 CFR 242.605(a)(1)(ii)(M) through (Q)
(requiring rule 605 statistics relative to the best available
displayed price). Entities that prepare rule 605 reports will be
required to use the BOLO to compare the best available odd-lot price
to the NBBO and determine the best available displayed price. In
some cases, the best available displayed price may be the NBBO.
\909\ See Rule 605 Amendments, supra note 10.
\910\ See Rule 605 Amendments, supra note 10.
---------------------------------------------------------------------------
Several commenters stated that odd-lot quotations should be
included in the NBBO.\911\ Odd-lot quotations are currently included in
the calculation of the NBBO when they are aggregated into round lots
for purposes of providing an exchange's best bids and offers to the
exclusive SIPs.\912\ Pursuant to Regulation NMS, bids and offers can
only be in round lot sizes,\913\ therefore, the NBBO can only be
reflected in round lot sizes. The round lot definition as adopted in
the MDI Rules, will categorize certain orders that are currently odd-
lots as round lots based on their price, and as a result, quotations
and orders that were previously defined as odd-lots will be eligible to
establish the NBBO.\914\ Orders that remain odd-lots under the new
definitions are not bids and offers and therefore do not independently
contribute to establishing the NBBO.
---------------------------------------------------------------------------
\911\ See supra note 902.
\912\ See, e.g., NYSE Rule 7.36(b)(3); Nasdaq Equity 4, Rule
4756(c)(2); Cboe BZX Rule 11.9(c)(2). See also supra note 732.
\913\ 17 CFR 242.600(b)(16) (defining ``bid or offer'' to mean
``the bid price or the offer price communicated by a member of a
national securities exchange or member of a national securities
association to any broker or dealer, or to any customer, at which it
is willing to buy or sell one or more round lots of an NMS security,
as either principal or agent, but shall not include indications of
interest.'').
\914\ See supra section V.B.1.
---------------------------------------------------------------------------
The identification of a BOLO will assist investors in assessing the
current state of the market for individual NMS securities. The BOLO
will reflect the best odd-lot price consolidated across all national
securities exchanges and national securities associations and is
therefore consistent with the goals set forth in section 11A of the
Exchange Act because it will make information about quotations in NMS
stocks available to broker-dealers and investors and will enhance the
usefulness of odd-lot information. Although odd-lot liquidity better
than the NBBO often resides at multiple price levels and information
[[Page 81678]]
reflecting all of these odd-lot prices is already included in the
definition of odd-lot information, requiring the identification and
dissemination of the best of all such inside the NBBO odd-lots on both
the buy and sell side will help inform market participants of the best
possible prices at which their orders (or their customers' orders)
could--in whole or in part--be executed. The identification and
dissemination of the price, size, and market of the best odd-lot orders
will also enhance the ability of market participants to make effective
trading and order routing decisions using NMS information and
facilitate best execution. One commenter requested guidance on how to
treat odd-lot orders for order routing purposes.\915\ The Commission
stated in the MDI Adopting Release that odd-lot information may be
relevant to a broker-dealer's ability to analyze and achieve best
execution.\916\ As the Commission stated in the MDI Adopting Release,
while odd-lot information, which will now include the BOLO, ``may be
relevant to broker-dealers' best execution analyses and, in many cases,
will facilitate the ability of broker-dealers to achieve best execution
for their customer orders, the Commission . . . is not setting forth
minimum data elements needed to achieve best execution and does not
expect that all market participants will need to purchase the most
comprehensive or fastest consolidated market data product available.''
\917\
---------------------------------------------------------------------------
\915\ See supra note 901.
\916\ See MDI Adopting Release, supra note 10, at 18605 for a
discussion of the implications of expanded consolidated market data
on the duty of best execution.
\917\ Id. at 18605-06. Consolidated market data products will be
developed by competing consolidators once the decentralized
consolidation model is implemented. See rule 600(b)(25), 17 CFR
242.600(b)(25) (defining consolidated market data product).
---------------------------------------------------------------------------
D. Display of Round Lots and Odd-Lot Information
Currently, the exclusive SIPs represent quotation sizes in SIP data
in terms of number of round lots. For example, for an NMS stock for
which a round lot is 100 shares, a bid for 200 shares of that stock
would be represented as a bid for ``2'' in SIP data.
Under the MDI Rules, competing consolidators are required to
represent a round lot as the number of shares rounded down to the
nearest multiple of a round lot.\918\ For example, a 275-share buy
order at $25.00 for a stock with a 100-share round lot would be
disseminated as ``200.'' \919\ Accelerated implementation of the round
lot definition will require the exclusive SIPs to revise their systems
to reflect this change. Therefore, in proposed Rule 603(b)(3), the
Commission proposed to require each exclusive SIP to, among other
things, represent quotation sizes in consolidated information in terms
of the number of shares, rounded down to the nearest multiple of a
round lot.
---------------------------------------------------------------------------
\918\ Under the MDI Rules, the definition of ``core data''
requires competing consolidators to represent certain core data
elements, including the best bid and best offer, the NBBO, and
protected quotations in terms of the number of shares, rounded down
to the nearest multiple of a round lot. 17 CFR 242.600(b)(26)(iii).
See also 17 CFR 242.600(b)(26) (defining ``core data''). The MDI
Rules adopted the definition of ``core data'' in rule 600(b)(21).
This provision was subsequently renumbered to rule 600(b)(26) by the
Rule 605 Amendments. See Rule 605 Amendments, supra note 10.
\919\ See MDI Adopting Release, supra note 10, at 18615. Through
the definition of ``odd-lot information,'' the MDI Rules also
require odd-lots priced at or better than the NBBO to be represented
in the aggregate at each price level at each national securities
exchange or national securities association rather than on an order-
by-order basis. 17 CFR 242.600(b)(69)(ii). See also 17 CFR
242.600(b)(26)(i)(H) (including ``odd-lot information'' as an
element of core data). The MDI Adopting Release explained that
``[a]ggregating better-priced odd-lots at each price level at each
exchange . . . . means that better-priced odd-lot orders will be
represented in core data in terms of the total number of shares
available at each price level at each exchange rather than on an
order-by-order basis. For example, if the NBB for XYZ, Inc. is 100
shares at $25.00, and there are three orders of five shares and two
orders of ten shares at $25.01 on Exchange A, a competing
consolidator's core data product would show 35 shares at $25.01 on
Exchange A.'' MDI Adopting Release, supra note 10, at 18613 n.199.
Therefore, quotations for odd-lot orders priced better than the NBBO
are required to be displayed as the number of shares available in an
odd-lot size that are aggregated at the same price.
---------------------------------------------------------------------------
1. Comments and Response
The Commission received comments on the display requirement adopted
as part of the MDI Rules.\920\ One commenter stated that, for mixed lot
orders, the total number of shares--both the round lot and odd-lot
portions--should be included as consolidated market data.\921\ Another
commenter stated that, by requiring that the number of shares at each
price level be displayed at the round lot level, the display
requirement would make consolidated market data less useful and less
competitive relative to exchange proprietary data feeds, which display
the total number of shares at a price level.\922\ The commenter also
stated that basing quotations on the number of shares rounded to the
nearest round lot (rather than based on round lots) could result in
operational risk and investor confusion.\923\
---------------------------------------------------------------------------
\920\ See FIF Letter at 13; SIFMA Letter II at 34, 42.
\921\ See FIF Letter at 13.
\922\ See SIFMA Letter II at 42.
\923\ See SIFMA Letter II at 34.
---------------------------------------------------------------------------
As stated above, the display requirement was adopted as part of the
MDI Rules. The Commission discussed the reasons for adopting the
display requirement in the MDI Adopting Release.\924\ However, in
response to the commenter that suggested the inclusion of both the
round lot and odd-lot portions of mixed lot orders,\925\ since odd-lot
information will be disseminated by the exclusive SIPs,\926\ the total
number of shares of odd-lots priced at or better than the NBBO will be
included in SIP data. Through the definition of odd-lot information,
the MDI Rules require odd-lots priced at or better than the NBBO to be
represented in the aggregate at each price level at each national
securities exchange or national securities association rather than on
an order-by-order basis.\927\
---------------------------------------------------------------------------
\924\ See MDI Adopting Release, supra note 10, at 18615.
\925\ See FIF Letter at 13.
\926\ See amended Rule 603(b)(3).
\927\ 17 CFR 242.600(b)(69)(ii). See also supra note 919.
---------------------------------------------------------------------------
In response to the commenter that stated that displaying quotations
in round lot sizes would undermine the usability and competitiveness of
consolidated market data as compared to exchange proprietary data
feeds,\928\ the Commission described the reason for displaying
quotations in round lot sizes in the MDI Adopting Release, which
remains relevant to the implementation of the round lot definition by
the exclusive SIPs.\929\ Further, the Commission recognized in the MDI
Adopting Release that different market participants need differing
amounts of information to meet different trading objectives.\930\ The
MDI Rules are intended to reduce information asymmetries between users
of proprietary feeds and users of SIP data by enhancing the content of
information made available in the national market system to enable
market participants to trade efficiently and competitively.\931\ For
example, the inclusion of odd-lot quotations and the round lot
definition will allow investors to see, and more readily access, better
priced orders in smaller sizes.\932\ As discussed above, the total
number of shares of odd-lots priced at or better than the NBBO will be
included in data that is made available by the exclusive SIPs. Certain
market participants may choose to continue to purchase exchange
proprietary data products if
[[Page 81679]]
they require more granular information about odd-lots. Here, the
Commission's amendment is limited in reach--it extends the MDI Rules'
display requirement to the exclusive SIPs as part of the accelerated
implementation of the round lot and odd-lot information definitions,
changing only the entity responsible for displaying this information--
from competing consolidators to the exclusive SIPs. This change by
itself should not impact the utility or competitiveness of consolidated
market data. With respect to the commenter's concerns that the display
requirement would result in operational risk and investor
confusion,\933\ the required display of mixed lot orders rounded down
to the nearest multiple of a round lot was previously adopted in the
MDI Rules for competing consolidators.\934\ In the MDI Adopting
Release, the Commission stated that the preexisting convention of
displaying the number of round lots ``could be confusing'' when applied
to the MDI Rules' round lot definition, which, once implemented, will
assign varying round lot sizes to individual NMS stocks based on stock
price.\935\ Further, the Commission explained that rounding down to the
nearest round lot multiple would ensure that the elements of core data
would reflect orders of meaningful size, and that for the NBBO,
rounding down would help ensure that the protected portion of the order
is clearly represented, to address concerns about impacts on investor
confidence and investor confusion that potentially could result from
the display of unprotected size at the NBBO.\936\ The Commission also
stated that odd-lots priced at or better than the NBBO, including the
odd-lot portion of a mixed lot order at the NBBO, will be included in
core data.\937\ The Commission is extending this display requirement to
exclusive SIPs as part of the accelerated implementation of the round
lot and odd-lot information definitions. There should not be any new
operational risks or investor confusion arising from this change
because only the entity responsible for displaying the information is
changing.
---------------------------------------------------------------------------
\928\ See SIFMA Letter II at 42.
\929\ See MDI Adopting Release, supra note 10, at 18615.
\930\ See MDI Adopting Release, supra note 10, at 18600.
\931\ See MDI Adopting Release, supra note 10, at 18601.
\932\ See MDI Adopting Release, supra note 10, at 18601, 18607.
\933\ See SIFMA Letter II at 34.
\934\ 17 CFR 242.600(b)(26)(iii).
\935\ See MDI Adopting Release, supra note 10, at 18615.
\936\ See MDI Adopting Release, supra note 10, at 18615 n.236.
Rule 611 of Regulation NMS requires trading centers to have policies
and procedures that are reasonably designed to prevent ``trade-
throughs'' on that trading center of protected quotes in NMS stocks,
subject to specified exceptions. 17 CFR 242.611. Rule 611 currently
only applies to round lots. Specifically, rule 611 applies to
``protected quotations'' which means ``protected bid[s] or [
]protected offer[s].'' 17 CFR 242.600(b)(82). ``Protected bid or
protected offer,'' as defined in rule 600(b)(81), refers to ``a
quotation,'' defined in rule 600(b)(86), which in turn refers to ``a
bid or an offer,'' defined in rule 600(b)(16), which, as noted
above, only applies to round lots. See supra note 913 and
accompanying text.
\937\ See MDI Adopting Release, supra note 10, at 18615 n.236;
see also supra note 919.
---------------------------------------------------------------------------
The Commission is adopting Rule 603(b)(3) as proposed.
E. MDI Rules Implementation
Some commenters discussed the benefits of the MDI Rules \938\ and
expressed concern that the accelerated implementation of the round lot
and odd-lot information definitions could indefinitely delay
implementation of the remainder of the MDI Rules,\939\ stating that
these proposed changes were not a substitute for implementation of all
of the MDI Rules.\940\ Some commenters suggested that the changes
proposed in the Regulation NMS Proposal should be postponed until the
full implementation of the MDI Rules.\941\ Some commenters also raised
concerns that the Commission was separately accelerating the
implementation of the round lot and the odd-lot information definitions
apart from the other components of the MDI Rules.\942\ One commenter
stated that ``full implementation of the MDI Rules'' is ``a necessary
first step for any significant changes to market structure,'' and
whether or not fully implemented, ``an indispensable component of the
`baseline' against which this proposal must be measured and
justified.'' \943\
---------------------------------------------------------------------------
\938\ See, e.g., Schwab Letter II at 36; FIA PTG Letter II at 5;
Robinhood Letter at 46-47.
\939\ See, e.g., JPMorgan Letter at 7; Citadel Letter I at 26;
ICI Letter I at 2-3 n.8; Robinhood Letter at 5. See also Schwab
Letter II at 36.
\940\ See SIFMA Letter II at 44. This commenter and another
commenter stated that the Commission has not taken action to ensure
the implementation of the full set of MDI Rules since disapproving
the proposed fees and proposed amendments to the current NMS plans
for consolidated market data in 2022. Id.; SIFMA AMG Letter I at 9.
See also Fidelity Letter at 4; FIA PTG Letter II at 5.
\941\ See, e.g., SIFMA Letter II at 44; Robinhood Letter at 5.
\942\ See, e.g., Robinhood Letter at 5, 38, 41-42, 43, 47, 48,
49; JPMorgan Letter at 7; Citadel Letter I at 26.
\943\ Robinhood Letter at 49. See also Robinhood Letter at 44,
46-49. See also infra section VII.C.3.
---------------------------------------------------------------------------
Despite delays in the process,\944\ the implementation of the MDI
Rules continues to be a Commission priority. In September 2023, the
Commission issued an amended order directing the SROs to file a
proposed new single national market system plan regarding consolidated
equity market data.\945\ Consolidation of the multiple Equity Data
Plans into a single, new equity data plan would modernize the
governance of the existing Equity Data Plans.\946\
---------------------------------------------------------------------------
\944\ See supra notes 73-76 and accompanying text.
\945\ See supra note 78.
\946\ Rule 614(e) of the MDI Rules requires that an amendment to
the effective national market system plan(s) be filed with the
Commission to conform such plan(s) to the decentralized
consolidation model. 17 CFR 242.614(e).
---------------------------------------------------------------------------
The Commission disagrees with comments that recommended delaying
implementation of the Regulation NMS Proposal until the full
implementation of the MDI Rules.\947\ Due to the delayed implementation
of the MDI Rules, the Commission proposed to accelerate the
implementation of the round lot and odd-lot definitions because these
definitions can be efficiently implemented under the current exclusive
SIP model.\948\ Not doing so would unnecessarily delay the benefits of
the round lot and odd-lot information definitions to investors and
market participants. One goal in adopting the round lot definition was
to increase transparency about the better priced orders available in
the market by allowing each exchange's BBO and the NBBO for an NMS
stock to be based upon smaller, potentially better priced orders, which
will also improve market participants' ability to access these
orders.\949\ Waiting to implement the round lot definition would delay
these benefits for market participants. Further, full implementation of
the MDI Rules will not address the issues discussed above related to
the minimum pricing increments for certain NMS stocks, the access fee
caps for protected quotations and exchange fees.\950\ Finally, as
discussed below, the eventual implementation of the MDI Rules is part
of the baseline for the amendments to Rules 610 and 612.\951\
---------------------------------------------------------------------------
\947\ See, e.g., SIFMA Letter II at 44; Robinhood Letter at 5;
see also Robinhood Letter at 42, 49.
\948\ See Proposing Release, supra note 11, at 80296 n.359.
\949\ See Proposing Release, supra note 11, at 80294. For more
details, see MDI Proposing Release, supra note 744, at 16743.
\950\ See supra sections III.A, IV.C, and IV.E.
\951\ See infra section VII.C.3.
---------------------------------------------------------------------------
VI. Compliance Dates
The Commission proposed different compliance dates for the
individual proposed rule amendments. As discussed below in the relevant
sections, the Commission received several comments on the proposed
compliance dates.\952\ The Commission is adopting compliance dates that
are longer than proposed.\953\
---------------------------------------------------------------------------
\952\ See, e.g., NYSE Letter I at 7-8; CTA, CQ, UTP Plans
Operating Committees Letter at 3; FISD Letter at 2, 3, 4; Nasdaq
Letter I at 3; Cboe Letter II at 11; FIF Letter at 14; Cboe Letter
III at 10 n.18.
\953\ In addition, with respect to the compliance dates, several
commenters requested the Commission consider the interaction between
the proposed rules and other recent Commission rules. In determining
compliance dates, the Commission considers the benefits of the rules
as well as the costs of delayed compliance dates and the potential
overlapping compliance dates. For reasons discussed throughout the
release, to the extent that there are costs from overlapping
compliance dates, we expect the benefits of the rules to justify
such costs. See infra section VII.D.6 for a discussion of the
interactions of the final rules with certain other Commission rules.
---------------------------------------------------------------------------
[[Page 81680]]
Specifically, for the reasons discussed below, the amendments
adopted herein will have the following compliance dates:
Rules 600(b)(89)(i)(F) and 612: The first business day of November
2025.
Rules 600(b)(89)(iv), 600(b)(93) and 603(b)(3) (with respect to the
requirement that the effective national market system plans to
disseminate consolidated information shall provide for the
dissemination of all consolidated information for an individual NMS
stock through an exclusive SIP, and that the exclusive SIPs must
represent quotation sizes in such consolidated information in terms of
the number of shares, rounded down to the nearest multiple of a round
lot): The first business day of November 2025.
Rule 610: The first business day of November 2025.
Rules 600(b)(69) and 603(b)(3) (with respect to the requirement
that every national securities exchange on which an NMS stock is traded
and national securities association must make available to the
exclusive SIPs all data necessary to generate odd-lot information, and
the collection, consolidation and dissemination of odd-lot information
by the exclusive SIPs): The first business day of May 2026.
A. Final Rule 612 Compliance Date
In the Proposing Release, the Commission detailed a staggered
implementation period that would cover five quarters for the proposed
amendments to Rule 612. The Commission proposed the implementation
period to provide the market and market participants with time to
implement the proposed variable minimum pricing increments as well as
to facilitate an orderly transition. The adopted amendments to Rule 612
are modified from those that were proposed. Accordingly, the Commission
is adopting a modified implementation schedule and compliance date.
One commenter suggested that the Commission direct the SROs to
develop a phased implementation schedule for the reduced minimum
pricing increment, ``[c]onsistent with the prior implementation of
decimalization'' and described several steps to be considered in an
implementation plan.\954\ Because the amendments to Rule 612 have been
modified to require the addition of only one minimum pricing increment,
the compliance date discussed below will provide the SROs and other
market participants sufficient time to implement the changes and a
further phased implementation schedule is unnecessary. The move to
decimalization in 2000-2001 was more complicated as it involved changes
to SRO rules that specified several different increments that were
fractions of a dollar and required systems changes to accommodate
decimals instead of fractions.\955\ The amendment to Rule 612 will be
less complicated because it will not require changes to SRO rules and
the systems that are in place today, while needing updates, already can
accommodate sub-penny increments. Further, the amendments adopted
require less changes than what were proposed.
---------------------------------------------------------------------------
\954\ See FIF Letter at 2, 8.
\955\ For a description of the move to decimalization, see Staff
Decimalization Report, supra note 26.
---------------------------------------------------------------------------
The amendments to Rule 612 will require the primary listing
exchanges to evaluate each NMS stock during an Evaluation Period to
calculate their TWAQS and the primary listing exchanges will have to
provide a minimum pricing increment indicator to the exclusive SIPs for
dissemination. The Evaluation Periods will be conducted on a semiannual
basis, rather than a quarterly basis as proposed. Further, the
amendments will require market participants to update and modify their
systems, such as order handling and processing systems, to accommodate
the one new minimum pricing increment, rather than the three new
minimum pricing increments that were proposed. Market participants'
systems will have to be updated and modified to accommodate the
assignment of minimum pricing increments for quotes and orders priced
$1.00 or greater to each NMS stock on a semiannual basis, rather than a
quarterly basis as proposed. The systems updates necessary for
implementing the amendment to Rule 612 are less burdensome than what
was proposed.
The Commission has considered the systems changes that will be
necessary to implement the amendments to Rules 600(b)(89)(i)(F) and 612
and is assigning the compliance date for amended Rule 612 to be the
first business day of November 2025. In determining this compliance
date, the Commission considered the systems changes that must be
completed and the date by which the TWAQS can be calculated during an
Evaluation Period after the systems changes could be completed. This
compliance date is sufficient for facilitating an orderly transition to
the amended Rule 612.
B. Final Rule 610 Compliance Date
The Commission proposed that compliance with the amendments to Rule
610 would have occurred during the implementation period proposed for
the amendments to Rule 612, discussed above.\956\ The proposed access
fee caps would have also had a staggered implementation to reflect the
proposed implementation of the proposed minimum pricing increments.
Specifically, compliance with the proposed 10 mils access fee cap would
have been at the same as the proposed $0.005 minimum pricing increment,
and compliance with the proposed 5 mils access fee cap would have been
at the same time as the proposed $0.001 minimum pricing increment.\957\
---------------------------------------------------------------------------
\956\ See Proposing Release, supra note 11, at 80284.
\957\ See Proposing Release, supra note 11, at 80284 n.249.
---------------------------------------------------------------------------
As described above, the Commission has modified amendment to Rule
612 and has also modified the compliance date for the Rule 612
amendments. Further, the Commission has modified the amendment to Rule
610 such that the access fee cap structure is retained and only the
level of the caps has been reduced. The Commission is reducing the
access fee caps under Rule 610 to accommodate the new pricing
increments as well as to address distortions in the market associated
with fee and rebate models.\958\ Accordingly, the Commission is
modifying the compliance date for the final Rule 610 amendments to
coincide with the compliance date for Rule 612. The national securities
exchanges will have to file proposed rule changes with the Commission
pursuant to section 19(b) and rule 19b-4 \959\ to adjust their fee
schedules to reflect the new lower access fee caps. Further, the
national securities exchanges will have to file proposed rule changes
to adjust any fee or rebate that is not determinable at the time of
execution.
---------------------------------------------------------------------------
\958\ See supra section IV.D.1.
\959\ See 15 U.S.C. 78s(b); 17 CFR 240.19b-4.
---------------------------------------------------------------------------
The Commission is adopting a first business day of November 2025
compliance date for the amendments to Rule 610. This date provides the
national securities exchanges with time to assess their fee schedules
and file proposed rule changes pursuant to section 19(b) and rule 19b-4
to adjust
[[Page 81681]]
their fee schedules in order to comply with Rule 610 as amended.
C. Final Compliance Date for Round Lot and Odd-Lot Information
In the Proposing Release, the Commission proposed to require
compliance with the odd-lot information and round lot definitions,
including, as required under proposed Rule 603(b), that national
securities exchanges and associations make the data available to the
exclusive SIPs, that the exclusive SIPs represent quotation sizes in
consolidated information in terms of the number of shares, rounded down
to the nearest multiple of a round lot, and that the exclusive SIPs
disseminate odd-lot information as defined in Rule 600(b)(69) \960\ 90
days from Federal Register publication of any Commission adoption of an
earlier implementation of the round lot and odd-lot information
definitions.\961\ The Commission explained that the proposed compliance
date would significantly move up the date by which round lot and odd-
lot information would be more widely available in the national market
system.\962\
---------------------------------------------------------------------------
\960\ See supra note 719.
\961\ See Proposing Release, supra note 11, at 80300; see id. at
n.399 and accompanying text.
\962\ See Proposing Release, supra note 11, at 80298.
---------------------------------------------------------------------------
Several commenters raised concerns about the proposed compliance
date, stating that 90 days was not enough time to implement the round
lot and odd-lot information definitions.\963\ In stating that a longer
timeframe was needed, some commenters stated their views of the
challenges entailed in implementing the changes.\964\ One commenter
stated, ``[t]he technical and operational requirements to implement the
definition changes will necessitate distinct product changes in the
systems of literally hundreds of exchanges, vendors, and subscribers,
each with different development priorities and system capabilities.''
\965\ The commenter stated that 90 days would not be enough time for
the exclusive SIPs, data vendors and subscribers to accommodate the
changes, and cautioned that ``hastily made changes or missed delivery
dates could result in not just a failure to provide odd-lot quotation
data but also disrupt the flow of other core data to the market.''
\966\ The commenter urged the Commission not to adopt the 90-day
compliance timeframe and instead, after adoption of the Proposing
Release, allow time for industry consultation to develop an
implementation plan.\967\ Two commenters, while supporting the odd-lot
information definition adopted in the MDI Rules, recommended only
implementing the BOLO due to the complexity and time it would take to
implement the odd-lot information definition by many market
participants.\968\
---------------------------------------------------------------------------
\963\ See, e.g., NYSE Letter I at 7-8; CTA, CQ, UTP Plans
Operating Committees Letter at 3; MEMX Letter at 7; FISD Letter at
2, 3, 4; Nasdaq Letter I at 3; Cboe Letter II at 11; FIF Letter at
14; Cboe Letter III at 10 n.18. See also Fidelity Letter at 16.
\964\ See, e.g., NYSE Letter I at 7-8; CTA, CQ, UTP Plans
Operating Committees Letter at 2, 3; FISD Letter at 2, 3, 4; FIF
Letter at 14; Cboe Letter II at 11.
\965\ See FISD Letter at 3.
\966\ Id.
\967\ See FISD Letter at 4. The commenter stated that the
Operating Committees of the Equity Data Plans considered a 10-12
month implementation process in their proposal to add odd-lot data
to the exclusive SIP feeds. Id.
\968\ See BlackRock Letter at 12. See also MEMX Letter at 7.
---------------------------------------------------------------------------
Three commenters suggested an implementation timeframe of at least
one year.\969\ One commenter explained that the changes to the round
lot definition would require programming changes by the exclusive SIPs
and the market participants that receive SIP data, as well as testing
of the changes at the exchanges, exclusive SIPs and customer
levels,\970\ and that it would likely take longer than one year for the
exclusive SIPs and exchanges to implement the proposed odd-lot
changes.\971\ The Operating Committees for the Equity Data Plans stated
that the implementation timeframe for the exclusive SIPs would likely
extend beyond one year due to, among other things, ``the time needed
for system design, [to] procure necessary equipment, and accommodate
industry testing.'' \972\ Another commenter stated that the proposed
90-day timeframe was too aggressive, did not consider ``technical
realities,'' and suggested an implementation period of at least one
year.\973\
---------------------------------------------------------------------------
\969\ See NYSE Letter I at 8; CTA, CQ, UTP Plans Operating
Committees Letter at 3; Nasdaq Letter I at 3.
\970\ See NYSE Letter I at 7.
\971\ See NYSE Letter I at 8.
\972\ CTA/CQ/UTP Plans Operating Committees Letter at 2, 3.
\973\ See Nasdaq Letter I at 3.
---------------------------------------------------------------------------
In light of the comments, the Commission is modifying the
compliance date for the round lot and odd-lot information definitions.
For implementation of the round lot definition \974\ and the round lot
indicator,\975\ the compliance date will be the first business day of
November 2025. The Commission calculated this deadline based on two
main factors. First, the compliance date is approximately 12 months
after the effective date, which is consistent with what commenters
suggested was necessary for systems changes and testing. Second, the
compliance date provides sufficient time for any exchanges that have
defined round lots in their rules to file proposed rule changes
pursuant to section 19(b) of the Exchange Act \976\ and rule 19b-4
\977\ thereunder to reflect the new round lot definition.\978\
---------------------------------------------------------------------------
\974\ Rule 600(b)(93).
\975\ Rule 600(b)(89)(i)(E) and Rule 600(b)(89)(iv).
\976\ 15 U.S.C. 78s(b).
\977\ 17 CFR 240.19b-4.
\978\ See Proposing Release, supra note 11, at 80300 n.408.
---------------------------------------------------------------------------
The compliance date for the odd-lot information definition \979\
and Rule 603(b)(3) (with respect to the requirement that every national
securities exchange on which an NMS stock is traded and national
securities association must make available to the exclusive SIPs all
data necessary to generate odd-lot information, and the collection,
consolidation and dissemination of odd-lot information by the exclusive
SIPs) will be the first business day of May 2026, which is
approximately 18 months after the effective date of the Adopting
Release. The Commission is providing a modified compliance date for the
odd-lot information definition, consistent with what industry comment
suggested was necessary for technical and operational
requirements,\980\ due to several factors. First, the exclusive SIPs
will likely have to make more changes to their systems to accommodate
the odd-lot information definition than to implement the round lot
definition. Specifically, the exclusive SIPs will need to collect more
data, consolidate it, and disseminate it as odd-lot information. In
addition, the exclusive SIPs will need to calculate and disseminate the
BOLO. The Commission continues to believe that both the changes to the
odd-lot information definition and the dissemination of the BOLO are
independently important, and the additional time allotted to comply
with the odd-lot information definition addresses the concerns from
commenters regarding the complexity or operational risks that may arise
with making odd-lot information changes in a compressed timeline.
Second, the effective national market system plan(s) may also need to
assess whether plan amendments will be necessary to conform such plans
to the odd-lot information definition, and to file any such amendments
with the Commission
[[Page 81682]]
pursuant to rule 608. Finally, market participants may need to update
their systems that accept SIP data to reflect odd-lot information.
---------------------------------------------------------------------------
\979\ Rule 600(b)(69).
\980\ See supra notes 969 and 973 (suggesting an implementation
process of approximately one year).
---------------------------------------------------------------------------
Accordingly, extending the compliance deadlines for the
implementation of the round lot and odd-lot information definitions
will address the concerns raised by commenters and provide additional
time for market participants to make the changes necessary to implement
the definitions.
VII. Economic Analysis
A. Introduction
The most common method of trading in NMS stocks by registered
exchanges today is the limit order book matching system, a mechanism
that securities exchanges use to bring together orders of multiple
buyers and sellers of securities and have those orders interact. It
acts as a central hub where participants' priced buy and sell orders
can be ranked, displayed, and matched based on programmed rules
established by the providing registered exchange. As such, the limit
order book matching system facilitates efficient and competitive
markets.\981\
---------------------------------------------------------------------------
\981\ See Anthony Clarke, Demystifying the Central Limit Order
Book (CLOB): Everything You Need to Know (Apr. 21, 2023), available
at https://www.nasdaq.com/articles/demystifying-the-central-limit-order-book-clob-everything-you-need-to-know.
---------------------------------------------------------------------------
Imagine an order book in which a buyer's or seller's order could be
displayed at any pricing increment, no matter how small. In this
scenario, assume a liquidity provider wants to buy a stock. The
provider sees the book with the prices at which others are willing to
buy. Because in this hypothetical market there are no restrictions on
an entry price point, the liquidity provider can jump ahead of those
other providers by offering to buy at a price that is infinitesimally
higher. This is what is known as ``pennying.'' \982\ The problem with
pennying is that it creates a disincentive for liquidity providers to
post buy or sell orders, because they know that a second trader can
step ahead with an infinitesimally better price. This leads to lower
priced offers to buy and higher priced offers to sell--namely a wider
quoted bid-ask spread.
---------------------------------------------------------------------------
\982\ See supra section I.A.1.
---------------------------------------------------------------------------
Recognizing this market failure, the Commission in 2005 adopted
\983\ a market-wide requirement that venues could not display, rank, or
accept orders in increments less than a penny.\984\ The 2005 adoption
of Rule 612 limited the scope of pennying, but it did so at the
inevitable cost of introducing a floor, namely one cent, below which
the quoted bid-ask spread could not fall.
---------------------------------------------------------------------------
\983\ See generally, Regulation NMS Adopting Release, supra note
4.
\984\ Specifically, preexisting Rule 612 of Regulation NMS
prohibited a national securities exchange, national securities
association, ATS, vendor, or broker or dealer from displaying,
ranking, or accepting quotations, orders, or indications of interest
in any NMS stock priced in an increment smaller than $0.01 if the
quotation, order, or indication of interest is priced equal to or
greater than $1.00 per share. If the quotation, order, or indication
of interest is priced less than $1.00 per share, the minimum pricing
increment is $0.0001.
---------------------------------------------------------------------------
Though a minimum tick is necessary, placing a floor on the spread
introduces distortions into the market. The price of liquidity will be
artificially high for some stocks, leading to a surplus, similar to a
goods market for which prices were artificially high. This creates
rents which accrue to some market participants at the expense of
others. By reducing the minimum pricing increment for a defined subset
of stocks, the adopted amendments to Rule 612 free the price of
liquidity from its current constraint, allowing it to approach its
natural level. At the same time, as described in greater detail below,
the adopted amendments maintain a minimum (but smaller) pricing
increment necessary for the proper functioning of financial markets'
limit order books.
Freeing the spread from the binding constraint of one penny will
bring a number of benefits, including lower transaction costs. For some
stocks currently constrained at a penny, the spread will, under the
amended rules, at times be a half-penny, a substantial reduction in the
quoted price of accessing liquidity. This reduction, while beneficial,
brings into the spotlight the cap on the access fee, which has been
0.30 cents. Absent a reduction in the maximum access fee, a round-trip
buy and sell for stocks quoted at the new half-penny tick would require
paying more in fees (0.60 cents) than in the spread itself (0.50
cents).
The practice of charging at or near the access fee cap has
persisted over time. Regulation NMS establishes the NBBO. Because the
NBBO is protected,\985\ many exchanges charge the maximum amount
allowed to access the quote. This allows the exchange to subsidize
liquidity providers with a rebate, reducing spreads (to acquire more
volume, due to traders' need to access the protected quote). While the
quoted spread may be lower, the cost to investors is not; this is
because gains from the lower spread are counteracted by the access fee.
On the other hand, the high access fee and rebate can lead to a loss of
price coherence when the spread is less than twice the fee. For stocks
that remain tick constrained, as some may, the rebate distorts the
supply of liquidity. Finally, fees and rebates that are high as a
percentage of the quoted spread introduce complexity, and potential
conflicts of interest. Lowering the access fee to 10 mils restores
price coherence and alleviates these costs.
---------------------------------------------------------------------------
\985\ See supra note 42 and accompanying text discussing and
defining protected quotes.
---------------------------------------------------------------------------
The Commission is also requiring that these fees and rebates be
determinable at the time of trade execution. Opacity and complexities
in current exchange fees and rebates make these more distortive than
otherwise.\986\ With new Rule 610(d), the Commission is taking an
incremental step in ameliorating the opacity in fees and rebates,
reducing information asymmetries and lessening the potential for agency
conflict between brokers and their customers.
---------------------------------------------------------------------------
\986\ As discussed in sections VII.D.2, VII.D.3, and VII.E.1,
fees and rebates create a potential conflict for a broker in
situations where transaction fees, which are paid by the broker,
potentially conflict with execution quality, which is incurred by
the customer. This conflict, if acted on, can lead to inefficient
order routing and worse transaction outcomes for customers; it can
also lead to an inefficient incorporation of information into stock
prices, harming market efficiency.
---------------------------------------------------------------------------
Finally, the Commission has accelerated the implementation of the
round lot, and odd-lot information definitions while providing more
time for the necessary systems changes to implement the definitional
changes than what was proposed. These amendments will allow the
benefits of these rules to accrue to market participants in a timely
manner.
Below, we explain why these amendments increase efficiency and
competition and bring benefits that will accrue to the broad range of
participants in U.S. equity markets. We also discuss the costs of these
amendments. The Commission has considered the economic effects of the
amendments and, wherever possible, the Commission has quantified the
likely economic effects of the amendments.\987\ The Commission is
providing both a qualitative assessment and quantified estimates of the
potential economic
[[Page 81683]]
effects of the amendments where feasible. The Commission incorporated
data and other information to assist it in the analysis of the economic
effects of the amendments. However, as explained in more detail below,
the Commission is unable to quantify certain economic effects because
the Commission does not have, and in certain cases cannot reasonably
obtain, data that may inform the Commission on certain economic
effects. Further, even in cases where the Commission has data, it is
not practicable to quantify certain economic effects due to the number
and type of assumptions necessary, which render any such quantification
unreliable. Our inability to quantify certain costs, benefits, and
effects does not imply that such costs, benefits, or effects are less
significant.
---------------------------------------------------------------------------
\987\ Section 3(f) of the Exchange Act requires the Commission,
whenever it engages in rulemaking and is required to consider or
determine whether an action is necessary or appropriate in the
public interest, to consider, in addition to the protection of
investors, whether the action will promote efficiency, competition,
and capital formation. Additionally, section 23(a)(2) of the
Exchange Act requires the Commission, when making rules under the
Exchange Act, to consider the impact such rules will have on
competition. Exchange Act section 23(a)(2) prohibits the Commission
from adopting any rule that would impose a burden on competition not
necessary or appropriate in furtherance of the purposes of the
Exchange Act.
---------------------------------------------------------------------------
B. Broad Economic Considerations
1. Liquidity and Spread
A key component of market liquidity is the limit order book.
Liquidity providers submit limit orders to buy (``bid'') and sell
(``ask'') stock at specified prices and quantities. Liquidity demanders
trade against these limit orders. The quoted (bid-ask) spread for a
stock is the difference between the lowest displayed ask price and the
highest displayed bid price.\988\ As discussed in the Proposing
Release, standard economic theory suggests that liquidity providers in
a competitive market will compete to provide liquidity until the
spread--i.e., their compensation for providing liquidity--is equal to
the break-even point given the costs of liquidity provision.\989\
Absent fees, rebates, and a minimum pricing increment, this break-even
point for liquidity provision represents the lowest bid-ask spread at
which liquidity providers (as a whole) are willing to provide liquidity
(hereinafter ``economic spread'').\990\
---------------------------------------------------------------------------
\988\ Investors can also execute trades on other ``dark'' venues
that do not display quotes. But because quotes are not displayed on
these venues, the investor could not be certain of the execution
price or of the number of shares available. See Proposing Release,
supra note 11, at 80287 (addressing trading centers that do not
display protected quotes).
\989\ Proposing Release, supra note 11, at 80309 n.483. See also
Jonathan Brogaard & Corey Garriott, High-Frequency Trading
Competition, 54 J. Fin. & Quantitative Analysis 1469 (2019)
(documenting that as more high-frequency liquidity providers enter
the market, spreads decrease until they converge to competitive
levels).
\990\ Although the Proposing Release did not use the phrase
``economic spread,'' the release employed the same concept in
multiple places. See, e.g., Proposing Release, supra note 11, at
80317 (``In a competitive market, and in the absence of rebates or
other price distortions, the prevailing bid or ask price would be
the feasible price equal to just worse than the price that equates
liquidity supply and demand.''). In a number of places where the
release employed the concept of economic spread, it arose in
discussions about a stock that would trade at a given price or
spread absent the tick size. See Proposing Release, supra note 11,
at 80304, 80309, and 80317. The spread is composed of several
elements: adverse selection, inventory risk, and processing costs.
See Proposing Release, supra note 11, at 80304 n.447 and 80321. See
generally, Roger D. Huang & Hans R. Stoll, The Components of the
Bid-Ask Spread: A General Approach, 10 Rev. Fin. Stud. 995 (Winter
1997). As explained in the Proposing Release, supra note 11, at
80304 and n.447, the spread is unlikely to ever be zero due to
inventory costs, adverse selection risks, the direct costs
associated with providing liquidity, and trading rules meant to
prevent the locking and crossing of markets. See P.C. Kumar, Bid-Ask
Spreads in U.S. Equity Markets, 43 Q. J. Bus. & Econ 85 (2004).
---------------------------------------------------------------------------
2. Economics of Minimum Pricing Increments
When a market has a minimum pricing increment (hereafter ``tick
sizes'' or just ``ticks''), liquidity providers quote bid and ask
prices that are discrete whole number multiples of that tick. For
example, if the tick is a penny, then a liquidity provider quotes
prices that are a multiple of a penny, such as $10.00 or $10.01, but
not $10.015. There may, however, be a liquidity provider willing to
quote an ask of $10.015 and a bid of $10.005. Were this liquidity
provider to be allowed to do so, the stock would have a spread $0.01
(i.e., the difference between the lowest ask price and bid price).
However, in the presence of the $0.01 tick, liquidity providers will
quote at the best feasible ask price above $10.015, which is $10.02,
and the best feasible bid price below $10.005, which is $10.00.\991\
Consequently, the stock's quoted spread would be $0.02 instead of
$0.01, twice as wide than it would otherwise be.
---------------------------------------------------------------------------
\991\ As discussed in the Proposing Release, supra note 11, at
80309 nn.483-484 and accompanying text, this assumes that stock
prices are expected to revert to the next worse level. This may
occur because standard economic theory suggests that in a
competitive market liquidity providers will compete to provide
liquidity until the spread--i.e., their compensation for providing
liquidity--is equal to the break-even point for liquidity provision.
See also Jonathan Brogaard & Corey Garriott, High-Frequency Trading
Competition, 54 J. Fin. & Quantitative Analysis 1469 (2019)
(documenting that as more high-frequency liquidity providers enter
the market, spreads decrease until they converge to competitive
levels). The range of infeasible quoting prices narrows somewhat in
the presence of rebates for liquidity providers. section VII.B.3
discusses these effects.
---------------------------------------------------------------------------
Tick sizes present an economic tradeoff. As discussed in the
Proposing Release, in determining what tick size is optimal for any
given stock, there is a tradeoff between price competition on one hand,
and incentives for liquidity provision on the other.\992\ A smaller
tick allows liquidity providers to better compete on price which can
lead to narrower spreads, reducing costs for investors. On the other
hand, a smaller tick can also lead to pennying. Pennying increases
adverse selection costs for slower liquidity providers by making it
more likely that they trade when prices are moving in an unfavorable
direction relative to their positions.\993\ To compensate for these
costs, liquidity providers may post less aggressive quotes--lower bid
prices and higher ask prices--resulting in a wider quoted spread and
worse liquidity.\994\ Both price competition and adverse selection from
pennying lie on a continuum.\995\ As explained in infra section
VII.D.1, the degree to which pennying versus price competition
dominates in determining whether increasing the tick will improve
market quality depends on the relation between the tick and the spread.
The greater the tick is in relation to the spread, the greater the
effect of price competition, and the lower the risk of pennying.\996\
Accordingly, the Commission defines a stock to be tick-constrained if
there is a reasonable probability that the stock would otherwise trade
with a spread less than the tick size in the course of normal trading,
were it allowed to do so, or one for which the tick is a substantial
portion of the quoted spread.\997\ That is,
[[Page 81684]]
while tick constrained stocks are not the only ones to potentially
benefit from a reduction in the tick size, they are the ones most
clearly likely to do so.\998\
---------------------------------------------------------------------------
\992\ Proposing Release, supra note 11, at 80305.
\993\ Id., at 80305 n.481 and accompanying text.
\994\ Id. at 80305-06. Pennying is defined in the Proposing
Release as occurring when a market participant gets to the front of
the limit order queue by posting economically trivial price
improvement. Id. at 80306 n.459. One commenter also described the
economics of pennying using option theory and the Commission agrees
with this characterization. See Harris Letter at 6. According to the
commenter, pennying results in a payoff structure that has unlimited
potential upside while the downside is capped. By pennying, a fast
trader jumps to the front of the queue and therefore has a high
chance of executing his trade and capturing the upside if prices
move favorably. If prices move in an unfavorable direction, the fast
trader can unwind his position against the slower liquidity supplier
(whom he undercut); in this case, the cost to the fast trader--i.e.,
the cost of the option--is only one tick. The fast trader thereby
captures value from the liquidity supplier and hence discourages
slow traders from offering liquidity. A small tick means that the
cost of pennying is low, which results in more pennying and thus
less incentive for liquidity provision. See also Lawrence E. Harris,
Minimum Price Variations, Discrete Bid-Ask Spreads, and Quotation
Sizes, 7 Rev. Fin. Stud. 149 (1994); Anne Dyhrberg, et al., When
Bigger is Better: The Impact of a Tiny Tick Size on Undercutting
Behavior, 58 J. Fin. & Quantitative Analysis (2023) (Dyhrberg et
al.).
\995\ Proposing Release, supra note 11, at 80305 n.458.
\996\ See infra section VII.D.1.b.i.
\997\ The Commission's definition of tick-constrained in this
release eliminates an unnecessary distinction drawn between tick-
constrained and near-tick-constrained stocks that appears in the
Proposing Release. Specifically, in the Proposing Release's economic
analysis, the Commission stated that it considered the term ``tick-
constrained'' to apply to ``stocks that would otherwise trade with a
spread less than the tick size, were they allowed to do so.''
Proposing Release, supra note 11, at 80304. That economic analysis
also stated that a ``near-tick-constrained'' stock was ``one that
has a reasonable probability of becoming tick-constrained in the
course of normal trading, or one for which the tick is a substantial
portion of the spread.'' Id. Given the economics of minimum pricing
increments discussed in this section, distinguishing near-tick-
constrained stocks from tick-constrained stocks is unnecessary. The
Proposing Release's discussion of these terms in its economic
analysis did not meaningfully differentiate between the effects on
stocks in each of these groups. As a result, applying the singular
term tick-constrained avoids confusion and streamlines the
discussion. In addition, the Proposing Release's empirical analysis
employed separate numerical definitions for tick-constrained and
near-tick-constrained stocks. Compare id. at 80268 n.17 (tick-
constrained stocks are those with a Time Weighted Average Quoted
Spread less than .011) with id. at 80304 n.449 (near-tick-
constrained stocks are those with a Time Weighted Average Quoted
Spread between .011 and .02). These numerical definitions served as
proxies for drawing distinctions in the Proposing Releases'
empirical analysis. See, e.g., id. at 80304 nn.448-449; at 80308
n.473 and accompanying text; and at 80319 n.549. Although we
continue to include specific explanations of what stocks are
included in each of our quantitative analyses where relevant, to
further simplify the discussion in the economic analysis, we do not
use the definitions employed in the empirical analysis more broadly.
\998\ See Proposing Release, supra note 11, at 80309 and the
discussion accompanying nn.474-477. See also CCMR Letter at 18 (``an
MPI that is too wide may set an artificial constraint on permissible
bids and offers, which can result in an unnecessarily wide spread
that can also increase transaction costs for investors'').
---------------------------------------------------------------------------
Benefits from a reduction in the tick size come in the form of
higher market quality and lower transaction costs to investors. As
explained above, relaxing the tick constraint (the price floor on
liquidity) directly allows competition for market orders. Moving toward
a more competitive market reduces distortions and economic rents.
As a general matter, a liquidity provider is incentivized to get
its quote to the front of the queue (i.e. establish price/time priority
on an order book).\999\ This is because stock exchange priority rules
give greater priority to better priced orders and generally factor
order entry time into the priority of limit orders at the same price.
When the NBBO is equal to the tick, liquidity providers cannot
establish price priority (other than by crossing the spread) \1000\
because there are no price points at which to do so.\1001\ Because
liquidity providers cannot establish price priority when the NBBO
spread is one tick, establishing time priority becomes more
important.\1002\ Consequently, an environment where stocks are tick-
constrained with artificially wider spreads and longer order queues
tends to favor traders who are better able to establish positions more
quickly so they can be at the front of the queue. Traders who are at
the back of the queue face slower executions and the risk of not being
executed against at all (a lower fill rate). In the latter case, they
will need to resubmit an order when the market has moved in an
unfavorable direction, increasing transaction costs. Adverse selection
amplifies these costs: the orders of slower traders are most likely to
be executed when such an execution is unfavorable to them and least
likely when they would be favorable. For example, a sell order at the
back of the queue will tend to be filled when there are many buy
orders, which tend to increase the price, implying that selling is
disadvantageous.
---------------------------------------------------------------------------
\999\ Material presented in this paragraph was discussed in the
Proposing Release, supra note 11, at 80309 and nn.478-481.
\1000\ ``Crossing the spread'' refers to switching from posting
a (non-marketable) limit order to sending a market order. For
example, if the national best bid were $10.00 and the national best
offer were $10.02, a limit order to buy, if executed, would entail
paying $10.00 for the security. However, a market order to buy would
entail pay $10.02, in other words, it would have crossed the spread
of $0.02. In the most common case of maker-taker, the difference
between the market and limit order is even greater because the buyer
using a market order would pay $10.02 plus the fee, whereas the
buyer using a limit order would pay $10.00 minus the rebate.
\1001\ The liquidity provider could submit an order at an
inverted exchange, though this is an inefficient solution. See
section VII.C.2.c.
\1002\ Under typical exchange rules, an order with time priority
is executed first when multiple orders are at the best price,
regardless of how many orders are at the best price. In longer order
queues, liquidity-providing orders deeper in the queue, which do not
have time priority, are less likely to be filled in a timely manner
and, conditional on being filled, the probability of the order
having been adversely selected tends to be greater compared to
orders with greater fill priority. Typically, liquidity providers
compete to gain priority over other resting orders by quoting a
better price, but tick-constraints make doing so difficult. In the
case when the spread is constrained to a single tick, it would be
impossible to improve on the displayed price without locking
markets. For tick-constrained stocks, when the quoted spread may be
greater than a single tick, improving the price by an entire tick
may be too much in the sense that doing so may narrow the spread
beyond what the liquidity providers could tolerate. A narrower tick
de-emphasizes time priority on a stock exchange by making it easier
to compete on price. See Edwin Hu, et al., Tick Size Pilot and
Market Quality (DERA White Paper, Jan. 31, 2018), available at
https://www.sec.gov/dera/staff-papers/white-papers/dera_wp_tick_size-market_quality; and Todd G. Griffith & Brian S.
Roseman, Making Cents of Tick Sizes: The Effect of the 2016 U.S. SEC
Tick Size Pilot on Limit Order Book Liquidity, 101 J. Banking Fin.
104 (2019).
---------------------------------------------------------------------------
To summarize, current wider quoted spreads mean greater cost to
liquidity demanders and greater revenue to liquidity providers.\1003\
An artificially wide spread, due to a price floor imposed by the tick
constraint, effectively subsidizes liquidity provision. Because there
is an increased incentive to provide liquidity via limit orders, queues
of limit orders tend to be longer, and wait times to get a limit order
executed also tend to be longer. This makes it more likely that the
market moves away from an investor's limit order and leads to lower
overall fill rates for limit orders.\1004\ Thus the floor on liquidity
leads to rents accruing to fast liquidity providers \1005\ at the
expense of slower ones as well as liquidity demanders.\1006\
---------------------------------------------------------------------------
\1003\ Market participants can use inverted exchanges or ISOs to
help ameliorate some of the negative effects of tick size
constraints.
\1004\ See, e.g., Barbara Rindi & Ingrid M. Werner, U.S. Tick
Size Pilot (working paper Mar. 4, 2019), available at https://ssrn.com/abstract=3041644 (retrieved from SSRN Elsevier Database);
Mao Ye & Chen Yao, Tick Size Constraints, Market Structure and
Liquidity (working paper Dec. 26, 2019), available at https://ssrn.com/abstract=2359000 (retrieved from SSRN Elsevier database);
Phil Mackintosh, Why Ticks Matter, NASDAQ (May 19, 2022), available
at https://www.nasdaq.com/articles/why-ticks-matter; and MEMX, Tick-
Constrained Securities (Aug. 2021) (``MEMX Report''), available at
https://memx.com/wp-content/uploads/MEMX-Market-Structure-Report-Tick-Constrained-Securities.pdf.
\1005\ This phenomenon is sometimes referred to as excessive
intermediation. In this context, excessive intermediation refers to
excessive quoting in sufficiently liquid securities in order to
profit from the tick-constraint-induced price floor on liquidity,
which crowds out investors from being able to supply liquidity. Such
price floors can increase quoting activity from high-frequency
traders looking to earn the artificially high spread. Because
profiting off of the spread is easiest when the marketable orders
filled are small, obtaining high priority in the queue at each tick
is essential to such strategies. High-frequency, proprietary traders
are generally better able to obtain such priority, and consequently
investors may have less opportunity to profitably fill their trades
using limit orders. Rebates on limit orders further increase the
incentives of these traders to engage in such intermediation,
thereby exacerbating the problem.
\1006\ In support of this point, one commenter stated that the
subsidization of liquidity providers resulting from the tick
constraint leads to greater competition on the basis of speed to
provide liquidity, which increases complexity and related costs to
investors; See Budish Letter at 4.
---------------------------------------------------------------------------
3. Economics of Access Fees
Trading venues can choose to charge an access fee, or pay a rebate,
to their participants--liquidity providers and liquidity takers--who
trade at their venue. The trading venue can further choose to levy the
fee (or pay the rebate) on either the liquidity taker or liquidity
provider, or on both. As discussed in infra section VII.C.2.b, the most
common fee structure is maker-taker, in which liquidity takers are
assessed an access fee and liquidity providers are paid a rebate, which
is typically funded through the access fee. That is, for a buy order
the liquidity taker pays the price plus the access fee. For a sell
order, the liquidity taker receives the price, less
[[Page 81685]]
the fee. Assuming the broker-dealer is the principal to the trade, then
the economic price of accessing or providing liquidity would be
equivalent to the displayed nominal price net of the applicable fees.
If the broker-dealer is an agent, then there maybe a wedge between
economic price of accessing or providing liquidity and the price net of
the fee and rebate. It is possible that the fee and rebate may be
passed on directly to the customer. The fee and rebate may be passed on
indirectly and in part through fees, commissions, or as part of a
bundle of services to the customer.
Section VII.C.2 describes the current market structure as it
relates to access fees and rebates. A key feature of the current market
structure is that many exchanges charge at the current cap and pay out
nearly all of the fee as a rebate. For this reason, in practice the
Commission expects access fees to be near the cap under the amended
rule, just as they are near the pre-existing cap under the current
structure. As discussed in the Proposing release, several basic
economic considerations are among those governing the analysis of
access fees. First, access fees should be such that net and quoted
prices satisfy coherence.\1007\ Second, under simplifying assumptions,
fees and rebates are approximately neutral provided that the stock is
not tick constrained,\1008\ although outside of those simplifying
assumptions lowering the access fee cap can have additional benefits as
discussed in section VII.D.2.d. Finally, for tick constrained stocks,
access fees and rebates can distort liquidity supply and demand,
increasing transaction costs for investors.\1009\
---------------------------------------------------------------------------
\1007\ See Proposing Release, supra note 11, at 80348 (``Net and
nominal price rankings are coherent if sorting trading venues on the
competitiveness of their nominal quoted prices yields the same
ordering as sorting on prices net of fees and rebates.'').
\1008\ Proposing Release, supra note 11, at 80328.
\1009\ Id.
---------------------------------------------------------------------------
In response to the Proposing Release, the Commission received
extensive comment regarding the role of fees and rebates on the supply
of and demand for liquidity.\1010\ Below, to address comments, the
Commission supplements its discussion on how access fees and fee-funded
rebates may affect trading in the presence of the commonly used maker-
taker fee structure.\1011\ This section addresses certain aspects of
fees and rebates in developing a basic framework for evaluating the
principle economic effects and responding to comments. The remaining
aspects and effects are considered in sections VII.C.2, VII.D.2, and
VII.D.3.
---------------------------------------------------------------------------
\1010\ Specifically, commenters had different views on whether
reducing access fees and rebates will adversely affect the provision
of liquidity on exchanges, either generally or for particular
categories of stocks. Compare Cboe Letter II at 9; Cboe Letter III
at 6, 8; Cboe Letter IV at 3, 5; CCMR Letter at 27; IEX Letter I at
2; IEX Letter IV at 18, 21; IEX Letter V at 4; IEX Letter VI at 7;
Nasdaq Letter I at 2, 20, 22, 25; Nasdaq Letter II at 3, 6; Nasdaq
Letter III at 2-3; Themis Letter at 7-8; Virtu Letter II at 7-8.
Commenters likewise had differing perspectives on whether reducing
access fees and rebates will reduce overall transaction costs,
thereby increasing demand for liquidity, or cause offsetting costs
related to, e.g., wider spreads, volatility, or a less
representative NBBO (which could reduce demand for liquidity).
Compare Cboe Letter III at 5-6; Cboe Letter IV at 5; CCMR Letter at
27; IEX Letter I at 26; IEX Letter IV at 16, 22-23; Nasdaq Letter I
at 2, 20, 22-24; Nasdaq Letter II at 4-7; Nasdaq Letter III at 5;
Themis Letter at 7; Virtu Letter II at 8. As one example of this
debate, Nasdaq identifies a ``vicious cycle'' that could result from
a reduction in the minimum access fee, whereas IEX identifies a
``virtuous cycle'' from the identical change. See Nasdaq Letter II
at 6; IEX Letter I at 26. See section VII.D.2 for a response to
these comments on the effect of the reduction in the access fee cap
on liquidity and transaction costs.
\1011\ The economic theory laid out below allows the Commission
to create a common framework for the competing claims of commenters
and to disentangle the complex forces at work in determining
spreads. While the Commission has supplemented this discussion from
the Proposing Release, the essential point in this framework--the
equilibrium resulting from the supply and demand for liquidity,
modified as necessary for the presence of a minimum tick--was
discussed throughout the Proposing Release. See, e.g., Proposing
Release, supra note 11, at 80228-29, 80317, 80321, 80336, 80338.
---------------------------------------------------------------------------
a. Liquidity With Access Fees and Rebates
In the absence of ticks, the market for liquidity, discussed in
section VII.B.1, may generally be represented by an economic model of
supply and demand, as shown below in panel A of figure 1.\1012\ The
vertical axis represents the price of liquidity, here the quoted half-
spread (i.e. because the liquidity taker is not typically on both sides
of the trade, we use the quoted half-spread to measure price of
liquidity \1013\), while the horizontal axis represents the quantity of
liquidity.\1014\ Liquidity providers supply liquidity, and the supply
curve is upward sloping because liquidity providers are willing to
supply more liquidity when the price of liquidity is higher. Liquidity
takers demand liquidity, and the demand curve is downward sloping
because liquidity takers demand less liquidity at higher prices of
liquidity (i.e., they trade less when they have to pay higher
transaction costs). The supply and demand curves intersect at the point
where the amount of liquidity supplied equals the quantity demanded,
which indicates the equilibrium price and quantity of liquidity in the
market.
---------------------------------------------------------------------------
\1012\ This model presents an abstraction of the market for
liquidity. As explained in section VII.B.2, the willingness to quote
a bid or offer depends in part on the degree of adverse selection in
the market which in turn depends on the tick size. The key point in
this section is that, for stocks that have a spread that is
sufficiently wide, liquidity providers are indifferent between
receiving compensation in the form of spread or in the form of a
rebate; similarly, liquidity demanders are indifferent between
paying the spread or access fee, and thus fees and rebates tend to
be neutral assuming a spread that is sufficiently wide. This point
is unaffected by the presence of adverse selection arising from a
tick that may be too narrow.
\1013\ The price of liquidity is represented by the quoted half-
spread because the half-spread represents the price which liquidity
takers must pay for the immediacy of executing their trade while
liquidity providers stand to capture the half-spread.
\1014\ Figure 1 shows the supply and demand for liquidity, with
the quoted half-spread as representing the price of liquidity on the
y-axis. This should not be confused with supply and demand for
shares of the stock, where the price of the stock would be on the y-
axis.
---------------------------------------------------------------------------
[[Page 81686]]
[GRAPHIC] [TIFF OMITTED] TR08OC24.001
[[Page 81687]]
Consider next the effect on liquidity of a 30 mils access fee used
to fund a 30 mils rebate.\1015\ With a rebate of 30 mils, liquidity
providers who submit buy orders are willing to increase their bid price
by 30 mils, while liquidity providers who submit sell orders are
willing to lower their ask price by 30 mils. For this reason, the bid-
ask spread narrows by 60 mils, and the quoted half-spread (i.e., price
of liquidity) narrows by 30 mils. Accordingly, in panel B of figure 1,
the supply curve for liquidity shifts down by 30 mils. Likewise, the 30
mils access fee acts as a tax on liquidity takers. This means that, for
the same amount of liquidity, liquidity takers will reduce the price
they are willing to pay by 30 mils to account for the access fee (since
their net cost to take liquidity is the price they pay plus the access
fee). In panel B of figure 1, this effect is represented by the
liquidity demand curve shifting down by 30 mils. Because both the
supply curve and the demand curve shift down by 30 mils, they continue
to intersect at the same quantity of liquidity.\1016\ That is, the
equilibrium amount of liquidity remains unchanged, but the displayed
price is 30 mils lower. The net cost to take liquidity is not affected
since it equals the price of liquidity plus the 30 mils access fee;
\1017\ similarly, the net proceeds from providing liquidity are not
affected since they equal the price of liquidity plus the 30 mils
rebate.\1018\ Thus, in the absence of frictions (e.g., discrete prices,
minimum pricing increments, or agency problems), the level of fees and
rebates (when fees and rebates are equal in size) does not affect the
total costs of trading.\1019\ As one commenter put it, ``when liquidity
suppliers are subsidized at the cost of liquidity takers, spreads
decline. If they did not, everyone would want to be a liquidity
supplier, and no trade would occur. So, maker-taker pricing created
narrower quoted spreads on average, but it does not affect the net cost
of providing liquidity.'' \1020\
---------------------------------------------------------------------------
\1015\ This level of access fee and rebate is similar to current
fees and rebates on maker-taker lit exchanges and may vary based on
pricing tier; see infra section VII.C.2.b. To illustrate the salient
economic points, this discussion assumes that liquidity providers
and demanders know what the resulting access fees and rebates from a
transaction will be. As discussed below in sections VII.C.2.b and
VII.D.3, in the baseline this is only ever approximatively true
since fees and rebates are often determined using current and future
volumes. But as long as market participants are able to approximate
their fees and rebates, then they will generally behave as described
in this paragraph--liquidity providers will adjust their quotes on
account of the expected rebate, and liquidity demanders will adjust
the quoted price they are willing to pay on account of the expected
fee.
\1016\ If the demand curve were vertical (namely if liquidity
demanders were not sensitive to price), the curve would not shift.
However, the conclusions would be the same in that the supply curve
shift would cause the same quantity to be supplied at a lower price.
\1017\ The concept that net cost (or net spread) is the correct
way to measure the cost of liquidity is supported by basic economics
and by commenter statements. See Citigroup Letter at 6 (stating,
``Many of CGMI's institutional clients are increasingly measuring
their execution costs all-in, inclusive of exchange fees.'').
\1018\ This discussion refers to the cost to take liquidity and
proceeds from providing liquidity at the time of the execution of
the trade.
\1019\ The term frictions here refers to factors that prevent
prices from perfectly reflecting the forces of liquidity supply and
demand. It does not imply that a frictionless market is the optimal
market construct. As discussed throughout this release, a tick size
that is too small creates pennying concerns which can harm market
quality outcomes. See infra section VII.D.1.b for additional
discussion.
\1020\ Harris Letter at 2-3, describing the equilibrium spreads
model (citing Kalman J. Cohen, et al., Transaction Costs, Order
Placement Strategy and Existence of the Bid-Ask Spread, 89 J. Pol.
Econ. 287 (1981)).
---------------------------------------------------------------------------
Academic work on the effect of a fee change on the Toronto Stock
Exchange supports the model.\1021\ In 2005, the exchange began offering
a rebate to liquidity suppliers in a pre-defined subset of securities.
For securities in which the total fee remained constant but was split
into a maker rebate and a taker fee, the authors find that quoted
spreads narrow, but the net spread--which includes the quoted spread
and the take fee--did not change.\1022\
---------------------------------------------------------------------------
\1021\ See Katya Malinova & Andreas Park, Subsidizing Liquidity:
The Impact of Make/Take Fees on Market Quality, 70 J. Fin. 509
(2015).
\1022\ The authors also present evidence suggesting that adverse
selection costs decreased with the introduction of the maker-taker
model by increasing retail trader participation. See Malinova and
Park (2015), supra note 1021. In the United States, most retail
orders in NMS stocks are handled by wholesalers, who execute a large
majority of the dollar volume of the retail orders they handle via
internalization. See, e.g., Lewis Letter attached to Virtu Letter II
at p 8-12 and 40-45. Consequently, should there be a decrease in
retail participation, we do not expect this to cause an increase in
adverse selection on exchanges, because so much of retail order flow
passes through a wholesaler before being executed.
---------------------------------------------------------------------------
b. Liquidity With Ticks, Access Fees, and Rebates
Section VII.B.3.a shows that the quoted spread adjusts to a fee and
rebate by narrowing by the amount of the fee and the rebate. The supply
and demand curves in section VII.B.3.a are continuous, whereas in fact
displayed liquidity has discrete price points, namely ticks (the focus
of section VII.B.2). In the presence of ticks, access fees and rebates
need no longer be neutral, namely the quoted spread may not adjust in
the same seamless way to the presence of a rebate. The basic intuition
of section VII.B.3.a states that a liquidity provider is indifferent
between receiving compensation from the spread and compensation from
the rebate. If a rebate is offered, the liquidity provider in a
competitive market responds by accepting a lower spread. However, if
the quoted spread is already at its floor, as specified by the tick, it
is not possible to further lower the spread. In this case, unlike in
section VII.B.3.a, the rebate and access fee are therefore not neutral.
Rather, the floor creates rents that in this case accrue to those
liquidity suppliers that are able to get to the front of the queue the
fastest. These rents are earned at the expense of liquidity takers and
slower liquidity providers.\1023\
---------------------------------------------------------------------------
\1023\ See Budish Letter and Harris Letter; see also supra notes
999 to 1008, and surrounding discussion.
---------------------------------------------------------------------------
For stocks that are not constrained by the tick, rebates and access
fees are again on average neutral, as we now show. Consider, for
example, a stock with an economic spread of 2 cents. Absent a fee or
rebate, the quoted spread would equal the economic spread rounded up to
the next smallest tick, or this case, also 2 cents.\1024\ Given a 30
mil rebate, a liquidity provider would be willing to quote a spread of
1.4 cents (the economic spread of 2 cents minus twice the 0.3 cents
rebate, or 1.4 cents). However, given the tick, it is likely that the
quoted spread would be a full 2 cents. If it were any lower, the
profit-maximizing liquidity provider would incur a marginal cost (the
economic spread of 2 cents) that exceeds the marginal benefit (i.e.,
the next smaller quoted spread of 1 cent plus twice the 0.3 cent
rebate, or 1.6 cents). It is unlikely that the liquidity provider would
be willing to do this.\1025\ Now, consider the perspective of the
market participant taking liquidity. Because the liquidity taker is not
typically on both sides of the trade, we use the half-spread to measure
their costs due to the bid-ask spread.\1026\ This liquidity taker pays
the half-spread along with the access fee. In this case, the half-
spread is 1 cent and the access fee is 0.3 cents, so the total is 1.3
cents. As stated above, in the absence of rebates, the quoted spread
would equal its economic spread of 2 cents. The half-spread would be 1
cent, less than 1.3 cents, meaning that the liquidity taker pays more
when there are fees and rebates.
---------------------------------------------------------------------------
\1024\ This stock may become tick-constrained in the future, and
indeed some stocks that have an average spread of 2 cents over a
prior period could be tick-constrained during that time.
\1025\ Equivalently, the liquidity provider is only willing to
quote at 1.4 cents or above, and therefore the quoted spread must be
at least 2 cents.
\1026\ See section VII.B.3.a.
---------------------------------------------------------------------------
However, consider a stock with an economic spread of 2.5 cents.
Given a
[[Page 81688]]
30 mil rebate, a liquidity provider would be willing to quote a spread
of 1.9 cents (the economic spread of 2.5 cents minus twice the 0.3
cents rebate, or 1.9 cents). Again, given the tick, it is likely that
the quoted spread would be a full 2 cents. The liquidity taker would
again pay 1.3 cents. In this case, in the absence of rebates, it is
likely that the quoted spread would be 3 cents. Otherwise, the profit-
maximizing liquidity provider would incur a marginal cost (the economic
spread of 2.5 cents) that exceeds the marginal benefit (i.e., the next
smallest quoted spread of 2 cents). Given that the quoted spread is 3
cents, the quoted half-spread would be 1.5 cents, so the liquidity
taker would likely pay more if there were no fees and rebates.
This same reasoning can be used to show that, except for stocks for
which the economic spread is one cent or below, the effect of fees and
rebates cancels out mathematically.\1027\ The main intuition is that,
while liquidity providers will quote one tick lower if the rebate
pushes the spread below the next smaller tick, liquidity demanders pay
the access fee even if the rebate does not change the quoted spread. We
show that the gains to liquidity demanders from this situation are
exactly offset by their losses when the rebate does not result in
quoting one tick lower, all provided that the economic spread is
greater than one cent. Thus, the supply-demand curve reasoning in
section VII.B.3.a is robust to the introduction of the tick, provided
that the economic spread is greater than 1 tick. In other words, for
stocks for which the economic spread exceeds 1 tick, fees and rebates
are neutral on average.
---------------------------------------------------------------------------
\1027\ Let S* equal the economic spread, R the rebate, and the
tick size. Then, for any integer N >=1, the liquidity provider
collects t- 2R more in profits under no rebates versus rebates when
S* [isin] (tN, tN + 2R], and 2R less in profits when S* [egr] (tN +
2R,tN + t]. These amounts sum to zero assuming a uniform
distribution over the interval. The liquidity taker is on the other
side of the trade; in total, the profits to providers are losses to
takers, and so these also sum to zero. In contrast, for N >1, the
liquidity provider collects profits 2R over the interval (0,t], The
assumption of a uniform distribution over the interval is not
necessary for the result. It is a reasonable and standard assumption
given the lack of specific information on the properties S* of over
intervals of length equal to the tick size.
---------------------------------------------------------------------------
One commenter provides a numerical example that might appear to go
against this neutrality result. In particular, the commenter states
that, ``because rebates also increase depth, it is possible that the
costs of access fees for liquidity takers are more than offset by the
tighter spreads and depth that they create.'' \1028\ The numerical
example is as follows: A liquidity taker wants to buy 1,000 shares, and
the tick size is 1 cent. Without access fees and rebates, the liquidity
provider is willing to sell 651 shares at a price of $10.02 and 349
shares at a price of $10.03, in which instance the liquidity provider
earns $10.02349 per share.\1029\ The liquidity taker then pays an
average of $10.02349 per share to buy the 1,000 shares.\1030\ In the
presence of an access fee and rebate of 30 mils per share (i.e., $0.003
per share), the commenter states that the liquidity provider is willing
to sell all 1,000 shares at $10.02, in which instance the liquidity
provider earns only $10.023 per share (i.e., $10.02 per share + $0.003
rebate per share = $10.023 per share). The liquidity taker thus pays
only $10.023 per share to buy the 1,000 shares in the presence of an
access fee and rebate of 30 mils (i.e., $10.02 per share + $0.003
access fee per share = $10.023 per share).
---------------------------------------------------------------------------
\1028\ See Nasdaq Letter I at 23.
\1029\ In the commenter's example, prices account for the
presence of a 1 cent tick. This explains why the liquidity provider
is offering shares at $10.02 and $10.03 and not at prices in
between. The liquidity provider earns: ($10.02 * 651 + $10.03 *
349)/1,000 = $10.02349 per share to sell 1,000 shares.
\1030\ ($10.02 * 651 + $10.03 * 349)/1,000 = $10.02349 per share
to buy 1,000 shares.
---------------------------------------------------------------------------
However, it is not clear from the example why the liquidity
provider would be willing to offer all of the 1,000 shares at $10.02
per share and earn only $10.023 per share in the presence of the rebate
when it earned $10.02349 per share absent the rebate. Rather, in the
presence of the rebate, the liquidity provider would be willing to sell
951 shares at $10.02 and 49 shares at $10.03, in which instance it
would also earn $10.02349 per share on average.\1031\ Depth does
increase, as the commenter states. However, consistent with the
neutrality argument, the liquidity taker pays exactly the same as
without the fees and rebates. Thus, while tighter spreads offset the
cost of access fees, they do not ``more than offset'' these fees.\1032\
---------------------------------------------------------------------------
\1031\ (951 * $10.02 + 49 * $10.03)/1,000 + $0.003 = $10.02349
per share, where $0.003 is the rebate per share.
\1032\ See section VII.E.3 for discussion of the commenter's
concerns regarding capital formation as it relates to reduced depth
at the NBBO.
---------------------------------------------------------------------------
The previous argument for neutrality pertained to stocks for which
the economic spread was greater than one tick. For stocks for which the
economic spread is less than 1 tick, however, fees and rebates are not
neutral. Rather, because the quoted spread cannot fall to compensate
for the rebate, provision of liquidity is overpriced at the spread of
one tick. Because the price is artificially high, the supply of
liquidity is distorted,\1033\ leading to rents for liquidity providers
who can get to the top of the queue the fastest. It is harder as a
result for slower liquidity providers, such as retail investors and
institutions to have their limit orders executed. It is also more
expensive for investors seeking to access liquidity. For these stocks,
lowering the access fees lowers rents and improves market quality,
making it cheaper to transact for investors as a whole.
---------------------------------------------------------------------------
\1033\ Proposing Release, supra note 11, at 80328-29.
---------------------------------------------------------------------------
C. Baseline
The baseline against which the costs, benefits, and the effects on
efficiency, competition, and capital formation of the amendments are
measured consists of the current state of the trading environment for
NMS stocks, including pricing increments; current practice as it
relates to order routing, quotes, fees, and rebates; and availability
of data about quotes, fees, and rebates; and the current regulatory
framework. The economic analysis appropriately considers existing
regulatory requirements, including recently adopted rules, as part of
its economic baseline against which the costs and benefits of the
amendments are measured.\1034\
---------------------------------------------------------------------------
\1034\ See, e.g., Nasdaq v. SEC, 34 F.4th 1105, 1111-15 (D.C.
Cir. 2022). This approach also follows Commission staff guidance on
economic analysis for rulemaking. See Current Guidance on Economic
Analysis in SEC Rulemaking, (Mar. 16, 2012), available at https://www.sec.gov/divisions/riskfin/rsfi_guidance_econ_analy_secrulemaking.pdf (``The economic
consequences of proposed rules (potential costs and benefits
including effects on efficiency, competition, and capital formation)
should be measured against a baseline, which is the best assessment
of how the world would look in the absence of the proposed
action.''); id. at 7 (``The baseline includes both the economic
attributes of the relevant market and the existing regulatory
structure.''). The best assessment of how the world would look in
the absence of the proposed or final action typically does not
include recently proposed actions, because that would improperly
assume the adoption of those proposed actions.
---------------------------------------------------------------------------
[[Page 81689]]
Several commenters requested that the Commission consider
interactions between the economic effects of the proposed rule and
other recent Commission rules.\1035\ Since the date of the Proposing
Release, the Commission has adopted eight rules mentioned by
commenters,\1036\ namely the Settlement Cycle Adopting Release,\1037\
the February 2024 Form PF Adopting Release,\1038\ the May 2023 SEC Form
PF Adopting Release,\1039\ the Dealer Adopting Release,\1040\ the
Beneficial Ownership Adopting Release,\1041\ Rule 10c-1a Adopting
Release,\1042\ the Short Position Reporting Adopting Release,\1043\ and
the Rule 605 Amendments.\1044\ The Commission has also considered the
potential effects on entities that are implementing other recently
adopted rules during the compliance period for these amendments,
including the Treasury Clearing Adopting Release \1045\ and the
[[Page 81690]]
Customer Notification Adopting Release.\1046\ These recently adopted
rules were not included as part of the baseline in the Proposing
Release because they were not yet adopted at that time, but they are
part of the baseline in this analysis.\1047\ In response to commenters,
this economic analysis considers potential economic effects arising
from any overlap in compliance dates between these amendments and the
other recent amendments.\1048\ It also considers interactions between
these amendments and the Rule 605 Amendments.\1049\
---------------------------------------------------------------------------
\1035\ See, e.g., CCMR Letter; Independent Trustees Letter;
SIFMA Letter I; Citadel Letter I; Equity Market Structure Citadel
Letter; ICAN Letter; Virtu Letter II; AIC Letter; AIMA Letter;
Antitrust Division of the DOJ Letter; Wagner Letter; Danny Mulson
Letter. See also supra notes 135 to 137, and surrounding text,
discussing these comments. Many commenters referred to the ``Equity
Market Structure Proposals'' as a group, and we understand that
commenters intended this to mean the four rulemaking proposals the
Commission issued on Dec. 14, 2022. Of those four, only one was
adopted prior to these amendments. One commenter described
``Interconnected Rules'' to include those four proposed rules issued
on Dec. 14, 2023, as well as certain rules that were previously or
later adopted, see infra notes 1036 to 1046; and a variety of other
proposed rules including Safeguarding Advisory Client Assets,
Investment Advisers Act of 1940 Release No. 6384 (Aug. 23, 2023), 88
FR 14,672 (Mar. 9, 2023). See AIC Letter at 1 n.3, 9 n.30 (listing
rules and proposed rules, but not explaining a specific connection
to the Proposing Release).
\1036\ In addition, commenters also mentioned the proposal that
was ultimately adopted as Private Fund Advisers; Documentation of
Registered Investment Adviser Compliance Reviews, Advisers Act
Release No. 6383 (Aug. 23, 2023), 88 FR 63206 (Sept. 14, 2023)
(``Private Fund Advisers Adopting Release''). On June 5, 2024, the
Fifth Circuit issued a ruling that vacated the rules and amendments
adopted in the Private Fund Advisers Adopting Release. Nat'l Ass'n
of Priv. Fund Managers v. SEC, 103 F.4th 1097 (2024).
\1037\ Shortening the Securities Transaction Settlement Cycle,
Securities Exchange Act Release No. 96930 (Feb. 15, 2023), 88 FR
13872 (Mar. 6, 2023) (``Settlement Cycle Adopting Release''). The
rules and rule amendments adopted in the Settlement Cycle Adopting
Release shorten the standard settlement cycle for most broker-dealer
transactions from two business days after the trade date to one
business day after the trade date. To facilitate an orderly
transition to a shorter settlement cycle, a new rule also
establishes requirements related to completing allocations,
confirmations, and affirmations no later than the end of trade date
for the processing of institutional transactions subject to the
rule; requires registered investment advisers to make and keep
records of each confirmation received, and of any allocation and
each affirmation sent or received, with a date and time stamp for
each allocation and affirmation indicating when it was sent or
received; and requires clearing agencies that provide a central
matching service to establish, implement, and enforce policies and
procedures reasonably designed to facilitate straight-through
processing and to file an annual report regarding progress with
respect to straight-through processing. With certain exceptions, the
rule has a compliance date of May 28, 2024. See Settlement Cycle
Adopting Release, section VII.
\1038\ Form PF: Reporting Requirements for All Filers and Large
Hedge Fund Advisers, Advisers Act Release No. 6546 (Feb. 8, 2024),
89 FR 17984 (Mar. 12, 2024) (``February 2024 Form PF Adopting
Release''). The Form PF amendments are designed to enhance the
Financial Stability Oversight Council's ability to monitor systemic
risk as well as bolster the SEC's regulatory oversight of private
fund advisers and investor protection efforts. The compliance date
for the rule is Mar. 12, 2025. February 2024 Form PF Adopting
Release, section II.F.
\1039\ Form PF; Event Reporting for Large Hedge Fund Advisers
and Private Equity Fund Advisers; Requirements for Large Private
Equity Fund Adviser Reporting, Investment Company Act of 1940
Release No. 6297 (May 3, 2023), 88 FR 38146 (June 12, 2023) (``May
2023 SEC Form PF Adopting Release''). The Form PF amendments adopted
in May 2023 require large hedge fund advisers and all private equity
fund advisers to file reports upon the occurrence of certain
reporting events. The compliance dates were Dec. 11, 2023, for the
event reports in Form PF sections 5 and 6, and June 11, 2024, for
the remainder of the Form PF amendments in the May 2023 SEC Form PF
Adopting Release. See May 2023 SEC Form PF Adopting Release, section
II.E.
\1040\ Further Definition of ``As a Part of a Regular Business''
in the Definition of Dealer and Government Securities Dealer in
Connection with Certain Liquidity Providers, Securities Exchange Act
Release No. 34-99477 (Feb. 6, 2024), 89 FR 14938 (Feb. 29, 2024)
(``Dealer Adopting Release''). New Rules 3a5-4 and 3a44-2 further
define the phrase ``as a part of a regular business'' as used in the
statutory definitions of ``dealer'' and ``government securities
dealer.'' The compliance date is Apr. 29, 2025, for persons engaging
in activities that meet the qualitative factors under the final
rules. See Dealer Definition Adopting Release, section II.B.
\1041\ Modernization of Beneficial Ownership Reporting,
Securities Act of 1933 Release No. 11253 (Oct. 10, 2023), 88 FR
76896 (Nov. 7, 2023) (``Beneficial Ownership Adopting Release'').
Among other things, the amendments generally shorten the filing
deadlines for initial and amended beneficial ownership reports filed
on Schedules 13D and 13G, and require that Schedule 13D and 13G
filings be made using a structured, machine-readable data language.
The amendments are effective Feb. 5, 2024. The new filing deadline
for Schedule 13G will not be required before Sept. 30, 2024, and the
rule's structured data requirements will not be required until Dec.
18, 2024. Beneficial Ownership Adopting Release, section II.G.
\1042\ Reporting of Securities Loans, Securities Exchange Act
Release No. 98737 (Oct. 13, 2023), 88 FR 75644 (Nov. 3, 2023)
(``Rule 10c-1a Adopting Release''). This rule requires any covered
person who agrees to a covered securities loan on behalf of itself
or another person to report specified information about the covered
securities loan to a registered national securities association
(currently FINRA is the only registered national securities
association)--or rely on a reporting agent to do so--and requires
the registered national securities association to make certain
information it receives available to the public. Covered persons
will include market intermediaries, securities lenders, and broker-
dealers, while reporting agents include certain brokers, dealers, or
registered clearing agencies. The rule's compliance dates require
that the registered national securities association propose rules
pursuant to Rule 10c-1a(f) by May 2, 2024, and the proposed rules
shall be effective no later than Jan. 2, 2025; that covered persons
report Rule 10c-1a information to a registered national securities
association on or by Jan. 2, 2026 (which requires that the
registered national securities association have implemented data
retention and availability requirements for reporting); and that the
registered national securities association publicly report Rule 10c-
1a information by Apr. 2, 2026. Rule 10c-1a Adopting Release,
section VIII.
\1043\ Short Position and Short Activity Reporting by
Institutional Investment Managers, Securities Exchange Act Release
No. 98738 (Oct. 13, 2023), 88 FR 75100 (Nov. 1, 2023) (``Short
Position Reporting Adopting Release''). Under the new rule,
institutional investment managers that meet or exceed certain
specified reporting thresholds are required to report, on a monthly
basis using the related form, specified short position data and
short activity data for equity securities. The compliance date is
Jan. 2, 2025. See Short Position Reporting Adopting Release, section
VI. In addition, the Commission adopted an amendment to the national
market system (``NMS'') plan governing the consolidated audit trail
(``CAT'') created pursuant to the Exchange Act to require the
reporting of reliance on the bona fide market making exception in
the Commission's short sale rules. The Commission published the text
of the amendment to the NMS plan governing the CAT (``CAT NMS
Plan'') in a separate notice. The compliance date for the amendment
to the CAT NMS Plan is July 1, 2025. See SEC, Notice of the Text of
the Amendment to the National Market System Plan Governing the
Consolidated Audit Trail for Purposes of Short Sale-Related Data
Collection, Securities Exchange Act Release No. 98739 (Oct. 13,
2023), 88 FR 75079 (Nov. 1, 2023).
\1044\ Rule 605 Amendments, supra note 10. The Commission
adopted amendments to rules requiring disclosures for order
executions in NMS stocks, including expanding the scope of reporting
entities, modifying the scope of orders covered by the rule, and
modifying the information required to be reported under the rule.
The rule has an effective date of June 14, 2024, and, with a few
exceptions, a compliance date of Dec. 14, 2025. See Rule 605
Amendments, section VII.
\1045\ Standards for Covered Clearing Agencies for U.S. Treasury
Securities and Application of the Broker-Dealer Customer Protection
Rule with Respect to U.S. Treasury Securities, Securities Exchange
Act Release No. 99149 (Dec. 13, 2023), 89 FR 2714 (Jan. 16, 2024)
(``Treasury Clearing Adopting Release''). Among other things, the
amendments require covered clearing agencies for U.S. Treasury
securities to have written policies and procedures reasonably
designed to require that every direct participant of the covered
clearing agency submit for clearance and settlement all eligible
secondary market transactions in U.S. Treasury securities to which
it is a counterparty. The compliance date was Mar. 18, 2024, for
covered clearing agencies to file any proposed rule changes pursuant
to Rules 17Ad-22(e)(6)(i), 17Ad-22(e)(18)(iv)(C), and 15c3-3, which
must be effective by Mar. 31, 2025. With respect to the changes to
Rule 17Ad-22(e)(18)(iv)(A) and (B), (i) covered clearing agencies
were required to file any proposed rule changes regarding those
amendments no later than June 14, 2024, and (ii) those changes must
be effective by Dec. 31, 2025, for cash market transactions
encompassed by section (ii) of the definition of an eligible
secondary market transaction, and by June 30, 2026, for repo
transactions encompassed by section (i) of the definition of
eligible secondary market transactions. Finally, the Commission
amended the broker-dealer customer protection rule to permit margin
required and on deposit with covered clearing agencies for U.S.
Treasury securities to be included as a debit in the reserve
formulas for accounts of customers and proprietary accounts of
broker-dealers, subject to certain conditions. Compliance by the
direct participants of a U.S. Treasury securities covered clearing
agency with the requirement to clear eligible secondary market
transactions is not required until Dec. 31, 2025, and June 30, 2026,
respectively, for cash and repo transactions. See Treasury Clearing
Adopting Release, section III.
\1046\ Regulation S-P: Privacy of Consumer Financial Information
and Safeguarding Customer Information, Securities Exchange Act
Release No. 100155 (May 15, 2024), 89 FR 47688 (June 3, 2024)
(``Customer Notification Adopting Release''). The Commission amended
Regulation S-P to require brokers, dealers, investment companies,
registered investment advisers, and transfer agents registered with
the Commission or another appropriate regulatory agency to adopt
written policies and procedures for incident response programs to
address unauthorized access to or use of customer information. These
must include procedures for providing timely notification to
individuals affected by an incident involving sensitive customer
information with details about the incident and information designed
to help affected individuals respond appropriately. Among other
things, the amendments also broadened the scope of information
covered by the safeguards rule and the disposal rule, and extended
the requirements to safeguard customer records and information to
all transfer agents. The compliance date for larger entities is Dec.
3, 2025, and for smaller entities, June 3, 2026. Customer
Notification Adopting Release, section II.F.
\1047\ Some commenters assumed that the four proposed rules
issued on Dec. 14, 2022, would be implemented simultaneously, and
therefore stated that the baseline in the Proposing Release was
inaccurate to the extent it did not contemplate that the other rules
have gone into effect. See, e.g., Equity Market Structure Citadel
Letter at 15. As discussed above, however, our baseline does not
assume the adoption of proposed rules. Instead, the baseline changes
incrementally with each adopted rule. To the extent those or other
proposals are adopted in the future, the baseline in those
subsequent rulemakings will reflect the existing regulatory
requirements at that time.
\1048\ See infra section VII.D.6.b.
\1049\ See, e.g., SIFMA Letter II (stating that variable tick
sizes could diminish the ability to compare execution quality using
the Rule 605 disclosures); Citadel Letter II (stating execution
quality statistics are important to understanding the effects of
these amendments on market quality); AIMA Letter (suggesting
finalization of the proposed Rule 605 amendments would, along with
the MDI round lot order definition, provide a much more informed
economic baseline against which to assess other equity market
structure proposals); see also infra section VII.D.6.a, discussing
this topic.
---------------------------------------------------------------------------
1. Tick Sizes
Preexisting Rule 612 of Regulation NMS restricts the ability of
venues to display, rank, or accept quotations in NMS stocks beyond a
certain minimum quoting increment (or tick). This section discusses
current regulation on tick sizes and Commission analysis showing that
the current tick size acts as a binding price floor on the quoted
spread a significant portion of the time for a large fraction of the
share volume transacted in NMS stocks.
a. Current Regulations
Preexisting Rule 612 of Regulation NMS, which came into effect on
August 29, 2005, prohibited a national securities exchange, national
securities association, ATS, vendor, or broker or dealer from
displaying, ranking, or accepting quotations, orders, or indications of
interest in any NMS stock priced in an increment smaller than $0.01 if
the quotation, order, or indication of interest is priced equal to or
greater than $1.00 per share. If the quotation, order, or indication of
interest is priced less than $1.00 per share, the minimum pricing
increment is $0.0001. Most listing exchanges require stocks listed on
their exchanges to maintain a price greater than $1.00 per share, and
consequently $0.01 is the prevailing tick size for most quotes and
orders for NMS stocks.\1050\ Preexisting Rule 612 of Regulation NMS
effectively establishes $0.01 as the minimum spread that can be quoted
for stocks priced equal to, or greater than, $1.00 per share because
the NBBO is determined by the best displayed round lot quotes, and
exchanges are required to have rules in place to avoid and reconcile
locked and crossed quotations.\1051\
---------------------------------------------------------------------------
\1050\ See, e.g., NYSE Continued Listing Standards, Sec.
802.01C, available at https://www.nyse.com/listings/resources;
Rulebook--The Nasdaq Stock Market, Sec. 5400, available at https://listingcenter.nasdaq.com/rulebook/nasdaq/rules.
\1051\ See Reg NMS Rule 610(d). A locked market occurs when the
bid and ask price for a security are identical. A crossed market
occurs when the bid is higher than the ask.
---------------------------------------------------------------------------
While preexisting Rule 612 of Regulation NMS restricts quoting or
submitting orders in sub-penny increments for NMS stocks priced greater
than or equal to $1.00, it does not restrict trading in sub-penny
increments. Sub-penny trading on exchanges and ATSs occurs primarily as
a result of midpoint orders and benchmark trades. Benchmark trades,
such as volume weighted average price (``VWAP'') and time weighted
average price (``TWAP'') orders, may not be explicitly priced in an
impermissible sub-penny increment, but the ultimately determined
execution price may be in a sub-penny increment. Trading at sub-penny
increments also occurs as a result of broker-dealers, including some
OTC market makers known as wholesalers, internalizing customer order
flow at sub-penny prices.\1052\
---------------------------------------------------------------------------
\1052\ The term ``wholesaler'' is not defined in Regulation NMS,
but commonly refers to a broker-dealer acting as an OTC market maker
that primarily focuses on attracting orders from broker-dealers that
service the accounts of a large number of individual investors,
referred to in this release as ``retail brokers.''
---------------------------------------------------------------------------
Sub-penny trading on registered exchanges may also occur as a
result of their RLPs. The Commission granted exemptions from Rule 612
to various national securities exchanges' RLPs as a means to allow them
to compete with OTC sub-penny price improvement.\1053\ Under the RLPs,
exchanges can accept and rank certain quotes and orders from certain
participants in sub-penny increments as small as $0.001.\1054\ The
national securities exchanges designed the RLPs to attract retail
orders by providing a potential for price improvement at sub-penny
levels because ``most marketable retail order flow is executed in the
OTC markets, pursuant to bilateral agreements, without ever reaching a
public exchange'' and OTC market makers typically pay retail brokers
for their order flow.\1055\ Quotes in RLP programs are not displayed.
Instead, the appropriate SIP disseminates a flag indicating the side of
the market for which an exchange has an RLP quote available at a price
better than the NBBO. Because the exclusive SIP does not make known the
price or the size of the RLP quote, market participants do not see the
full liquidity available in RLP programs.\1056\ To date, RLPs have not
attracted a significant volume of retail order flow.\1057\
---------------------------------------------------------------------------
\1053\ See Securities Exchange Act Release No. 67347 (July 3,
2012), 77 FR 40673 (July 10, 2012) (approving retail liquidity
programs on a pilot basis for NYSE and NYSE Amex and granting rule
612 exemption) (``NYSE Retail Liquidity Program Approval Order'');
see also CBOE BYX Rule 11.24; Securities Exchange Act Release No.
68303 (Nov. 27, 2012), 77 FR 71652 (Dec. 3, 2012) (CBOE BYX Retail
Pilot Program Approval Order); and Nasdaq BX Equity Rule 4780;
Exchange Act Release No. 73702 (Nov. 28, 2014), 79 FR 72049 (Dec. 4,
2014) (NASDAQ BX Retail Pilot Program Approval Order).
\1054\ See discussion in supra section III.C.1.
\1055\ See specifically, NYSE Retail Liquidity Program Approval
Order, supra note 1053, at 40679. The Commission stated that
``[i]nternalizing broker-dealer[s] can offer sub-penny executions,
provided that such executions do not result from impermissible sub-
penny orders or quotations'' by ``typically select[ing] a sub-penny
price for a trade without quoting at that exact amount or accepting
orders from retail customers seeking that exact price.''
\1056\ See, e.g., UTP Participant Input Specification (April
2024), available at https://www.utpplan.com/DOC/UtpBinaryInputSpec_Fractional.pdf.
\1057\ See Proposing Release, supra note 11, at 80272 n.70
(citing to industry and academic discussions on why RLPs may not
attract significant retail order flow).
---------------------------------------------------------------------------
[[Page 81691]]
One commenter stated the ability to trade in increments of smaller
than a penny in certain circumstances as means of suggesting that there
may be no need for rulemaking.\1058\ However, it is not possible to
post a displayed quote at an increment other than a penny. The economic
forces that govern the tradeoffs in determining the tick size depend on
the quote being displayed, ranked, and accepted. The empirical analysis
also pertains to displayed quotes.
---------------------------------------------------------------------------
\1058\ See Lewis Letter at 33-34, attached to Virtu letter II.
---------------------------------------------------------------------------
b. Analysis of Quoted Spreads Under Current Ticks
This section discusses empirical analysis by the Commission on the
prevalence of trading at different quoted spread ranges, defined in the
Proposing Release using the concept of TWAQS.\1059\ To document the
prevalence of trading at different quoted spread ranges, table 3
presents data on trading volume in 2023 based on average time weighted
quoted spreads throughout the entire year for NMS stocks with a
quotation, order, or indication of interest priced equal to or greater
than $1.00 per share.\1060\ The analysis breaks trading volume each day
into one of 17 average quoted spread bins, beginning with stocks that
have TWAQS less than or equal to $0.011.\1061\ Table 3 also reports the
daily average number of stocks in each bin.
---------------------------------------------------------------------------
\1059\ See Proposing Release, supra note 11, table 4.
\1060\ The data derived in table 3 was derived using the same
methodology as corresponding table 4 in the Proposing Release (PR
table 4). See Proposing Release, supra note 11, at 80308. The only
differences between the tables are that table 3 uses data for all
trading days in 2023 (instead of from Jan. to May 2022 in PR table
4), and table 3, because of our policy choice to set the minimum
tick size at 0.5c (for quotes and orders priced $1.00 or more for
NMS stocks that have a TWAQS of $0.015 or less), adds tick size bins
for quoted spreads from 1.1c to 1.5c and 1.5c to 2c.
\1061\ Because of the $0.01 minimum quoting increment for NMS
stocks priced equal to or greater than $1.00 per share, a stock
cannot have a quoted spread less than $0.01 unless markets become
locked or crossed. The existence of locked and crossed markets can
in some cases result in time weighted quoted spread that are very
slightly lower than $0.01. However, even for stocks with spreads
most constrained by the tick, even a faction of a second spent with
a higher spread would likely result in an average quoted spread
higher than $0.01. For example, a large trade can exhaust liquidity
deeper in the limit order book such that the stock's quoted spread
temporarily increases from $0.01. Thus, time weighed quoted spreads
will virtually always be greater than $0.01. This makes $0.011 a
more pragmatic minimum cutoff for empirical analysis than $0.01.
---------------------------------------------------------------------------
The data analysis in table 3 indicates that under the current tick
size regime in 2023, 65.2% of share trading volume (31.5% of dollar
volume) occurred in stocks in the first average quoted spread bin of
$0.011 or less. Table 3 also reports that an additional 9.1% of share
volume (15.3% of dollar volume) occurred in stocks with quoted spreads
between $0.011 and $0.015. In sum, in table 3, in 2023, approximately
74.3% of share volume (46.8% of dollar volume) transacted in NMS stocks
(specifically, NMS stocks with a quotation, order, or indication of
interest priced equal to or greater than $1.00 per share) which have a
quoted spread that is likely to be constrained by the minimum pricing
increment,\1062\ which as discussed both above (section VII.B.2) and
below (VII.D.1) increases transaction costs.
---------------------------------------------------------------------------
\1062\ For the share volume, 74.3% = 65.2% (i.e., Quoted Spread
<= $0.011) + 9.1% (i.e., $0.015 < Quoted Spread <= $0.02). For the
dollar volume, 46.8% = 31.5% (i.e., Quoted Spread <= $0.011) + 15.3%
(i.e., $0.015 < Quoted Spread <= $0.02). See supra section VII.B.2
for a definition of tick-constrained; see also Proposing Release,
supra note 11, table 4, which used the same methodology and
estimated that, in the first six months of 2022, 56% of share volume
transacted in NMS stocks with a quoted spread of < $0.011, while an
additional 15% of share volume traded in stocks with a quoted spread
of $0.011 < and <= $0.02.
Table 3--Share Volume by Quoted Spread 2023 \a\
----------------------------------------------------------------------------------------------------------------
Share volume Dollar volume Average #
Quoted spread (%) (%) stocks
----------------------------------------------------------------------------------------------------------------
Quoted Spread <= $0.011......................................... 65.2 31.5 1,782
$0.011 < Quoted Spread <= $0.015................................ 9.1 15.3 638
$0.015 < Quoted Spread <= $0.02................................. 4.2 5.2 560
$0.02 < Quoted Spread <= $0.03.................................. 5.7 8.4 1,036
$0.03 < Quoted Spread <= $0.04.................................. 3.6 7.6 811
$0.04 < Quoted Spread <= $0.05.................................. 2.1 3.6 673
$0.05 < Quoted Spread <= $0.06.................................. 1.5 2.3 582
$0.06 < Quoted Spread <= $0.07.................................. 1.2 2.1 522
$0.07 < Quoted Spread <= $0.08.................................. 1.0 1.8 445
$0.08 < Quoted Spread <= $0.09.................................. 0.8 1.6 377
$0.09 < Quoted Spread <= $0.10.................................. 0.7 1.6 317
$0.10 < Quoted Spread <= $0.11.................................. 0.6 1.5 271
$0.11 < Quoted Spread <= $0.12.................................. 0.5 1.3 222
$0.12 < Quoted Spread <= $0.13.................................. 0.4 1.1 187
$0.13 < Quoted Spread <= $0.14.................................. 0.3 1.1 163
$0.14 < Quoted Spread <= $0.15.................................. 0.3 1.0 144
$0.15 < Quoted Spread........................................... 2.8 13.0 2,177
----------------------------------------------------------------------------------------------------------------
\a\ This table provides share volume by stocks with different quoted spread profiles. To create this table, for
each day the universe of stocks (identified by a unique stock variable) covered in the WRDS Intra-Day
Indicators data are assigned into one of the 17 quoted spread bins based on that day's time weighted quoted
spread as computed by WRDS Intra-Day Indicators. Then all share and dollar trading volume across all trading
days in 2023 is aggregated for each of the 17 quoted spread bins. Percentages based on these totals are then
computed. This table also presents the daily average number of stocks in each bin. To compute this variable,
for each trading day in 2023, the number of stocks in each bin is tabulated, then the average across all
trading days is presented here. Certain items in this table 3 may also be affected by the MDI Rules once they
are fully implemented. See infra section VII.C.3.
c. Reverse Stock Splits
A reverse split exchanges a fixed number of existing shares for a
smaller number of new shares. The new shares have a higher price, but
there are fewer of them.\1063\ For example, if an issuer undergoes an
8-for-1 reverse split, eight shares are exchanged for one share that is
worth the same as the eight,
[[Page 81692]]
increasing the price by a factor of eight. All else equal, one would
expect the quoted spread, which represents the per-share trading costs
to also rise by a factor of eight. Thus, the cost of transacting in the
stock, for a given dollar exposure, would remain constant. However, if
a stock were tick constrained, undergoing a reverse split would cause
the tick to become smaller relative to the (higher) stock price,
thereby alleviating the tick constraint and reducing transaction
costs.\1064\
---------------------------------------------------------------------------
\1063\ See, e.g., General Electric, GE Reverse Stock Split
Frequently Asked Questions as of September 21, 2021, at 1, available
at https://www.ge.com/sites/default/files/GE_Reverse_Stock_Split_FAQs.pdf (last accessed July 27, 2024). See
also FINRA, Stock Splits, available at https://www.finra.org/investors/investing/investment-products/stocks/stock-splits.
\1064\ For example, suppose a stock trades at $10 per share and
is tick-constrained such that it trades with a quoted spread of
$0.01, but could sustain a quoted spread of $0.004 if not for the
penny tick. If the stock undergoes a 5-for-1 reverse split, then the
share price would rise to $50 (5*$10) and the quoted spread would
rise to $0.02 (5*$0.004). The reverse split thereby alleviates the
tick constraint. The reverse split also reduces trading costs in
this case. To see the reduction in costs, suppose an investor wants
to submit a market order for $100 worth of the stock. With a $10
price and a $0.005 half-spread, this would entail purchasing five
shares and paying a transaction cost of $0.025 (i.e., five times the
half-spread of $0.005). With a $50 price and a $0.01 half-spread,
however, the market order would only entail purchasing two shares
and paying a transaction cost of $0.02 (i.e., two times the half-
spread of $0.01), thus reducing transaction costs by 25%. See infra
note 1295 and surrounding discussion for empirical studies
documenting a reduction in trading costs when tick-constrained
stocks complete a reverse split.
---------------------------------------------------------------------------
Using this logic, one commenter pointed out that currently the
problem that the Commission identifies with regard to tick constrained
stocks could be solved without Commission action by issuers making the
decision to undergo a reverse stock split.\1065\ The commenter further
states, ``[t]he fact that issuers do not discuss this indicates the
issue is immaterial to them.'' \1066\ While the Commission agrees that
a reverse split alleviates the tick constraint, there may be reasons
that an issuer may choose not to undergo a reverse split apart from an
assessment of the immateriality of tick constraints. Reverse splits
have historically been associated with negative stock returns.\1067\
Research suggests that this may be because the market views stock
splits as a signal,\1068\ and issuers may not wish to give the
appearance of bad news with a reverse split. Reverse splits also change
the share price, which may affect the trading characteristics of the
stock.\1069\ In addition, the direct administrative costs of a stock
split for a large issuer are estimated to be between $250,000 and
$800,000 (as of 2009).\1070\ Finally, an issuer may not capture all of
the benefits of the liquidity arising from the reverse split and the
alleviated constraints. In sum, though reverse stock splits may address
tick constraints, they are an inefficient solution for tick constrained
securities. One commenter agreed, stating that ``while issuers can make
pricing in their securities more or less granular through reverse or
forward splits, it's not practical for the SEC and the industry to rely
on them to do so.'' \1071\
---------------------------------------------------------------------------
\1065\ See Virtu Letter II at 25.
\1066\ Id. at 25.
\1067\ See, e.g., Hemang Desai & Prem C. Jain, Long-Run Common
Stock Returns Following Stock Splits and Reverse Splits, 70 J.
Business 3 (July 1997); Kim et al., Return Performance Surrounding
Reverse Stock Splits: Can Investors Profit?, 37 Fin. Mgmt. 2 (Summer
2008).
\1068\ For a discussion on the signal conveyed by stock splits,
see, e.g., Maureen McNichols & Ajay Dravid, Stock Dividends, Stock
Splits, and Signaling, 45 J. Fin. 3 (July 1990); David L. Ikenberry
et al., What Do Stock Splits Really Signal?, 31 J. Fin. &
Quantitative Analysis 3 (Sept. 1996). The academic literature on
signaling generally uses stock splits (rather than reverse splits)
because standard databases do not include the announcement date of
reverse splits, which makes the measurement of the market reaction
to the announcement of a reverse split imprecise. Once the reverse
split is completed, however, the academic literature documents
negative returns--see id.
\1069\ See, e.g., Kelly Shue & Robert Townsend, Can the Market
Multiply and Divide? Non-Proportional Thinking in Financial Markets,
76 J. Fin 5 (Oct. 2021). This paper documents greater return
responses to news in lower-priced stocks and hypothesizes that some
investors think about stock price changes in terms of dollars rather
than returns. The paper studies reverse stock splits and finds
evidence consistent with this hypothesis.
\1070\ See Weld et al., The Nominal Share Price Puzzle, 23 J.
Econ. Perspectives 2 (Spring 2009). The authors estimate the
administrative costs of stock splits. Reverse stock splits are
likely to have similar administrative costs. After adjusting for
changes in the consumer price index from 2009 to 2024, the paper's
estimated direct administrative cost of a split for a large issuer
in 2024 is $365,000 to $1,170,000.
\1071\ See The Tick Size Debate Revisited attached to MEMX
Letter at 69.
---------------------------------------------------------------------------
2. Access Fees
This section discusses current regulation on the access fee cap,
the current practices at exchanges for setting access fees and rebates,
and Commission analysis showing that most exchanges assess access fees
close to the current access fee cap and use these access fees to
principally fund rebates.
a. Current Regulations
Preexisting Rule 610(c) limits the fees that trading centers can
charge for accessing protected quotations with prices of $1.00 per
share or greater to $0.0030 per share (or 30 cents per 100 shares).
This level is commonly referred to as 30 mils.\1072\ Preexisting Rule
610 also prohibits access fees in excess of 0.3% of the price for
stocks priced less than $1.00 per share. The 30 mil fee cap was adopted
as a part of Regulation NMS in conjunction with the order protection
rule and was implemented to prevent trading centers from charging
excessive fees to orders that were required to trade with a protected
quote.\1073\ The 30 mil fee cap was also set based on existing market
practices at the time.\1074\ Rule 610(c) only regulates fees to access
protected quotes; it does not regulate fees to access non-protected
quotes, nor does it regulate rebates that exchanges can offer. However,
the 30 mil fee cap has become a central component of the structure of
fees and rebates as access fees for non-protected quotes generally do
not exceed the 30 mil fee cap, nor do typical rebates.\1075\
---------------------------------------------------------------------------
\1072\ See Proposing Release, supra note 11, at 80309.
\1073\ See Regulation NMS Adopting Release, supra note 4, at
37545 (justifying the 30 mil limit: ``For quotations to be fair and
useful, there must be some limit on the extent to which the true
price for those who access quotations can vary from the displayed
price . . . . To protect limit orders, orders must be routed to
those markets displaying the best-priced quotations. This purpose
would be thwarted if market participants were allowed to charge
exorbitant fees that distort quoted prices''); see also supra note
434 and surrounding discussion from the Regulation NMS Adopting
Release on the potential--absent a fee cap--for high fee markets to
take advantage of intermarket price protections.
\1074\ See id. at 37503 (the 30 mil access fee cap was chosen
because ``it will not seriously interfere with current business
practices'' and ``[i]n the absence of a fee limitation, some
`outlier' trading centers might take advantage of the requirement to
protect displayed quotations by charging exorbitant fees to those
required to access the outlier's quotations''); see also supra note
357.
\1075\ See infra table 4 for a summary of transaction-based fee
schedules for U.S. national equities exchanges as of Feb. 2024.
---------------------------------------------------------------------------
b. Current Practices at Exchanges
The transaction fee structure on an exchange currently takes one of
three forms. The most common is maker-taker, in which liquidity
demanders (i.e., takers) are assessed the access fee and liquidity
providers (i.e., makers) are offered a rebate. Exchanges can also be
inverted (also known as taker-maker), in which liquidity demanders are
offered a rebate and liquidity providers are assessed an access
fee.\1076\ The last form of fee structure is flat; a flat exchange
either charges one or both sides a fee but does not offer rebates.
While the exchanges are free to subsidize rebates beyond what they earn
through collecting access fees, in practice this does not appear to
happen.\1077\ The difference between the average access fee charged and
the average rebate paid
[[Page 81693]]
is the net capture earned by the exchanges for facilitating a
transaction.\1078\
---------------------------------------------------------------------------
\1076\ While Rule 611 creates incentives for exchanges to use
high fees and rebates in order to quote at the NBBO--see infra note
1120--not all exchanges compete on this margin (e.g., inverted
exchanges).
\1077\ See infra section VII.D.2 for more discussion on why
exchanges may not subsidize rebates from other sources of revenue;
see also Eric Budish, et al., A Theory of Stock Exchange Competition
and Innovation: Will the Market Fix the Market? (working paper May
22, 2019) available at https://ssrn.com/abstract=3391008 (retrieved
from SSRN Elsevier database).
\1078\ See Proposing Release, supra note 11, at 80291 n.304
(``Net capture'' is the amount earned by the trading center for
facilitating a transaction, which is typically the difference
between the average access fee charged by the trading center and the
average rebate paid by the trading center''). It is common practice
across exchanges to fund their rebates with transaction fees. In
principle rebates could exceed access fees as, unlike access fees,
there is no regulatory cap restricting the rebates that can be
offered. However, it is unlikely that trading venues would offer
rebates in excess of the fees collected as doing so would expose
them to the possibility of large losses. See supra note 1101 and
accompanying text for further discussion on exchanges' net capture
rates.
---------------------------------------------------------------------------
The regulatory access fee cap is most relevant for maker-taker
markets where the trader accessing a protected quote must pay the
access fee. This is because the access fee cap applies only to fees for
accessing protected quotations and does not apply to fees for posting
quotations. On an inverted venue, the exchange is not restricted by
preexisting Rule 610 in terms of the rebate that it can offer to access
a protected quote or the fee to post a protected quote.\1079\ Flat rate
venues, which do not offer rebates, do not appear to be economically
constrained by the preexisting Rule 610(c) as their fees for both
taking and adding liquidity are significantly lower than the 30 mil fee
cap.\1080\
---------------------------------------------------------------------------
\1079\ As can be seen from table 4, which presents information
on access fees and rebates for the 16 operating exchanges, in
practice the fee that is charged on an inverted fee venue to post
liquidity is generally very close to the 30 mil access fee cap even
though not constrained by Rule 610.
\1080\ As can be seen from table 4, the only flat exchange
(LTSE) is one that does not levy either a fee or rebate and prior to
adopting a maker-taking fee model another exchange (IEX) had charged
both sides of the transaction 9 mils. See Proposing Release, supra
note 11, at 80311.
---------------------------------------------------------------------------
Fee/rebate schedules can be quite complex, and the fee schedules
change frequently.\1081\ As was discussed in the Proposing
Release,\1082\ the actual fee or rebate that an exchange member is
assessed on most exchanges also generally depends on which tier a
market participant falls into based on trading volume in that month,
with higher-volume market participants typically receiving a higher
rebate or a lower fee.\1083\ Exchanges file their fee and rebate
schedules with the Commission and post them on their websites. While
this means that the rebate and fee rates associated with each volume-
based tier can be known at the time a market participant trades, market
participants may not know which volume-based tier they will fall under
at the time of the trade (and thus the fee or rebate rate that will
apply to their particular trade) because the tier they will fall under
is typically determined based on their trading volume during the
current month, which is not finalized until the end of the month.\1084\
More specifically, the volume-based fees or rebates a market
participant receives from an exchange are often determined by a market
participant's average total daily traded share volume on the exchange
during the month as a percentage of either the average total daily
market volume reported by one of the consolidated tapes during the
month or as a percentage of the average total daily market volume
reported by all consolidated tapes during the month.\1085\ It is
therefore, as one commenter stated, difficult for market participants
to forecast trading costs.\1086\ Hence, market participants currently
typically have to make trading decisions without the ability to
determine their full trading costs.
---------------------------------------------------------------------------
\1081\ See table 4 for information on how often exchanges amend
their fees.
\1082\ See Proposing Release, supra note 11, at 80292.
\1083\ See Letter from Richard Steiner, Electronic Trading
Strategist, RBC Capital Markets, to Brent Fields, Secretary,
Commission (Oct. 16, 2018), available at https://www.sec.gov/comments/s7-05-18/s70518-4527261-176048.pdf (commenting on the
transaction fee pilot); see also the Fee Tiers Proposal, supra note
437.
\1084\ See Chester Spatt, Is Equity Market Exchange Structure
Anti-Competitive? (Dec. 28, 2020), available at https://www.cmu.edu/tepper/faculty-and-research/assets/docs/anti-competitive-rebates.pdf. However, not all exchanges offer volume-based tiers in
their fee structures. For example, LTSE does not charge fees to
transact. For exchanges like these, it is possible to determine with
certainty the cost to transact prior to executing a trade.
\1085\ The Equity Data Plans disseminate SIP data over three
separate networks: (1) Tape A for securities listed on the New York
Stock Exchange (``NYSE''); (2) Tape B for securities listed on
exchanges other than NYSE and Nasdaq; and (3) Tape C for securities
listed on Nasdaq. These tapes are referred to as the ``consolidated
tapes.'' The CTA Plan governs the collection, consolidation,
processing, and dissemination of last sale information for Tape A
and Tape B securities. The CQ Plan governs the collection,
consolidation, processing, and dissemination of quotation
information for Tape A and Tape B securities. Finally, the UTP Plan
governs the collection, consolidation, processing, and dissemination
of last sale and quotation information for Tape C securities. For
details on exchange volume-based fees and rebates, see, e.g., Add
and Remove Rates, Nasdaq, available at http://www.nasdaqtrader.com/Trader.aspx?id=PriceListTrading2; New York Stock Exchange Price List
2024, NYSE, available at https://www.nyse.com/publicdocs/nyse/markets/nyse/NYSE_Price_List.pdf; and Cboe U.S. Equities Fee
Schedules EDGX Equities, Cboe, available at https://www.cboe.com/us/equities/membership/fee_schedule/edgx/. See also the Fee Tiers
Proposal supra note 437 at 76284-88, describing Commission concerns
about the effect of tiers on competition among exchange members,
conflicts of interest between members and their customers, and
competition between exchanges.
\1086\ See Council of Institutional Investors Letter at 4
(stating ``It is our understanding that currently exchanges use
volume-based tier schedules that depend on the current month's
trading volume. As a result, the per-transaction fee or rebate
cannot be known when the trade occurs. This significantly impedes
the ability of institutional investors and other market participants
`to evaluate the total price of a trade at the time of execution and
. . . [the] ability to evaluate best execution and order routing.'
'').
---------------------------------------------------------------------------
Broker-dealers trading in an agency capacity may pass fees and
rebates received from exchanges while working a customer's order
through to the customer, either directly or through a change in the
commission charged for the order. One commenter stated that passing
through transaction fees and rebates is not common.\1087\ The
Commission is uncertain the extent to which transaction fees and
rebates are passed through to customers, or in what form. It is
possible that fees and rebates are passed through indirectly in the
form of payment for order flow or in price improvement. For example,
one broker-dealer's 606 reports, in discussing orders routed to a
wholesaler, acknowledged that exchange rebates may affect the
wholesaler's subsequent routing decision, but that those rebates could
be used to provide price improvement to the broker-dealer's customers
or order flow payments to the broker-dealer.\1088\ As discussed in the
Proposing Release, the Commission is uncertain of how much demand
currently exists for rebates to be passed through to end
customers.\1089\
---------------------------------------------------------------------------
\1087\ See Harris Letter at 12 (``Almost all retail and many
institutional brokers pay the taker fee and keep the maker rebate
when trading on behalf of their clients.'').
\1088\ See E*Trade.com, Morgan Stanley Smith Barney LLC--Held
NMS Stocks and Options Order Routing Public Report, 1st Quarter,
2024, at 2, available at https://cdn2.etrade.net/1/24043013500.0/aempros/content/dam/etrade/retail/en_US/documents/pdf/order-routing-reports/2024/606-MSWM-2024Q1.pdf.
\1089\ See Proposing Release, supra note 11, at 80330.
---------------------------------------------------------------------------
Market participants who are not themselves broker-dealers may
access information on exchange fees and rebates through reports
available under Rule 606. With respect to held orders, Rule 606(a)(1)
requires broker-dealers to produce quarterly public reports regarding
their routing of non-directed orders \1090\ in NMS stocks that are
submitted on a held basis. Along with other information, these reports
require the broker-dealer to report both the total dollar amount and
per share average of net transaction fees paid and net transaction
rebates received for different
[[Page 81694]]
order types for each trading venue to which the broker-dealer reports
routing orders.\1091\ Additionally, Rule 606(b)(3) requires broker-
dealers to produce reports pertaining to order handling upon the
request of a customer that places, directly or indirectly, one or more
orders in NMS stocks that are submitted on a not held basis, subject to
a de minimis exception.\1092\ For each venue to which the broker-dealer
routed the customer's orders, these reports require the broker-dealer
to disclose, among other things, the average net execution rebate or
fee for shares of orders providing liquidity and the average net
execution rebate or fee for shares of orders removing liquidity.\1093\
However, these reports provide market participants with information
only on historical average transaction fees and rebates and may not
accurately reflect the current exchange fees and rebates a market
participate will encounter at the time of its transaction.\1094\
---------------------------------------------------------------------------
\1090\ A ``non-directed order'' means any order from a customer
other than a directed order. See 17 CFR 242.600(b)(56). A ``directed
order'' means an order from a customer that the customer
specifically instructed the broker or dealer to route to a
particular venue for execution. See 17 CFR 242.600(b)(27).
\1091\ Rule 606(a)(1) requires broker-dealers to report separate
information for market orders, marketable limit orders, non-
marketable limit order, and other orders. See 17 CFR 242.606(a)(1)
for the items that need to be disclosed in reports under rule
606(a)(1).
\1092\ See 17 CFR 242.606(b)(3). In addition, under rule
606(b)(5)'s customer-level de minimis exception, broker-dealers need
not provide upon request execution quality reports for customers
that traded on average each month for the prior six months less than
$1,000,000 of notional value of not held orders in NMS stocks
through the broker-dealer. See 17 CFR 242.606(b)(5).
\1093\ See 17 CFR 242.606(b)(3)(iii) and (iv).
\1094\ Reports under rule 606(a)(1) are produced by broker-
dealers at the end of the quarter and disclose information on
average fees and rebates for each month in that quarter. Reports
issued by broker-dealers to their customers under rule 606(b)(3)
disclose summarized information on the handling of the customer's
orders for each calendar month over the prior six months. The
broker-dealer must issue these reports to the customer within seven
business days of receiving the customer's request.
---------------------------------------------------------------------------
c. Analysis of Current Access Fees and Rebates
The Commission analyzes, in table 4, current fee and rebate
schedules, based on Rule 19b-4 filings with the Commission, for each of
the equity exchanges operating in the United States as of February 8,
2024,\1095\ as well as the transaction prices that each exchange
posts.\1096\ What is apparent from this analysis is that the current
structure of fees and rebates is complex and constantly changing.\1097\
Each exchange, except LTSE which does not charge transaction fees,
filed an average of 13.6 Rule 19b-4 equity market fee filings with the
Commission in 2023. Market participants interacting with all exchanges
had to adjust to 218 total fee filings in 2023. Each filing can contain
changes for numerous fee and rebate categories.
---------------------------------------------------------------------------
\1095\ Table 4 is constructed using the same methodology as
table 5 of the Proposing Release, supra note 11, at 80311; the only
difference is that table 4 herein uses data as of Feb. 2024 and
computes fee revisions during the 2023 calendar year, whereas table
5 of the Proposing Release used data as of May 2022 and computed
annual fee revisions from 2018 to June of 2022. Any differences
between these two tables are due to changes in exchange fee
schedules from May 2022 to Feb. 2024.
\1096\ Panel A of table 4 provides the category of exchange,
maker-taker, inverted, or flat/free, the number of fee revisions
since Jan. 2018 as indicated by the number of transaction fee
specific rule 19b-4 filings that the exchange has filed with the
Commission, the date that each exchange's website states that the
fee schedule posted there is effective and the range of fees and
rebates along with the number of categories of fees and rebates for
transactions priced equal to, or greater than, $1.00 per share.
\1097\ Some commenters have also stated that current transaction
pricing practices introduce complexity into the market and reduce
transparency. See Themis Letter at 7; BMO Letter at 3.
---------------------------------------------------------------------------
The effect of the 30 mils fee cap as an anchor point is also
apparent. For most exchanges the maximum fee assessed, presumably for
non-protected quotes, is close to the 30 mils fee cap for protected
quotes. The maximum rebate is generally in the vicinity of 30 mils,
further suggesting the 30 mils access fee cap effectively limits what
the exchanges offer as rebates. Some exchanges offer different access
fees and rebate schedules for retail versus non-retail trades.\1098\
---------------------------------------------------------------------------
\1098\ See, e.g., New York Stock Exchange (NYSE), Equity Fees
and Charges, NYSE.com, available at https://www.nyse.com/publicdocs/nyse/markets/nyse-arca/NYSE_Arca_Marketplace_Fees.pdf (accessed June
18, 2024); see also Virtu Letter II at 23 suggesting that the
Commission could consider an alternative which would allow exchanges
to offer differing fee schedules to retail and non-retail orders.
---------------------------------------------------------------------------
Panel B of table 4 provides information on the exchange's fee
schedules for stocks priced lower than $1.00. For these transactions,
the fee schedules tend to be simpler. Most exchanges do not offer a
rebate for transactions lower than $1.00 even if the exchange offers
rebates for other transactions--only three exchanges offer any sort of
baseline rebate.\1099\ Additionally, the exchanges tend to charge an
access fee of 0.1% with some also charging the maximum access fee of
0.3% of the share price. Only one exchange charges a fee of 0.1% to
both sides of a transaction.
---------------------------------------------------------------------------
\1099\ The three are Cboe EDGX, MEMX, and MIAX Pearl.
Table 4--Summary of Transaction-Based Fee Schedules for U.S. National Equities Exchanges as of February 2024 a
--------------------------------------------------------------------------------------------------------------------------------------------------------
Number of Date of fee Rebates (# of
Exchange Fee model revisions 2023 schedule Fees (# of categories) categories)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Panel A: Fees and Rebates for Transactions Greater Than $1.00
--------------------------------------------------------------------------------------------------------------------------------------------------------
Cboe BZX \b\............................ Maker-Taker............... 42 1/2/2024 $0.0030 (1) $0.0016-$0.0031 (7)
Cboe BYX \c\............................ Inverted.................. 9 2/1/2024 $0.0012-$0.0020 (10) $0.0015-$0.0020 (4)
Cboe EDGA \d\........................... Inverted.................. 14 2/7/2024 $0.0000-$0.0030 (12) $0.0016-$0.0024 (5)
Cboe EDGX \e\........................... Maker-Taker............... 44 2/1/2024 $0.00275-$0.0030 (3) $0.0020-$0.0034 (8)
BX \f\.................................. Inverted.................. 6 2/1/2024 $0.0020-$-$0.0030 (2) $0.0005-$0.0018 (7)
Phlx (PSX) \g\.......................... Maker-Taker............... 4 2/1/2024 $0.0030 (1) $0.0020-$0.0033 (5)
Nasdaq \h\.............................. Maker-Taker............... 6 2/1/2024 $0.0030(1) $0.0013-$0.00305 (26)
NYSE Arca \i\........................... Maker-Taker............... 11 2/1/2024 $0.0000-$0.0030 (6) $0.0000-$0.0032 (6)
NYSE American........................... Maker-Taker............... 7 1/3/2024 $0.0025-$0.0030 (3) $0.0016-$0.0030 (3)
NYSE.................................... Maker-Taker............... 16 1/12/2024 $0.0000-$0.00275 (5) $0.0004-$0.0030 (11)
NYSE National........................... Inverted.................. 5 1/3/2024 $0.0022-$0.0029 (4) $0.0007-$0.0030 (5)
NYSE Chicago............................ Maker-Taker............... 3 1/8/2024 $0.0010-$0.0030 (1) $0.0000 (0)
IEX \j\................................. Maker-Taker............... 2 1/24/2024 $0.0000-$0.0010 (2) $0.0004 (1)
MEMX \k\................................ Maker-Taker............... 22 2/1/2024 $0.00295-$0.0030 (2) $0.0015-$0.0033 (6)
MIAX Pearl \l\.......................... Maker-Taker............... 10 1/17/2024 $0.0024 (1) $0.00295 (1)
LTSE \m\................................ Free...................... NA N/A $0.0000 (1) $0.0000 (1)
--------------------------------------------------------------------------------------------------------------------------------------------------------
[[Page 81695]]
----------------------------------------------------------------------------------------------------------------
Exchange Fee Model Rebate Fee (%) Charged both sides
----------------------------------------------------------------------------------------------------------------
Panel B: Fees and Rebates for Transactions Under $1.00
----------------------------------------------------------------------------------------------------------------
Cboe BZX......................... Maker-Taker........ 0.................. 0.30
Cboe BYX......................... Inverted........... 0.................. 0.10
Cboe EDGA........................ Inverted........... 0.................. 0
Cboe EDGX........................ Maker-Taker........ 0.00009 (per share) 0.30
BX............................... Inverted........... 0.................. 0.10
Phlx (PSX)....................... Maker-Taker........ 0.................. 0.30
Nasdaq........................... Maker-Taker........ 0.................. 0.30
NYSE Arca........................ Maker-Taker........ 0.................. 0.10
NYSE American.................... Maker-Taker........ 0.................. 0.10
NYSE............................. Maker-Taker........ 0.................. 0.10
NYSE National.................... Inverted........... 0.................. 0
NYSE Chicago..................... Maker-Taker........ 0.................. 0.10 Yes.
IEX.............................. Maker-Taker........ 0.................. 0.09
MEMX............................. Maker-Taker........ 0.075% (of value).. 0.10
MIAX Pearl....................... Maker-Taker........ 0.15% (of value)... 0.250
LTSE............................. Free............... 0.................. 0
----------------------------------------------------------------------------------------------------------------
\a\ The number of fee revisions is obtained by counting each Rule 19b-4 filing for each exchange that is not
clearly marked for a non-transaction fee related purpose such as connectivity fees, listing fees, options
fees, etc. To determine the fee and rebate information, the staff searched each exchange's webpage for its
current posted access fee and rebate schedule and collected information on access fees and rebates pertaining
to non-auction trading in stocks priced equal to, or greater than, $1.00 per share. Sources for Current Access
Fee Data were effective on the dates shown in panel A of table 4, and were accessed during February 2024 at
the websites shown beneath the table.
\b\ https://www.cboe.com/us/equities/membership/fee_schedule/bzx/.
\c\ https://www.cboe.com/us/equities/membership/fee_schedule/byx/.
\d\ https://www.cboe.com/us/equities/membership/fee_schedule/edga/.
\e\ https://www.cboe.com/us/equities/membership/fee_schedule/edgx/.
\f\ https://www.nasdaqtrader.com/trader.aspx?id=bx_pricing.
\g\ https://www.nasdaqtrader.com/trader.aspx?id=psx_pricing.
\h\ https://www.nasdaqtrader.com/Trader.aspx?id=PriceListTrading2.
\i\ All NYSE Exchange Family fees: https://www.nyse.com/markets/fees.
\j\ https://exchange.iex.io/resources/trading/fee-schedule/. (Note: that the majority of IEX trading occurs via
non-displayed orders. IEX only pays rebates on displayed orders.)
\k\ https://info.memxtrading.com/equities-trading-resources/us-equities-fee-schedule/.
\l\ https://www.miaxglobal.com/sites/default/files/fee_schedule-files/MIAX_Pearl_Equities_Fee_Schedule_01172024.pdf.
\m\ https://ltse.com/trading/faqs.
[[Page 81696]]
Complex fee schedules and volume-based tiers mean that it is
difficult for the Commission to determine the net capture on a given
exchange (the difference between average fees levied and rebates
paid).\1100\ Additionally, financial statements for exchange groups
generally do not break down performance on a per-venue level and the
financial statements generally combine auction access fees collected
with regular trading access fees. Furthermore, some exchanges are
privately held and thus do not release the same financial statements
that public exchanges do. Using information from the financial
statements of the three major exchange groups which collectively
account for the overwhelming majority of trading volume on exchanges,
the Commission estimates that the average total net capture is around 4
mils for all trading types.\1101\ However, the Commission understands
based on staff conversations with industry members that the net capture
for non-auction trading in stocks that have a price equal to or greater
than $1.00 is likely close to 2 mils on most exchanges,\1102\ and in
the analysis in later sections where the net capture needs to be
assumed, we use 2 mils unless otherwise stated.\1103\ A commenter
agreed stating that the net fee is typically only 2 mils per share,
stating, ``it is worth pausing to reflect on just how competitive this
net fee is--this is about as close to Bertrand price competition as one
sees.'' \1104\ Another commenter reported estimated trading related
``all-in'' costs to trade ranging 1.9 to 3.4 mils over the 2017 to 2021
period for the three largest exchange groups (Cboe, Nasdaq, and NYSE);
\1105\ these estimates are broadly in line with the estimated net-
capture rates and consistent with the 2 mil net capture rate assumption
used in the subsequent analysis.\1106\ Given the low net capture rate
of 2 mils on most exchanges, the primary reason that access fees remain
near 30 mils on most exchanges is likely to fund rebates.\1107\ For
stocks trading below $1.00 the Commission estimates an average net
capture of around 0.24% of the transaction volume.\1108\ This amount is
close to the 0.30% access fee cap and arises because, as seen in panel
B of table 4, much of the trading for sub $1.00 priced volume takes
place on exchanges which set their baseline fee at or near 0.30% but do
not offer baseline rebates for transactions under $1.00. On a per-share
basis, this net capture of 0.24% of transaction dollar volume
corresponds to a net capture of 7.3 mils.\1109\
---------------------------------------------------------------------------
\1100\ Volume-based tiers imply that the net capture varies by
exchange member. To calculate an exchange's net capture would
require knowing the number and types of orders that are executed by
each member, mapping these orders to the exchange's fee schedule,
calculating the net capture for each member, and then aggregating
the net capture over all exchange members.
\1101\ Intercontinental Exchange, the parent firm of NYSE,
reports on page 53 of its 2021 Form 10-K filing that their net
capture for U.S. equity transactions was approximately 4.2 mils in
2021. Nasdaq did not report its net capture in its Form 10-K filing,
however Nasdaq provides information on its investor relations web
page which, when we average the relevant 2021 volumes, indicates
that the average net capture across all Nasdaq platforms for U.S.
equity transactions was 5.9 mils. See Nasdaq 2022/2021 Monthly
Volumes, Nasdaq, available at https://ir.nasdaq.com/static-files/465d2157-c476-4546-a9f7-8d7ad0c9be77). Cboe reports in its Form 10-K
filing that its net capture for U.S. equity transactions was
approximately 2 mils.
\1102\ Non-auction orders exclude opening, closing, and
reopening auctions. See table 7, note a for additional details
regarding which orders are considered for estimation.
\1103\ One commenter stated that exchanges subsidize rebates
with other sources of revenue as manifest by the fact that some
market participants could receive rebates in excess of the 30 mil
fee cap. See Healthy Markets Letter I at 23. Table 4 presents
evidence consistent with the notion that in some cases rebates
received may be in excess of 30 mils. However, the commenter did not
provide any analysis to suggest that the net capture of the
exchanges was, on average, negative. As discussed, the Commission
believes that most exchanges, on average, earn approximately 2 mils
per transaction priced greater than $1.00. One exception is IEX
which earns an estimated 6 mils based on the information in table 4.
\1104\ See Budish Letter at 3. Bertrand competition is an
economics model of competition on the basis of prices whereby firms
set their price--net price in this instance--at marginal cost.
\1105\ The commenter also included similar estimates for IEX
which ranged from 6.8-8.9 mils, see Nasdaq Letter II at 3.
\1106\ Subsequent analysis in section VII.D.2 assumes that the
assumed 2 mil net capture will continue to be valid following the
implementation of the amendments.
\1107\ See Retirement Coalition Letter at 1 (``in practice, this
`cap' has come to be used as the standard rate charged to access
quotes at most exchanges, and almost all of those fees are then
`rebated' to liquidity providers'').
\1108\ The estimate for the 0.24% net capture is obtained by
taking the total estimated net transaction fee across all exchanges
for trading in shares priced below $1.00 ($83.2 million) and
dividing this number by the total sub $1.00 dollar volume from Panel
B of table 5 below ($34.2 billion). 100*83.2 million/34.2 billon
%0.24.
\1109\ To estimate the net capture in terms of mils, the
Commission divides the dollar revenue from fees by the number of
shares traded. The dollar net capture is 0.24%, see id.,) and the
dollar volume is $34.2 billion, see table 5, Panel B, resulting in a
dollar revenue of $82 million (0.0024*$34.2 billion). With share
volume of 112.6 billion, see table 5, Panel A, the net capture is
7.3 mils ($82 million/112.6 billion).
---------------------------------------------------------------------------
Table 5 presents tabulations of the total share (Panel A) and
dollar (Panel B) trading volume executed on the 16 exchanges in
2023.\1110\ This table provides estimates for the total volume that
executed below $1.00, and that which executed above $1.00. These
numbers represent an estimate of the total number of shares that will
have been subject to the access fees and rebates discussed in this
release.
---------------------------------------------------------------------------
\1110\ Table 5 is constructed using the same methodology as
table 6 of the Proposing Release, supra note 11, at 80313; table 5
herein uses data for 2023, whereas table 6 of the Proposing Release
used data for the first half of 2022.
Table 5--Trading Volume by Exchange, Exchange Type 2023 \a\
----------------------------------------------------------------------------------------------------------------
>=$1 Volume >=$1 Volume QS
Exchange name Exchange type <$1 Volume QS<= $0.015 > $0.015 % of Exchange
(billions) (billions) (billions) volume
----------------------------------------------------------------------------------------------------------------
Panel A: Share Volume
----------------------------------------------------------------------------------------------------------------
Off-Exchange.................. ................ 192.9 620.3 241.3
Nasdaq........................ Maker-Taker..... 26.8 226.2 88.9 26.9
NYSE Arca..................... Maker-Taker..... 27.5 139.3 30.4 15.5
NYSE.......................... Maker-Taker..... 3.4 127.4 37.4 13.2
Cboe BZX...................... Maker-Taker..... 14.0 86.3 20.9 9.5
Cboe EDGX..................... Maker-Taker..... 21.1 98.2 23.3 11.2
MEMX.......................... Maker-Taker..... 8.1 61.9 12.0 6.5
IEX........................... Maker-Taker..... 1.7 38.9 19.5 4.7
Cboe EDGA..................... Inverted........ 2.7 33.8 5.6 3.3
Cboe BYX...................... Inverted........ 2.6 20.4 2.8 2.0
MIAX Pearl.................... Maker-Taker..... 2.4 41.9 2.7 3.7
NYSE National................. Inverted........ 0.5 11.0 1.4 1.0
[[Page 81697]]
Nasdaq OMX PSX................ Maker-Taker..... 0.3 7.5 1.9 0.8
Nasdaq OMX BX................. Inverted........ 0.5 6.7 2.6 0.8
NYSE American................. Maker-Taker..... 1.0 4.8 1.0 0.5
NYSE Chicago.................. Flat............ 0.2 0.8 1.8 0.2
LTSE.......................... Flat............ 0.0 0.0 0.0 0.0
---------------------------------------------------------------------------------
Total..................... ................ 305.5 1,525.4 493.5
Exchange Total............ ................ 112.6 905.1 252.2
----------------------------------------------------------------------------------------------------------------
Panel B: Dollar Volume
----------------------------------------------------------------------------------------------------------------
Off-Exchange.................. ................ 65.3 19,744.9 24,508.4
Nasdaq (TapeC)................ Maker-Taker..... 9.1 9,008.3 9,402.3 30.4
NYSE Arca..................... Maker-Taker..... 7.6 6,433.3 3,107.1 15.8
NYSE.......................... Maker-Taker..... 1.5 3,985.9 3,658.6 12.6
Cboe BZX...................... Maker-Taker..... 3.1 3,711.6 2,479.9 10.2
Cboe EDGX..................... Maker-Taker..... 6.2 3,450.0 2,371.1 9.6
MEMX.......................... Maker-Taker..... 2.5 2,148.8 1,130.4 5.4
IEX........................... Maker-Taker..... 0.8 1,383.8 2,119.0 5.8
Cboe EDGA..................... Inverted........ 1.0 1,038.0 494.8 2.5
Cboe BYX...................... Inverted........ 0.9 657.2 275.3 1.5
MIAX.......................... Maker-Taker..... 0.7 1,297.3 268.4 2.6
NYSE National................. Inverted........ 0.2 287.3 131.0 0.7
Nasdaq OMX PSX................ Maker-Taker..... 0.1 368.9 211.9 1.0
Nasdaq OMX BX................. Inverted........ 0.2 274.9 258.7 0.9
NYSE American................. Maker-Taker..... 0.4 165.5 97.7 0.4
NYSE Chicago.................. Flat............ 0.1 39.6 236.0 0.5
LTSE.......................... Flat............ 0.0 1.5 1.6 0.0
---------------------------------------------------------------------------------
Total..................... ................ 99.4 53,996.9 50,752.2
---------------------------------------------------------------------------------
Exchange Total............ ................ 34.2 34,252.0 26,243.9
----------------------------------------------------------------------------------------------------------------
\a\ This table aggregates all trade information from the TAQ database for every trading day in 2023. Only
trading volume reflecting normal trades during regular trading is included. Normal trades are identified in
TAQ data by sale conditions ``blank, @, E, F, I, S, Y'' which correspond to regular trades, intermarket sweep
orders, odd-lot trades, split trades, and yellow flag regular trades. The remaining share volume was
aggregated by exchange, and the table denotes exchange type (maker-taker, inverted, flat, free). Share and
dollar volume from exchange codes T and Q were combined into `Nasdaq.' Panel A presents share volume totals
and panel B presents dollar volume totals. Certain items in table 5 may also be affected by the MDI Rules once
they are fully implemented. See infra section VII.C.3.
Transaction fees for trades in stocks priced equal to or greater
than $1.00 are generally levied per share transacted. From table 5 we
see that in 2023, there were approximately 2 trillion shares transacted
at prices equal to or greater than $1.00 per share across all venues,
57% of which (1.16 trillion shares) were executed on a registered
exchange.\1111\ Of these on-exchange transactions priced equal to or
greater than $1.00 per share, approximately 78% were in stocks with
quoted spreads of $0.015 or less.\1112\ These numbers provide the basis
for estimating the total amount of access fees and rebates collected
and distributed in transactions priced equal to, or greater than, $1.00
per share. For transactions less than $1.00 per share the access fee is
generally levied as a percent of the transaction share price. In panel
B we see that in 2023 there was approximately $34 billion transacted on
exchanges in shares priced less than $1.00 per share.
---------------------------------------------------------------------------
\1111\ 2T [equiv] shares 1.5T narrow spread shares + 493 billion
wider spread shares. Also, off-exchange trading volume has increased
in recent years. See, e.g., Jonathan Brogaard & Jing Pan, Dark Pool
Trading and Information Acquisition, 35 Rev. Fin. Stud. 2625 (2022).
\1112\ The fourth column of Panel A shows 905.1 billion shares
traded on exchanges with a price greater than or equal to $1.00 and
a quoted spread of $0.015 or less; the fifth column shows 252.2
billion shares traded on exchanges with a price greater than or
equal to $1.00 and a quoted spread over $0.015. The total number of
shares traded on exchanges with a price greater than or equal to
$1.00 is therefore 1157.3 billion (905.1+252.2), and the fraction of
these that a spread of $0.015 or less is 78% (905.1/1157.3).
---------------------------------------------------------------------------
Panels A and B of table 6 break down the share and dollar volume
statistics presented in table 5 by venue type: maker-taker, inverted,
and flat/free.\1113\ The overwhelming majority (over 90%) of both
dollar and share exchange trading volume occurs on maker-taker venues.
Inverted exchanges capture about 5-7% of dollar and share volume, and
the remaining share volume transact on flat/free exchanges.
---------------------------------------------------------------------------
\1113\ Table 6 is constructed using the same methodology as
table 7 of the Proposing Release, supra note 11, at 80314. The only
difference is that table 6 herein uses data for 2023, whereas table
7 of the Proposing Release used data for the first six months of
2022.
[[Page 81698]]
Table 6--Volume by Exchange Type and Estimated Access Fee/Rebate Estimates 2023 \a\
----------------------------------------------------------------------------------------------------------------
Price>$1; TWAQS Price>$1; TWAQS >
Price<$1 <= $0.015 $0.015 % Total
(billions) (billions) (billions)
----------------------------------------------------------------------------------------------------------------
Panel A: Exchange Share Volume by Venue Type
----------------------------------------------------------------------------------------------------------------
Maker-Taker......................... 106.2 832.4 237.9 92.7
Inverted............................ 6.2 71.9 12.4 7.1
Flat/Free........................... 0.2 0.8 1.8 0.2
----------------------------------------------------------------------------------------------------------------
Panel B: Exchange Dollar Volume by Venue Type
----------------------------------------------------------------------------------------------------------------
Maker-Taker......................... 31.9 31,953.4 24,846.4 93.9
Inverted............................ 2.2 2,257.4 1,159.8 5.6
Flat/Free........................... 0.1 41.1 237.6 0.5
----------------------------------------------------------------------------------------------------------------
Panel C: Estimated Fees Collected and Rebates Distributed (Billions)
----------------------------------------------------------------------------------------------------------------
Fees Collected...................... ................. $3.41 ................. .................
Rebates Distributed................. ................. $3.08 ................. .................
Exchange Capture.................... ................. $0.34 ................. .................
----------------------------------------------------------------------------------------------------------------
Panel D: Total Estimated Net Fees by Liquidity Type (Billions)
----------------------------------------------------------------------------------------------------------------
Demander............................ ................. $2.97 ................. .................
Provider............................ ................. ($2.63) ................. .................
Exchange Capture.................... ................. $0.34 ................. .................
----------------------------------------------------------------------------------------------------------------
\a\ Certain items in this table 6 may also be affected by the amendments in the MDI Rules once they are fully
implemented. See infra section VII.C.3.
Panel C provides an estimate of the total amount of access fees
collected and rebates distributed.\1114\ In 2023 there were an
estimated $3.41 billion in access fees collected across all exchanges
and $3.08 billion in rebates distributed, resulting in a net capture to
all exchanges of $340 million.
---------------------------------------------------------------------------
\1114\ These estimates are computed by assuming a 30 mil access
fee and 28 mil rebate on all transactions that occur on maker-taker
or inverted exchanges and a 10 mil access fee (and 4 mil rebate) on
the volume priced equal to, or greater than, $1.00 per share that
occurs on IEX. For trading in sub $1.00 transactions, the various
access fees and rebates for each exchange presented in Panel B of
table 4 are multiplied by the corresponding dollar volume of trade
in transactions priced less than $1.00 per share to compute the
total access fees collected and rebates distributed for this volume.
The figures are summed together to provide the estimates of total
access fees collected and rebates distributed.
---------------------------------------------------------------------------
Panel D of table 6 provides estimates of the net access fee paid by
liquidity demanders and liquidity suppliers.\1115\ In 2023 liquidity
demanders paid an estimated $2.97 billion in net access fees and
liquidity providers received an estimated $2.63 billion in rebates. The
difference of $340 million is the exchanges' estimated net capture.
---------------------------------------------------------------------------
\1115\ This estimate presumes that for shares transacted in
prices equal to or greater than $1.00 per share on maker-taker
venues the liquidity demander pays a 30 mil access fee and the
liquidity provider receives a 28 mil rebate. On inverted exchanges
the opposite occurs. On IEX it is presumed that liquidity demanders
pay an 8 mil access fee and liquidity providers receive no rebate.
For trading in sub $1.00 transactions the various access fees and
rebates for liquidity suppliers and demanders are computed by taking
the respective fees and rebates for sub $1.00 transactions for each
exchange presented in Panel B of table 4 and multiplying them by the
corresponding dollar volume of trade in transactions priced less
than $1.00 to compute the total access fees collected and rebates
distributed for liquidity-providing and demanding trades. The
figures are summed together to provide the estimates of total access
fees collected and rebates distributed.
---------------------------------------------------------------------------
Although not subject to Rule 610(c), because they do not post
protected quotes, ATSs also often assess transaction fees.\1116\ As of
the third quarter of 2023 there were 32 ATSs that reported trading
volume to FINRA transacting a total of 73 billion shares.\1117\ Unlike
exchanges, the fees that ATSs charge generally do not have a standard
structure and are often negotiated between the ATS and the customer.
Based on a review of item 19 in form ATS-N, ATSs generally do not
provide rebates, and when transaction fees are explicitly discussed,
they are often in the range of 10 mils.\1118\
---------------------------------------------------------------------------
\1116\ IntelligentCross ATS, for example, offers matching
processes for all NMS stocks eligible for trading, and disseminates
bids and offers in real-time to subscribers to the ATS's proprietary
data feed, but these are not protected quotes. See IntelligentCross,
Form ATS-N, Item 15 (Display) (dated Apr. 11, 2022) available at
https://www.sec.gov/Archives/edgar/data/1708826/000170882622000002/xslATS-N_X01/primary_doc.xml.
\1117\ See FINRA, ATS Transparency Data Quarterly Statistics,
available at https://www.finra.org/filing-reporting/otc-transparency/ats-quarterly-statistics.
\1118\ See infra note 1442 for commenter discussion on ATS
transaction fees. See also IEX Letter VI at 5 for additional
analysis supporting the conclusion that 10 mils is a representative
transaction fee among ATSs.
---------------------------------------------------------------------------
Table 4 indicates that many exchanges charge the maximum allowed
fee, rebating nearly all of it as a compensation for liquidity
provision. One commenter states, ``access fees have been uniquely
impervious to market forces.'' \1119\ Rule 611 generally causes
marketable orders to be routed to those markets displaying the best-
priced quotations.\1120\ As discussed in section
[[Page 81699]]
VII.B.3, a liquidity provider is generally indifferent between
receiving compensation in the form of a rebate or in the form of a
quoted spread, implying that an exchange can use rebates to induce
quoting lower spreads, and hence the best prices.\1121\ The exchange
can fund the rebate with an access fee charged to the liquidity
demander, relying on Rule 611 to reduce the loss of liquidity demanding
customers that would otherwise occur from such an increase in
prices.\1122\ The exchanges profit from the difference between the
access fees collected and the rebates paid. Were exchanges to
unilaterally lower their access fees and rebates (without other
exchanges making similar changes), liquidity providers would likely
route their orders to another exchange.\1123\ Notably, research
surrounding a Nasdaq experiment where it unilaterally lowered fees and
rebates found that Nasdaq lost market share to other maker-taker venues
with a higher rebate.\1124\ Table 4 also shows that even the maximum
rebates are close to the access fees; doing otherwise would likely be
unprofitable or risky.\1125\
---------------------------------------------------------------------------
\1119\ See IEX letter V at 2, 3. See also Proposing Release,
supra note 11, at 80305.
\1120\ More specifically, Rule 611 requires trading centers to
have policies and procedures that reasonably prevent trade-throughs.
See Proposing Release, supra note 11, at 80286. A trade-through is a
trade that executes at a price lower than a protected bid or higher
than a protected offer. The NBBO is set by the best protected bid
and offer; therefore, a trade that executes outside the NBBO is a
``trade-through.'' If an exchange does not have a limit order at the
best quote, then it cannot execute against an incoming marketable
order; rather the exchange would generally need to cancel the order
or route it to another exchange with the best quote. The routing of
marketable orders is prevalent. A recently published academic
article finds that 34% of market orders sent to the NYSE in 2010-11
are routed. See Sida Li et.al., Refusing the Best Price? 2 J. FIN.
ECON. 147 (February 2023). For example, suppose two liquidity
providers want to sell a share in exchange for $10.002, net of fees
and rebates. Suppose the first seller posts at exchange X, which
offers a 30mil rebate, while the second seller posts at exchange Y,
which offers a 10mil rebate. The seller on exchange X is willing to
quote at $10.00 to receive a net price of $10.003, while the seller
on Y is not willing to quote at $10.00 because the net price would
be only $10.001--the seller on Y must quote at $10.01. Rule 611 will
therefore direct marketable orders to the lower quoted price at
exchange X.
\1121\ See Proposing Release, supra note 11, at 80305 (``the
NBBO restricts the routing behavior of marketable orders and often
forces liquidity demanders to pay the access fee to trade against a
NBBO order. Exchanges are thus incentivized to attract more
competitively priced liquidity with large rebates, which are funded
by similarly large access fees, in order to capture more trading
volume''). The high rebate allows the liquidity supplier to offer a
better quoted price--i.e., a higher bid or a lower offer--because
the liquidity supplier only cares about the total proceeds from the
sale (the liquidity supplier does not care whether the proceeds take
the form of a rebate).
\1122\ That is, the need to execute against the protected quote
first, before executing at other prices, would maintain a strong
incentive for broker-dealers to route orders to the exchange, even
in the face of high access fees.
\1123\ See Proposing Release, supra note 11, at 80305 n.457.
\1124\ See id.; see also Yiping Lin, et al., A Model of Maker-
Taker Fees and Quasi-Natural Experimental Evidence (working paper
Feb. 8, 2021), available at https://ssrn.com/abstract=3279712
(retrieved from SSRN Elsevier database). Consequently, it could be
harmful to an exchange to unilaterally reduce access fees and their
associated rebates if other exchanges do not follow suit. Further,
even if each of the exchanges lowered its fees, there would be the
risk that a new exchange would see the opportunity and enter the
market with high fees and rebates and thus capture market share,
inducing the other exchanges to abandon their low fee models to
remain competitive.
\1125\ See discussion in section VII.D.2.b.
---------------------------------------------------------------------------
As discussed in the Proposing Release, and as table 4 shows, the
NYSE, Nasdaq, and Cboe exchange families each operate both a maker-
taker venue as well as an inverted venue,\1126\ Commenters state that
inverted venues can be used to achieve intra-tick pricing.\1127\
Specifically, the net price of a trade is the quoted price adjusted for
exchange fees and rebates; the quoted price is constrained by the tick,
but the net price can be between ticks.\1128\ To the extent that intra-
tick pricing on inverted venues is a solution to the quoted price being
constrained by the tick, it is a costly one. First, quote protection
applies to the quoted bid or ask, not the net cost, implying that
routing a market order to an inverted venue runs the risk of the order
being routed elsewhere if the inverted venue is not at the NBBO.
Research shows that inverted venues are less likely to be at the NBBO,
a result that follows from revenue from rebates and from spreads being
interchangeable assuming the market participant receives both.\1129\ As
explained in section VII.D.1.b.ii, this leads to delays and increased
cost.
---------------------------------------------------------------------------
\1126\ See Proposing Release, supra note 11, at 80313.
\1127\ See Larry Harris, Quarter Penny Tick (working paper, Mar.
9, 2022) attached to Letter from Larry Harris (``Quarter Penny
Tick'').
\1128\ Suppose an exchange family operates both a maker-taker
venue and an inverted venue; further suppose that the maker-taker
venue offers a 30mil rebate to liquidity suppliers, while the
inverted venue charges a 30mil fee to liquidity suppliers. Liquidity
suppliers would therefore be able to transact at two different net
prices within the tick--e.g., the liquidity supplier could offer to
sell at $10.00 on the maker-taker venue, which would result in a net
price of $10.003; or the liquidity supplier could offer to sell at
$10.00 on the inverted venue to net $9.997. The variation in fee
schedules across the venues therefore allows for intra-tick pricing.
See also supra section VII.C.2.b for a discussion of current state
of the fees and rebates and the variation in pricing structure
across exchanges.
\1129\ See IEX Letter I at 14.
---------------------------------------------------------------------------
Second, the existence of inverted venues fragments liquidity
compared to the situation where an exchange is able to offer orders to
be placed at multiple prices within the quoted spread. One commenter
agreed when discussing inverted venues, stating that: ``different
exchanges are optimal to use for different prices within the penny,
which is a recipe for artificial fragmentation, a confusing paper
trail, and overall excess complexity. Excess complexity, in turn, is a
recipe for excess rents, agency conflict and distrust.'' \1130\
---------------------------------------------------------------------------
\1130\ See Budish Letter at 4.
---------------------------------------------------------------------------
Lastly, one commenter stated that exchanges could circumvent
barriers associated with the tick size and access fees through
innovation, such as new order types.\1131\ The commenter stated that:
``Lastly, if the tick size were really a significant barrier to
competition for exchanges, they could innovate solutions to solve for
this. For example, exchanges could develop an order type that functions
within the current structure (limit order pricing and priority-ranked
based on even penny ticks) but where an order could provide sub-penny
price improvement if matched to a marketable order from a counterparty
that met certain objective conditions (such as being sourced from a
retail customer).'' \1132\ The commenter proceeded to describe a second
novel order type that could allow for sub-penny price improvement
through reduced access fees for retail investors.\1133\ The order types
the commenter lists as examples appear to solve for the specific
problem of retail investors achieving price improvement on exchange, a
solution that already exists through RLPs (though which do not have
significant volume).\1134\ It is possible that a new order type
designed to mitigate this problem might not work as intended. In
contrast, allowing for more ticks in certain stocks as the Commission
is adopting is a more straightforward and predictable way of achieving
the same ends without requiring the need for order types designed to
allow for sub-tick executions on exchanges.
---------------------------------------------------------------------------
\1131\ See Virtu Letter II at 23.
\1132\ See id.
\1133\ See id.
\1134\ See supra section VII.C.1.a for further discussion of
exchange RLP programs.
---------------------------------------------------------------------------
3. Round Lots, Odd-Lots, and Market Data Infrastructure
Currently, information on odd-lot quotes inside the NBBO is
available only to investors who subscribe to proprietary data feeds,
and comprehensive odd-lot information is only available to market
participants who subscribe to the proprietary data feeds of all the
exchanges. The implementation of the MDI Rules will include odd-lot
information inside the NBBO.\1135\ The MDI Rules also defined a round
lot, which previously had not been defined in a Commission rule.
Specifically, the MDI Rules establish a uniform round lot size of 100
shares for stocks priced $250 or less; 40 shares for stocks priced
greater than $250 and less than or equal to $1,000; 10 shares for
stocks priced greater than $1,000 and less than or equal to $10,000;
finally, 1 share for stocks priced greater than $10,000. These
amendments modify the round lot definitions set by the MDI Rules by
changing the evaluation period in which a stock's share price is
measured. The MDI Adopting Release defined round lots based on the
stock's average price in the preceding month--i.e., a stock's round lot
was updated every month based on the most recent month's data. These
amendments
[[Page 81700]]
update round lots every six months, with the round lot determined by
the stock's average price with a one-month lag--i.e., a stock's round
lot is updated in May of every year using its average stock price in
March, and the round lot is updated again in November using its average
stock price in September.
---------------------------------------------------------------------------
\1135\ See supra section VI.C and section V.E for discussions on
the expected time of the implementation of the MDI Rules.
---------------------------------------------------------------------------
In the MDI Adopting Release, the Commission established a
transition period for the implementation of the MDI Rules.\1136\ The
Commission's approval of the MDI Plan Amendments will be the starting
point for the rest of the MDI implementation schedule.\1137\ After
approval of the MDI Plan Amendments, the next step will be a 180-day
development period, during which competing consolidators can register
with the Commission.\1138\ Based on the times provided in the
transition plan for implementation of the MDI Rules, the Commission
estimated that the full implementation of the MDI Rules will be at
least two years after the Commission's approval of the plan
amendment(s) required by Rule 614(e).\1139\
---------------------------------------------------------------------------
\1136\ See MDI Adopting Release, supra note 10, at 18698-18701.
\1137\ See id. at 18698.
\1138\ See id. at 18699-18700.
\1139\ See id. at 18700-18701.
---------------------------------------------------------------------------
The Operating Committees of the CTA/CQ Plan and UTP Plan filed the
MDI Plan Amendments on November 5, 2021.\1140\ The Commission
disapproved the proposed amendments on September 21, 2022.\1141\ As a
result, the participants to the effective national market system
plan(s) will need to develop and file new proposed amendments as
required by Rule 614(e),\1142\ before the implementation period
prescribed by the phased transition plan can commence. Because the
implementation of the MDI Rules has been delayed, the end date of the
implementation period cannot be estimated with certainty.
---------------------------------------------------------------------------
\1140\ The Operating Committees of CTA Plan and UTP Plan filed
proposed amendments on Nov. 5, 2021, which were published for
comment in the Federal Register. See Securities Exchange Act Release
Nos. 93615 (Nov. 19, 2021), 86 FR 67800 (Nov. 29, 2021); 93625 (Nov.
19, 2021), 86 FR 67517 (Nov. 26, 2021); 93620 (Nov. 19, 2021), 86 FR
67541 (Nov. 26, 2021); 93618 (Nov. 19, 2021), 86 FR 67562 (Nov. 26,
2021).
\1141\ See Securities Exchange Act Release Nos. 95848 (Sept. 21,
2022), 87 FR 58544 (Sept. 27, 2022); 95849 (Sept. 21, 2022), 87 FR
58592 (Sept. 27, 2022); 95850 (Sept. 21, 2022), 87 FR 58560 (Sept.
27, 2022); 95851 (Sept. 21, 2022), 87 FR 58613 (Sept. 27, 2022).
\1142\ The Commission ordered the exchanges and FINRA to file a
new plan regarding consolidated market data on Sept. 1, 2023. On
Jan. 19, 2024, the Commission published notice of filing of a
National Market System Plan for Consolidated Equity Market Data. See
supra note 78.
---------------------------------------------------------------------------
The following discussion reflects the Commission's assessment of
the anticipated economic effects of the MDI Rules described in the MDI
Adopting Release as they relate to the baseline for the adoption of
these amendments.\1143\ The MDI Rules are part of the regulatory
baseline for this rule because they have been adopted. Given that the
MDI Rules have not yet been implemented, they have not affected market
practice and therefore data that would be required for a quantitative
analysis of a baseline that includes the effects of the MDI Rules is
not available. It is possible that the baseline for this rule, and
therefore the economic effects relative to the baseline, could be
different depending on how the MDI Rules are implemented.\1144\
---------------------------------------------------------------------------
\1143\ See MDI Adopting Release, supra note 10, at 18741-18799.
\1144\ Commission staff will review and study the effects of the
amendments adopted herein. See the introduction to section VII.D.
---------------------------------------------------------------------------
When adopting the MDI Rules, the Commission enumerated numerous
economic effects specifically related to changing the round lot
definition and including odd-lot information as a part of core data.
For the change in the definition of round lots, these effects included:
(1) a mechanically tighter NBBO for higher priced stocks due to the
redefinition of the round lot sizes,\1145\ (2) increased transparency
and better order execution,\1146\ and (3) potentially more orders for
high priced stocks being routed to exchanges instead of ATSs.\1147\ The
costs of changing the round lot definition included upgrading systems
to account for additional message traffic, and modifying and
reprogramming systems.\1148\ The Commission also discussed the expected
effect that changing the round lot definition will have on other rules
and regulations.\1149\
---------------------------------------------------------------------------
\1145\ See MDI Adopting Release, supra note 10, at 18743 for the
full discussion of the effect of changing the round lot size on the
NBBO.
\1146\ See MDI Adopting Release, supra note 10, at 18744, 18747
for the full discussion of the effect of changing the round lot size
on transparency and execution quality.
\1147\ See MDI Adopting Release, supra note 10, at 18747 for the
full discussion of the effect of changing the round lot size on
exchange competition and order routing.
\1148\ See MDI Adopting Release, supra note 10, at 18748 for the
full discussion of the expected costs of changing the round lot
size.
\1149\ See MDI Adopting Release, supra note 10, at 18749 for the
full discussion of the effect of changing the round lot size on
other rules and regulations.
---------------------------------------------------------------------------
For the inclusion of odd-lot information inside the NBBO in core
data,\1150\ these effects include reducing information asymmetries
between investors who currently have access to odd-lot information
through proprietary data feeds and those who do not, leading to better
order execution and price efficiency.\1151\ Providing an alternative to
proprietary data for some market participants will allow these market
participants to reduce data expenses required for trading.\1152\ The
costs of including odd-lot information inside the NBBO include: \1153\
the cost of upgrading existing infrastructure and software to handle
the dissemination of additional core data message traffic; the cost to
SROs to implement system changes required in order to make regulatory
data and other data needed to generate consolidated market data
available to competing consolidators; the cost of technological
investments market participants might have to make in order to receive
the new core message traffic; and the cost to users of proprietary data
whose information advantage will dissipate somewhat.\1154\
---------------------------------------------------------------------------
\1150\ See MDI Adopting Release, supra note 10, at 18753 for the
full discussion of the effect of including odd-lot information
inside the NBBO in its definition of core data.
\1151\ Id.
\1152\ Id.
\1153\ See MDI Adopting Release, supra note 10, at 18759 for the
full discussion of the costs associated with expanding core data to
include odd-lot information inside the NBBO.
\1154\ Id.
---------------------------------------------------------------------------
The MDI Rules do not require the competing consolidators to
disseminate odd-lot information. However, the Commission estimated that
at least one competing consolidator will disseminate the odd-lot
information because the Commission believed that there will be demand
for the data.\1155\
---------------------------------------------------------------------------
\1155\ See MDI Adopting Release, supra note 10, at 18752 n.1945
and surrounding text.
---------------------------------------------------------------------------
4. Affected Entities and Markets
The amendments will affect trading in NMS stocks, particularly
either on exchanges that charge high access fees or in stocks with
lower quoted spreads, many odd-lots inside the spread, or higher
prices. Therefore, the amendments will affect a wide variety of market
participants, including national securities exchanges, other trading
venues, exclusive SIPs and their data users, any future competing
consolidators, broker-dealers operating order entry and order routing
systems, and others who engage in the trading of NMS stocks, including
investors.\1156\
---------------------------------------------------------------------------
\1156\ According to the 2022 Survey of Consumer Finances,
available at https://www.federalreserve.gov/econres/scfindex.htm?mod=article_inline, out of a total number of households
of approximately 131,000,000, 58% invested in equities in some
fashion (e.g., held stock directly, invested in a stock mutual fund,
etc.). Bd Gov. Fed. Res., Changes in U.S. Family Finances from 2019
to 2022: Evidence from the Survey of Consumer Finances (Oct. 2023)
at 19, available at financeshttps://www.federalreserve.gov/publications/files/scf23.pdf.
---------------------------------------------------------------------------
[[Page 81701]]
There are 16 national securities exchanges on which NMS stocks are
traded that will be affected by the amendments. The exchanges compete
with each other and other trading venues to attract order flow.
Exchanges compete by setting the rules that dictate how orders routed
to them interact given the broader requirements of the Exchange Act and
rules thereunder. Such rules are coded into the systems of exchanges
that match buy and sell orders. Exchanges also compete via their
services and fee structures; they differentiate themselves with the
access fees they charge or the rebates they pay out for particular
order types, as well as their data and connectivity options.\1157\ A
subset of national securities exchanges, the five listing exchanges,
also compete to attract stock listings by setting rules for listing
standards for securities. The listing exchanges are also responsible
for tracking certain regulatory information regarding their listed
stocks.
---------------------------------------------------------------------------
\1157\ Exchanges can also facilitate the routing of orders to
other exchanges.
---------------------------------------------------------------------------
Other trading venues, including 33 ATSs and 238 other FINRA
members, including OTC market makers, also compete with exchanges and
each other to attract order flow in NMS stocks and can route orders to
the various trading venues. The order flow they attract depends on a
number of factors such as fees and price improvement over the NBBO,
services such as order display features, segmentation of subscriber
order flow and the ability of subscribers to select which category of
order flow to interact with, among other aspects of execution quality.
Pending the full implementation of the MDI Rules, the market for
market data is serviced by the two exclusive SIPs and exchange
proprietary feeds. The two exclusive SIPs collect trade, quote, and
regulatory data from the 16 exchanges and three trade reporting
facilities,\1158\ consolidate the data, determine an NBBO, and
disseminate those data directly to users or through vendors and broker-
dealers. The exclusive SIPs can also collect information from the
alternative display facility (``ADF'') operated by FINRA, though no one
currently uses the ADF to display quotes. Upon full implementation of
the MDI Rules, the exclusive SIPs will be retired, and an unknown
number of competing consolidators will take over the collection,
consolidation, estimation, and dissemination of these data.\1159\ The
volume of data to be processed through these competing consolidators
will be greater than that currently processed through exclusive SIPs,
but competing consolidators will have flexibility to design data
products tailored to different user types. In addition to the exclusive
SIPs, the exchanges also disseminate market data to paying subscribers
via proprietary data feeds. Some of these proprietary data feeds
provide more data than the exclusive SIPs and are provided at a lower
latency; however, the proprietary feeds are limited to individual
exchanges while the SIPs contain consolidated data across all exchanges
and also contain all off-exchange trades.\1160\ Following the
transition to a competing consolidator model for market data, the
Commission expects total fees for market data are likely to
decline.\1161\
---------------------------------------------------------------------------
\1158\ Trade Reporting Facilities (TRFs) are facilities through
which FINRA members report off-exchange transactions in NMS stocks,
as defined in SEC Rule 600(b)(47) of Regulation NMS.
\1159\ While the Commission is uncertain about the number of
competing consolidators that will enter the market when exclusive
the SIPs are retired, the Commission believes that the most likely
outcome is three or more competing consolidators with at least one
competing consolidator that is not affiliated with one of the
exchanges currently operating the exclusive SIPs or an exchange that
has sufficient proprietary data revenue that would create
conflicting profit incentives. See MDI Adopting Release, supra note
10, at 18768-72 for further discussion on the number of competing
consolidators that may enter the market.
\1160\ See supra note 862 and infra note 1780 and associated
text for a further discussion on the nature of proprietary data
feeds.
\1161\ See MDI Adopting Release, supra note 10, at 18772-78.
---------------------------------------------------------------------------
Broker-dealers typically route their own orders or their customers'
orders for execution to trading venues. There were 3,494 registered
broker-dealers as of Q2 2023.\1162\ A portion of these broker-dealers
focus their business on individual and/or institutional investors in
the market for NMS stocks. According to CAT data, as of the end of
2022, there were approximately 1,006 registered broker-dealers that
originated NMS stock orders on behalf of individual investors and
approximately 837 broker-dealers that originated NMS stocks orders on
behalf of institutional investors.\1163\ Institutional investor orders
are typically ``not held'' orders, which provides the broker-dealer
with more time and price discretion to execute the order or to minimize
price impact.\1164\ In contrast, broker-dealers must attempt to execute
a marketable held order immediately; these orders better suit retail
investors because retail orders typically have much lower price impact,
which reduces the need for discretion in order handling.\1165\ Brokers-
dealers serving individual investors often distinguish themselves by
the customer service and financial advice they provide and the
accessibility and functionality of their trading platforms.
---------------------------------------------------------------------------
\1162\ Based on information from broker-dealers' Q2 2023 FOCUS
Report Form X-17A-5 Schedule I. This includes both carrying broker-
dealers, who maintain custody of customer funds and securities, and
introducing broker-dealers, who accept customer orders and introduce
their customers to a carrying broker-dealer that will hold the
customers' securities and cash. In addition, the Commission
acknowledges that the total number of broker-dealers is likely to
increase as a result of the recent Dealer Adopting Release. The
Dealer Adopting Release adopted new rules to further define the
phrase ``as a part of a regular business'' as used in the statutory
definitions of ``dealer'' and ``government securities dealer.'' The
Dealer Adopting Release estimated that up to 43 entities may be
required to register with the Commission as a dealer or government
securities dealer, which would increase the total number of broker-
dealers affected by the amendments.
\1163\ Customer accounts are identified in CAT as accounts
belonging to either the ``Institutional Customer'' account type,
defined as accounts that meet the definition in FINRA Rule 4512(c),
or the ``Individual Customer'' account holder type, defined as
accounts that do not meet the definition of FINRA Rule 4512(c) and
are also not a proprietary account.
\1164\ See Securities Exchange Act Release No. 84528 (Nov. 2,
2018), 83 FR 58338 (Nov. 19, 2018) (adopting new order handling
disclosure requirements) at nn.59-60 and corresponding text.
\1165\ FINRA's best execution obligation requires that, ``A
member must make every effort to execute a marketable customer order
that it receives fully and promptly.'' See FINRA Rule 5310 (Best
Execution and Interpositioning), Supplementary Material para. .01,
available at https://www.finra.org/rules-guidance/rulebooks/finra-rules/5310 (accessed Jun. 18, 2024).
---------------------------------------------------------------------------
Many broker-dealers that handle customer accounts do not directly
access national securities exchanges or ATSs for their orders. They use
other broker-dealers to facilitate market access for them through those
broker-dealers' order entry systems. The Commission estimates that
there are 1,161 broker-dealers with order entry systems that originate
orders in NMS stocks in the minimum pricing increments; the amendments
to Rule 612 may require changes to these order entry systems.\1166\ Of
these broker-dealers, an estimated 270 broker-dealers operate smart
order routers to facilitate order routing.\1167\
---------------------------------------------------------------------------
\1166\ See infra note 1656.
\1167\ See infra note 1660.
---------------------------------------------------------------------------
5. Amendments to Rule 605
Several commenters requested the Commission consider interactions
between the economic effects of these proposed amendments and the
proposed amendments to Rule 605.\1168\ The amendments to Rule 605 were
not included as part of the baseline in the
[[Page 81702]]
Proposing Release because they were not adopted at that time. The
Commission amended Rule 605 on March 6, 2024,\1169\ and the
requirements of that rule are part of the baseline considered here.
With certain exceptions, the amendments to Rule 605 have a compliance
date of Dec. 14, 2025,\1170\ which is after the compliance dates of the
amendments made by this adopting release. The following discussion
reflects the Commission's assessment of the anticipated economic
effects of the amendments to Rule 605 described in the Rule 605
Amendments as they relate to the baseline for the adoption of these
amendments. Specific interactions between the expected economic effects
of the amendments to Rule 605 and those of rules adopted herein will be
discussed in detail in a later section.\1171\
---------------------------------------------------------------------------
\1168\ See, e.g., SIFMA Letter II; Virtu Letter II; Citadel
Letter I; Equity Market Structure Citadel Letter; Citadel Letter II.
See also Rule 605 Proposal, supra note 117.
\1169\ See Rule 605 Amendments, supra note 10.
\1170\ See supra note 1044. As an exception, after odd-lot order
information sufficient to calculate best available displayed price
is made available pursuant to an effective NMS plan, market centers,
brokers and dealers will have six months to begin including price
improvement statistics relative to best available displayed price in
their Rule 605 reports. See Rule 605 Amendments, supra note 10, at
26497.
\1171\ See infra section VII.E.6.a; see also supra section II.
---------------------------------------------------------------------------
Rule 605 requires disclosures for order executions in NMS
stocks.\1172\ The Rule 605 amendments modified reporting requirements
in several ways. First, the amendments expanded the scope of reporting
entities subject to the rule to include larger-broker-dealers \1173\ in
addition to market centers.\1174\ The amendments also enhanced the
accessibility of the reported execution quality statistics by requiring
all reporting entities to make a summary report available.\1175\
---------------------------------------------------------------------------
\1172\ 17 CFR 242.605.
\1173\ The term ``larger broker-dealer'' refers to a broker-
dealer that meets or exceeds the ``customer account threshold,'' as
defined in Rule 605(a)(7) as broker-dealers that carry or introduce
orders on behalf of 100,000 or more customer accounts through which
transactions are affected for the purchase sale of NMS stocks. See
Rule 605 Amendments, supra note 10, at 26428 n.61; 17 CFR
242.605(a)(7).
\1174\ Regulation NMS defines the term ``market center'' to mean
any exchange market maker, OTC market maker, ATS, national
securities exchange, or national securities association. See 17 CFR
242.600(b)(55).
\1175\ See Rule 605 Amendments, supra note 10, at 26428.
---------------------------------------------------------------------------
The Rule 605 Amendments also included amendments to the information
required to be reported under Rule 605, some of which are expected to
be relevant to the amendments to this Rule. First, the amendments to
Rule 605 added requirements related to the reporting of price
improvement statistics relative to the best available displayed price,
which incorporates information about the best priced odd-lot orders, in
addition to the preexisting requirement to report price improvement
statistics relative to the NBBO.\1176\ The Rule 605 Amendments
acknowledged that, while under the MDI Rules odd-lot information will
include pricing information about odd-lots priced better than the
NBBO,\1177\ the MDI Rules have been approved but not yet implemented,
and thus this information is not yet available. Therefore, the
Commission stated that Rule 605's price improvement statistics that are
relative to the best available displayed price will not be required to
be reported until six months after odd-lot order information needed to
calculate the best available displayed price is made available pursuant
to an effective national market system plan.\1178\
---------------------------------------------------------------------------
\1176\ See 17 CFR 242.600(b)(14) (defining the ``best available
displayed price'' as, with respect to an order to buy, the lower of:
the national best offer at the time of order receipt or the price of
the best odd-lot order to sell at the time of order receipt as
disseminated pursuant to an effective transaction reporting plan or
effective national market system plan; and, with respect to an order
to sell, the higher of: the national best bid at the time of order
receipt or the price of the best odd-lot order to buy at the time of
order receipt as disseminated pursuant to an effective transaction
reporting plan or effective national market system plan. With
respect to a midpoint-or-better limit order, the best available
displayed price shall be determined at the time such order becomes
executable rather than the time of order receipt) and 17 CFR
242.605(a)(1)(ii)(M) through (Q).
\1177\ See MDI Adopting Release, supra note 10, at 18753.
\1178\ In the Rule 605 Amendments, the Commission acknowledged
that it was still considering the proposed changes discussed in the
Proposing Release and adopted herein, including accelerating the
implementation of the round lot and odd-lot information definitions
contained in the MDI Release and amending the definition of odd-lot
information to include a new data element for the best available
odd-lot orders available in the market. In the Rule 605 Amendments
the Commission stated that, if it determined to adopt an amendment
to the definition of odd-lot information to include a data element
that identifies the best odd-lot orders available in the market,
reporting entities would be required to use such information to
determine the best available odd-lot price. See Rule 605 Amendments,
supra note 10, at 26428 n.719.
---------------------------------------------------------------------------
Second, the amendments to Rule 605 require the separate reporting
of non-marketable limit orders that are priced at the midpoint of the
NBBO or better (``midpoint-or-better NMLOs''), and additionally
requires the reporting of information about the price improvement
offered to these orders.\1179\ An analysis by the Commission in the
Rule 605 Amendments indicates that a high percentage of midpoint-or-
better NMLO share volume is submitted with IOC designations as compared
to other NMLOs, confirming that many of these orders are submitted by
traders with the intention of executing immediately against hidden or
odd-lot inside-the-quote liquidity, and that these orders tend to have
different execution characteristics than other types of NMLOs.\1180\
Therefore, the Commission stated that market participants will benefit
from an increase in transparency by the separate reporting of these
orders, along with the required reporting of certain execution quality
statistics that measure the cost of executing immediately, such as
effective spreads.\1181\
---------------------------------------------------------------------------
\1179\ See 17 CFR 242.600(b)(57) (defining ``midpoint-or-better
orders'') and 17 CFR 242.605(a)(1)(ii).
\1180\ See Rule 605 Amendments, supra note 10, at 26528.
\1181\ See Rule 605 Amendments, supra note 10, at 26556-26557,
26568.
---------------------------------------------------------------------------
Third, the amendments to Rule 605 require the reporting of
information regarding the extent to which orders received an execution
at prices at or better than the quote for share quantities greater than
the displayed size at the quote, i.e., ``size improvement.'' This
information includes (1) a benchmark metric that measures the displayed
size at the time of order receipt, which can then be compared to the
number of submitted shares to determine the extent to which a trading
venue handled orders that outsized available displayed depth,\1182\ and
(2) for orders that outsized available displayed depth, the number of
shares that received size improvement.\1183\
---------------------------------------------------------------------------
\1182\ See 17 CFR 242.600(b)(72) (defining the ``order size
benchmark'') and 17 CFR 242.605(a)(1)(ii)(R).
\1183\ See 17 CFR 242.605(a)(1)(ii)(S), requiring the reporting
of ``the sum of, for each execution of a covered order, the greater
of: the total number of shares executed with price improvement plus
the total number of shares executed at the quote minus the order
size benchmark, or zero.'' The ``total number of shares executed
with price improvement plus the total number of shares executed at
the quote minus the order size benchmark'' (``net size
improvement'') will only be a strictly positive number for those
orders that are both eligible to receive size improvement and
actually receive size improvement, and thus is equivalent to a
measure of shares that are eligible to and that received size
improvement. See Rule 605 Amendments, supra note 10, at 26428
n.1544.
---------------------------------------------------------------------------
The amendments to Rule 605 also modified the definition of order
size categories from order size categories based on numbers of shares,
with orders less than 100 shares excluded, to order size categories
based on a notional dollar value range, along with an indication that
the category reflects orders that were for an odd-lot, a round lot, or
less than a share.\1184\ The Commission stated in the Rule 605
Amendments that one of the benefits of
[[Page 81703]]
this change is to ensure that round lots for stocks with prices greater
than $250 are not excluded from Rule 605 reports following the change
in round lot definition under the MDI Rules.\1185\
---------------------------------------------------------------------------
\1184\ See 17 CFR 242.600(b)(18).
\1185\ See Rule 605 Amendments, supra note 10, at 26523.
---------------------------------------------------------------------------
In the Rule 605 Amendments, the Commission stated that the
amendments to Rule 605 will promote increased transparency of order
execution quality, particularly for larger broker-dealers who were not
required to disclose execution quality information under preexisting
Rule 605, but also for market centers, whose execution quality
information will be more relevant and easier to access because of
improvements to existing Rule 605 disclosure requirements.\1186\ The
Commission stated in the Rule 605 Amendments that this increase in
transparency is expected to increase the extent to which market centers
and broker-dealers compete on the basis of execution quality, as well
as improvements in execution quality.\1187\ The Commission also stated
that the amendments to Rule 605 will result in initial and ongoing
compliance costs, the majority of which will be related to expanding
the scope of reporting entities to include larger broker-dealers, but a
significant portion of which will result from the need for market
centers to update their systems to process and store the data necessary
to prepare the amended reports.\1188\
---------------------------------------------------------------------------
\1186\ Id. at 26543.
\1187\ See Rule 605 Amendments, supra note 10, at 26543-26544.
\1188\ Id. at 26579-26580.
---------------------------------------------------------------------------
D. Benefits, Costs, and Other Economic Effects
The Commission expects the adopted minimum quoting increment will
alleviate tick constraints and better allow prices to be determined by
the forces of supply and demand, lowering transaction costs for
investors. A lower access fee cap will further reduce the transaction
costs of liquidity demanders in the predominant maker-taker structure.
Making fees and rebates determinable at the time of trade may enhance
broker-dealer order routing by helping mitigate a potential conflict of
interest and providing clarity in terms of all in execution costs.
Accelerating the inclusion of odd-lot information into the exclusive
SIPs, accelerating the implementation of the round lot definitions, and
amending the definition of odd-lot information to include the best odd-
lot order, will accelerate some of the benefits of the MDI Rules, and
could also lead to better order execution by enhancing benchmarking.
The amendments will also impose compliance costs on various market
participants.
The Commission continually monitors the national market system and
the operation of Federal securities laws. As discussed above, the
national market system continually changes and the Commission,
consistent with its oversight of the national market system, will
monitor the impact of the adopted rules. With regard to the amendments
adopted herein, by May 2029 (three years from the last implementation
date), Commission staff will review and study the effects of the
amendments in the national market system. Such a review and study might
include, but would not be limited to, an investigation of: (i) general
market quality and trading activity in reaction to the implementation
of the variable tick size, (ii) the reaction of quoted spreads to the
implementation of the amended access fee cap, and (iii) changes to
where market participants direct order flow, e.g., to exchange versus
off-exchange venues, following the implementation of the amendments.
In studying the effect on market quality, a number of different
metrics could be examined including quoted, realized, and effective
spreads; cumulative depth from the midpoint across multiple price
levels; and the cost of a round-trip trade for various trade
sizes.\1189\ In such analysis, improvements in market quality for
stocks affected by Rule 612 would correspond to reduced spreads
(adjusting for fees or rebates) or a reduced cost of a round-trip
trade.\1190\ To isolate the effect of Rule 610, the analysis might
focus on those stocks not directly affected by Rule 612. Such analysis
might focus on the effect of Rule 610 on quoted spreads (e.g., to
examine how the quoted spread adjusts in response to changes in fees
and rebates), on whether Rule 610 leads to any change in effective
spread off-exchange (due to adjustments to on-exchange quotes), and on
any migration of liquidity off-exchange.
---------------------------------------------------------------------------
\1189\ Compare table 8.
\1190\ One possible study design could focus on stocks close to
the TWAQS threshold. Comparing stocks with similar levels of
liquidity ex ante would better isolate the effect of the smaller
tick size on market quality.
---------------------------------------------------------------------------
1. Modification of Rule 612 To Create a Half-Penny Tick
The Commission is adopting amendments to Rule 612 that introduce
one minimum pricing increment that is less than $0.01, i.e., $0.005,
for quotes and orders priced $1.00 or more for NMS stocks that have a
TWAQS of $0.015 or less during the evaluation period.\1191\ Hence, the
amendments to Rule 612 will create a smaller tick size for some NMS
stocks.
---------------------------------------------------------------------------
\1191\ See infra section III.C
---------------------------------------------------------------------------
The Commission expects that, on average, market quality will
improve for the stocks receiving the smaller tick size. A smaller tick
has two competing effects on market quality. First, a smaller tick
leads to pricing that more effectively balances liquidity supply and
demand, limiting distortions, and thus lowering transaction costs.
Second, a smaller tick fragments liquidity in the order book into more
price levels, which can increase complexity associated with
implementing trades, and increases the incidence of pennying \1192\--
effects that can harm liquidity. A smaller tick can also increase
message traffic which can be costly for market participants. The
amendments will not change the tick for NMS stocks priced below $1.00,
nor for stocks with time weighted average quoted spread always greater
than $0.015 during an Evaluation Period and thus the tick size
amendments are expected to have minimal if any effect on the trading
environment for these stocks.
---------------------------------------------------------------------------
\1192\ See supra section I.A.1 and note 994 for the definition
and discussion of pennying.
---------------------------------------------------------------------------
a. Estimates of Percent of Trading Volume and Number of NMS Stocks
Affected
As discussed in section VII.C.1, prior to these amendments, the
tick size for orders in NMS stocks priced equal to or greater than
$1.00 was $0.01, and the tick size for orders in NMS stocks priced less
than $1.00 was and remains $0.0001.\1193\ The amendments assign each
NMS stock to one of two tick sizes: $0.005 or $0.01, depending on the
stock's time weighted average quoted spread during an Evaluation Period
(specifically, assigning $0.005 for stocks with a TWAQS of $0.015 or
less).\1194\ Table 7 presents estimates of the amount of share and
dollar trading volume that would have been associated with the two tick
sizes, as well as the sub $1.00 tick size, based on 2023 trading
volumes. It also presents estimates based on the Proposal which would
have reduced tick sizes for stocks with TWAQS of $0.040 or less.
---------------------------------------------------------------------------
\1193\ See supra section VII.C.1.a.
\1194\ See supra section III.C for further discussion.
[[Page 81704]]
Table 7--Estimated Number of Stocks and Trading Volume in Each Tick Size Group \a\
----------------------------------------------------------------------------------------------------------------
Number of Estimated % Estimated %
Average quoted spread Tick stocks share volume dollar volume
----------------------------------------------------------------------------------------------------------------
All Stocks
----------------------------------------------------------------------------------------------------------------
Spread <= $0.015...................... $0.005.................. 1,788 66.2 42.9
$0.015 < Spread....................... $0.01................... 9,047 33.8 57.1
Spread <= $0.04....................... (Proposed Reduction to 4,333 84.8 66.5
$0.005 or smaller).
----------------------------------------------------------------------------------------------------------------
Price < $1
----------------------------------------------------------------------------------------------------------------
$0.0001................. 1,106 12.3 0.1
----------------------------------------------------------------------------------------------------------------
\a\ In this table, quoted spreads, and thus tick sizes, are determined by computing the time weighted quoted
spread during regular trading hours as computed by the WRDS intra-day indicators for every sym_root and
sym_suffix combination in the WRDS intra-day indicators dataset and taking the equal weighted average across
all trading days in January-March 2023. Stocks with average quoted spreads less than $0.015 are assigned a
$0.005 tick. All other stocks are assigned a $0.01 tick. A stock with a price less than $1.00 will still be
assigned a tick size per the usual process, which would be in force should the stock's price rise above $1.00.
As long as the stock's price remains below $1.00 the $0.0001 tick size would prevail. The designated tick size
is applied to trading volume in May-October 2023 where share and dollar volume is obtained from the universe
of stocks in WRDS intra-day indicators. New stocks are given a tick size of $0.01. The number of stocks
assigned to each group is indicated in the Number of Stocks column and indicates the average number of stocks
in each category (listings and de-listings can affect the daily number of stocks trading as well as if a
stock's price falls below $1). If a stock has a VWAP of less than $1.00, then that stock, as well as all of
its trading volume for that day, is assigned to the $0.0001 tick size.
This estimate may be an upper bound. As discussed in section VII.B.3, supra and infra section VII.D.2, rebates
can lower the quoted spread (although not necessarily transaction costs). Thus, lowering the access fee, and
thus the associated rebates, may lead to wider quoted spreads. Because of this, some stocks may have quoted
spreads that meet the threshold for the smaller tick size in the current environment but may not meet that
threshold once the access fee cap is reduced, leading to lower rebates offered. Additionally, all stocks, even
those priced below $1.00, will be assigned a tick size via the usual process. If a stock price falls below
$1.00 the applicable tick size will be $0.0001. So not all stocks initially assigned the $0.005 tick size will
trade differently than the baseline. This table differs from table 3 because table 3 is based on daily average
TWAQs and does not attempt to analyze the effect of the adopted amendments.
Once implemented, the changes to the current arrangements for consolidated market data pursuant to the MDI Rules
may impact the number of stocks and their estimated percentage volumes anticipated for each tick level. In
particular, under the MDI Rules, NMS stocks priced $250 or more will receive reductions in round lot sizes
which is anticipated to lower their quoted spreads; however, the effect on the reported numbers is likely
small both because these stocks make up less than 4% of share volume and because they are unlikely to have
quoted spreads less than $0.015. Based on an analysis of data from May-October 2023, the average quoted spread
of a stock priced between $250 and $1,000 was $0.71, far greater from the $0.015 that will trigger a smaller
minimum increment. Similarly, for stocks priced between $1,000 and $10,000 the average quoted spread was $3.85
and the only stock that had a value weighted average price greater than $10,000 already has a round lot size
of one share and had an average quoted spread of $0.07.
Table 7 indicates that, had the amendments been in place in 2023,
approximately 66% of share volume and 43% of dollar volume, associated
with an estimated 1,788 individual stocks, would likely have been
assigned the $0.005 tick size. The adopted Rules represent a
significant reduction in the scope of the Rule compared to the
proposal. Table 7 provides estimates of the number of stocks and volume
that would have been affected if the Commission had implemented the
Rule with the tick size thresholds as proposed (the proposal would have
lowered the tick size for all stocks with TWAQS less than $0.04). The
Commission estimates that there would have been 4,333 stocks receiving
a smaller tick accounting for 84.8% (66.5%) of share (dollar) volume if
all stocks with a TWAQS less than or equal to $0.04 received a smaller
tick size. Consequently, the number of stocks receiving a lower tick
size is more than halved under the adopted amendments.
b. Effects on Market Quality
For the stocks that will receive the $0.005 tick, the Commission
expects market quality to improve. Smaller tick sizes present a market
quality tradeoff between increasing pennying and complexity concerns--
which can harm market quality--and reducing pricing constraints--which
can improve market quality by reducing pricing distortions leading to
an oversupply of liquidity relative to competitive levels. The
Commission believes that market quality will, on average, improve for
stocks receiving the smaller tick based on theoretical discussion, the
Commission's empirical analysis, as well as evidence and opinions
expressed by commenters. For example, one commenter agreed with the
presence of market distortions under current tick sizes, stating:
``[t]he SEC correctly describes the problem of tick-constrained
securities. Such securities are `not able to be priced by market
forces' because the current `rule 612 minimum pricing increment of
$0.01 may now be too large for certain stocks, which, in turn, results
in the pricing of such stocks being artificially constrained.' Trading
in these securities would be improved `if competitive market forces
could establish prices in sub-penny increments, which could reduce
quoted spreads,' allowing these securities to `be priced more
aggressively within the spread.' '' \1195\
---------------------------------------------------------------------------
\1195\ See Nasdaq Letter I at 11 (quoting the Proposing
Release).
---------------------------------------------------------------------------
The theoretical discussion provided below supports characterizing a
smaller tick size as providing a pennying/complexity versus pricing
constraint tradeoff, and the empirical analysis presented in table 8 as
well as other empirical research suggests that, for stocks with fewer
than approximately two ticks intra-spread,\1196\ a reduction in the
tick size on average improves market quality. A number of commenters
agreed, and some commenters presented analyses suggesting that 2 to 4
ticks intra-spread may be optimal. Combined, this evidence suggests
that the tick size reduction associated with these amendments will, on
average, improve
[[Page 81705]]
market quality for the subset of stocks receiving the lower tick
size.\1197\
---------------------------------------------------------------------------
\1196\ We use the terminology ``ticks intra-spread'' or ``ticks
within the spread'' to mean the number of quoting increments between
the NBB and NBO (the quoted spread). For example, if the quoted
spread is one penny wide (in a stock priced above $1), then we say
that there is one tick intra-spread under the baseline. Under the
baseline, symbols priced above $1.00 with a quoted spread between 2
and 4 pennies would have 2 to 4 ticks intra-spread.
\1197\ The amendments will take stocks trading with 1-1.5 ticks
intra-spread and increase the number of ticks intra-spread to up to
3.
---------------------------------------------------------------------------
i. Theoretical Discussion
Tick sizes present an economic tradeoff.\1198\ All else equal,
reducing the tick size improves market quality by reducing distortions
associated with markets not being able to set prices that equate
liquidity supply and demand in the presence of a discrete pricing
grid.\1199\ In a competitive market, and in the absence of rebates or
other price distortions, the prevailing bid or ask price will be the
feasible price equal to or just worse than the price that equates
supply and demand for the underlying asset.\1200\ This is because
liquidity providers will not post bids and offers that would result in
guaranteed trading losses--i.e., they will not post prices that do not
bring in sufficient revenue to cover their marginal cost of providing
liquidity.\1201\ Since there is competition along a finite pricing
grid, they choose the closest feasible price just worse than the
competitive one. The gap between the feasible price and the price that
equates liquidity supply and demand--i.e., the competitive price--is a
price distortion allowing liquidity providers to earn rents on
liquidity provision.
---------------------------------------------------------------------------
\1198\ See section VII.B.2
\1199\ See, e.g., Rindi and Graziani Letter at 2 (agreeing), see
also Barardehi et al., supra note 231 (for a more thorough
discussion of this tradeoff). See also NASAA Letter at 9 (stating
that the general concept that a narrower tick size will increase
pricing efficiency), as well as discussion in Ingrid M. Werner, et
al., Tick Size, Trading Strategies and Market Quality, 69 Mgmt. Sci.
3818 (2023). See also Budish Letter at 4 referring to a tick size
that is too wide as producing rents via regulatory price
constraints.
\1200\ Any price better than this will lead to an excess of
liquidity demand which will push prices out again.
\1201\ Marginal cost in this context refers to the cost of
providing an additional share of liquidity. If the revenue
associated with providing a share of liquidity is less than the cost
of providing that share, then liquidity providers are better off not
providing liquidity than incurring a loss to provide liquidity.
---------------------------------------------------------------------------
This pricing distortion is most relevant for stocks that are tick-
constrained and diminishes as quoted spreads widen. To understand this,
consider again the example of section VII.B.2. In that example, under a
tick size of $0.005, the ask would be $10.015 and the bid $10.005.
However, with a tick size of $0.01, the ask would be $10.02 and the bid
$10.00, implying a spread that is twice as wide. Now assume that the
same issuer reduced the number of shares so that the stock increases in
price 100-fold, but the underlying economics are the same. To achieve
the same reduction in spread would not require any change to the tick
size: an ask of $1,001.50 and a bid of $1,000.50 are feasible even with
a tick size of one penny.
While a smaller tick size increases competition, thereby reducing
distortions and reducing transaction costs, there are potential costs
raised in the proposing release and also by commenters which are
discussed below.
Pennying: The proposing release and commenters identified pennying
as a risk of a smaller tick which can harm market quality.\1202\
Pennying occurs when limit order providers get to the front of the
queue by providing economically trivial price improvement. It reduces
the importance of time priority.\1203\ The risk of being pennied could
discourage liquidity provision in lit markets, particularly by market
participants that are slower to respond to changes in market conditions
and could increase transaction costs for these investors.\1204\ To
compensate for additional costs associated with a fragmented order
book, liquidity providers may post less aggressive quotes leading to
wider quoted spreads and worse market quality.\1205\
---------------------------------------------------------------------------
\1202\ See, e.g., Robinhood Letter at 41, Virtu Letter II at 4,
Tastytrade Letter at 20, AIMA Letter at 2, Brandes Letter at 2, UBS
Letter at 10, and TradeStation Letter at 5, Lewis Letter attached to
Virtu Letter II at p 33. See also supra note 994 and section VII.B.2
for a definition and discussion of pennying. See also Proposing
Release, supra note 11 at section V.D.1.
\1203\ See, e.g., Antitrust Division of the DOJ Letter at 5 and
XTX Markets Letter at 3.
\1204\ See Dyhrberg et al., supra note 994 studying the effects
of imposing a tick size on a crypto exchange that previously did not
have a tick size. The authors report an improvement in market
quality due largely to a reduction in pennying behavior. See also
Virtu Letter II at 25 and Better Markets Letter I at 8. See also
Budish Letter at 5.
\1205\ See, e.g., Virtu Letter II at 8, Fidelity Letter at 10.
---------------------------------------------------------------------------
Market participants may respond to an increased risk of pennying by
increasing their use of hidden or off-exchange orders that do not
display prices, and thus avoid exposing the price needed to beat in
order to get to the front of the queue and increase the likelihood of a
fill.\1206\ Increased use of hidden orders has been associated with
worse market quality outcomes.\1207\ Some commenters expressed their
belief that a narrower tick and increased pennying could lead some
orders that previously were at protected prices to be traded
through.\1208\ However, it is not clear from the commenters' letters
why this would occur given the order protection rule and broker's best
execution responsibilities. One commenter also suggested that narrow
ticks could increase volatility.\1209\ However, existing research on
the topic would suggest, if anything, an opposite effect.\1210\
---------------------------------------------------------------------------
\1206\ See, e.g., IEX Letter I at 12, Danny Mulson Letter at 1,
Nasdaq Letter I at 2.
\1207\ See, e.g., Amy K. Edwards, et al., The Effect of Hidden
Liquidity: Evidence from an Exogenous Shock (working paper Mar. 1,
2021), available at https://ssrn.com/abstract=3766512 (retrieved
from SSRN Elsevier database) (``Edwards, et al. (2021)''). See also
Danny Mulson Letter at 3 stating that a smaller tick size would lead
to more hidden orders, specifically `peg offset dark orders' which
could harm price efficiency.
\1208\ See Themis Letter at 5,Virtu Letter II at 6, 10
discussing how a smaller tick can weaken protected quotes.
\1209\ See, e.g., Virtu Letter II at 8. See, also Edwin Hu et
al., 2018; supra note 1002; and Kee H. Chung et al., Tick Size
Liquidity for Small and Large Orders and Price Informativeness:
Evidence From the Tick Size Pilot Program, 136 J. Fin. Econ. 879
(2020), who both report the opposite effect in the context of the
Tick Size Pilot where stocks with wider ticks experienced more
volatility.
\1210\ See id. see also e.g., Edwards, et al., (2021), supra
note 1207.
---------------------------------------------------------------------------
In contrast to the tick size pricing distortion discussed above,
which is most relevant for stocks that are tick-constrained,\1211\ the
pennying effect will be most pronounced for stocks with wide quoted
spreads because there are more intra-spread price levels and the cost
of gaining priority over other liquidity providers, by updating the
best price by a single tick, is lower with a smaller tick.\1212\ For
example, a stock with a quoted spread of ten cents, and a $0.01 tick,
will have 10 price levels within the quoted spread, whereas a stock
with a $1.00 quoted spread and a $0.01 tick will have 100. Because
price has first priority in order execution, in a price-time priority
system where quote priority is awarded based on best price first and
then arrival order second, a primary way to gain priority for a trader
providing liquidity is to price-improve over existing orders. Without a
small tick size relative to the quoted spread, getting to the front of
the queue via price improvement will be more costly, requiring larger
relative price concessions.\1213\ Because the (beneficial) pricing
efficiency effect is greatest when quoted spreads are narrow, whereas
the (detrimental) pennying effect is greatest when quoted spreads are
wide, this
[[Page 81706]]
analysis suggests setting a minimum quoting increment on the basis of
average spread. Commenters agreed.\1214\
---------------------------------------------------------------------------
\1211\ See supra this section.
\1212\ The pennying effect would be particularly acute for wide-
quoted spread stocks with lower stock prices because a lower stock
price reduces the amount of capital needed to supply a round-lot
quote and hence make pennying less capital intensive.
\1213\ For example, if a stock has a quoted spread of ten cents
and a $0.01 tick, gaining priority through price improvement would
require narrowing the half- quoted spread (i.e., the distance
between the current quote and the midpoint) by 20%. If instead a
stock has a quoted spread of $1.00 with a $0.01 tick, a market
participant would only need to improve the half-quoted spread by 2%
to get to the front of the queue.
\1214\ See Budish Letter at 4 and Harris Letter at 7 supporting
the use of quoted spread as the determinate of the tick size. See
also infra section VII.D.1.b.iii.
---------------------------------------------------------------------------
Fragmenting liquidity: The proposing release and commenters also
discussed a cost of a lower tick size as spreading the displayed orders
over more price levels.\1215\ When tick increments are farther apart,
all else equal, liquidity providers that may have various prices at
which they are willing to provide liquidity must congregate their
quotes at only the available quoting increments. Thus, there will be
more depth at each level including at the NBBO. With more price levels
due to a smaller tick size, market participants can more accurately
tailor their quotes to the prices at which they are willing to provide
liquidity and thus liquidity will naturally spread over more levels and
there will be fewer resting orders at each price level, including the
NBBO.\1216\
---------------------------------------------------------------------------
\1215\ See Proposing Release, supra note 11 at section V.D.1 for
a discussion of fragmenting liquidity. See also e.g., GTS Letter at
5 and CCMR Letter at 24.
\1216\ See, e.g., GTS Letter at 5, CCMR Letter at 24, and UBS
Letter at 11.
---------------------------------------------------------------------------
Fragmenting liquidity across multiple price levels may decrease
costs associated with smaller orders, which would be able to source
liquidity at improved prices due to a finer price grid.\1217\ However,
it can increase the complexity and cost associated with sourcing
liquidity for larger orders,\1218\ as the reduction in shares available
at the top of the book will render it more likely that a market
participant must source liquidity beyond the NBBO in order to execute a
trade.\1219\ It could also increase the number of child orders a parent
order needs to be divided into to execute, which could increase the
overall complexity and likelihood of information leakage leading to
increased transaction costs via increased price impact.\1220\
---------------------------------------------------------------------------
\1217\ See UBS Letter at 11.
\1218\ Id. See also TradeStation Letter at 6 mentioning as an
example of increased complexity that brokers would have to put
systems in place to manage customers' good-till-canceled trades that
may remain open over a weekend when a tick size change is
implemented. See also discussion in Lewis Letter attached to Virtu
Letter II at 34-35.
\1219\ See, e.g., IEX Letter I at 13 discussing how order
shredding with smaller ticks can increase information leakage, such
as when quotes on other exchanges are cancelled when limit orders on
one exchange begin to be executed, potentially signaling a large
price moving trade. Brandes Letter at 2 states that increased
complexity associated with more pricing increments would be to the
detriment of longer-term investors. Equity Market Structure Citadel
Letter at 2 states that for institutional investors, ``[l]arger
orders will be more complex to execute, as filling the entire order
will require accessing multiple price levels, which can increase
price impact.''
\1220\ See, e.g., Citadel Letter I at 5, 9 and Virtu Letter II
at 8, 10 discussing the price impact of large trades under a regime
of smaller ticks; stating that smaller ticks could increase price
impact.
---------------------------------------------------------------------------
Quote Instability: Commenters also stated that less depth at the
NBBO can lead to increased NBBO quote instability as trades are more
likely to deplete depth at the NBBO prices.\1221\ One commenter
presented evidence that lower quote stability is empirically associated
with increased market making costs, which it states may deter liquidity
provision.\1222\ While increased quote instability may occur in stocks
receiving the lower tick, the lower tick itself will allow market
forces to adjust the price of liquidity--i.e., the quoted spread--such
that market makers are competitively compensated for the risks
associated with providing liquidity. Increased instability in the NBBO
could make it more difficult to determine which exchange has the best
price at a given point in time and thus where to route an order.\1223\
This could be particularly true when markets are volatile.\1224\
---------------------------------------------------------------------------
\1221\ Some commenters stated that smaller ticks would lead to
more ``flickering quotes,'' which are defined in the Reg NMS release
as quotes that flashed for a short period of time solely to earn
market data revenues, but were not truly accessible and therefore
did not add any value to the consolidated quote stream. However, the
Commission believes that this is unlikely for reasons discussed in
the Proposing Release note 195 and surrounding text relating to
advances in exchange technology. Other commenters defined
`flickering quotes' more broadly simply as periods of time where the
NBBO changes rapidly, see, e.g., IEX Letter I at 8 and Robinhood
Letter at 20. Much of the concern these commenters expressed was
with respect to the proposed $0.001 and $0.002 tick sizes which are
not part of the adopted amendments, see IEX Letter I at 8 and
Robinhood Letter at 20. As discussed here, the Commission
acknowledges that a smaller tick will likely lead to more frequent
changes to the NBBO and discusses those consequences herein.
\1222\ See IEX Letter I at 11-12.
\1223\ See, e.g., Themis Letter at 6 and CCMR Letter at 17 and
MFA Letter at 1. Quote instability could increase the complexity
associated with complying with Rule 611 as it could make it harder
to determine which exchange currently has the best price.
\1224\ See Citadel Letter I at 2, 7.
---------------------------------------------------------------------------
Increasing (or decreasing) rents to speed: The Proposing Release
stated that ``too small ticks may inefficiently award speed''.\1225\ As
discussed in the next few paragraphs, commenters also commented on the
effects of tick size on speed. Investments in speed are a fixed and
largely irreversible cost that some market participants choose to
incur. Changing the tick size could change the profitability of such
investments, that is, they could increase or decrease the rents to
speed. As noted in section VIII.B.2, a narrower tick reduces rents that
accrue when a liquidity provider can be first in line in a queue.\1226\
That is, narrowing the tick would be expected to reduce rents to speed.
However, speed confers an advantage in implementing a pennying
strategy: a trader can not only step ahead of another trader, but also
potentially sell (or buy) an asset back to the other trader if the
market moves unfavorably, replicating an option-like payoff.\1227\ The
amendments are limited to stocks with spreads for which pennying is
unlikely to be a dominant effect. Nonetheless, to the extent that
pennying increases, it has the potential to increase the rents to
speed.
---------------------------------------------------------------------------
\1225\ See Proposing Release, supra note 11, at 80306.
\1226\ See, e.g., Budish Letter at 1 (``Reducing the tick-size
constraint for tick-constrained stocks will reduce excess rents from
artificially constrained prices. These excess rents lead to a speed
race to the top of the book, which increases complexity, and the
rents come at the expense of investors via a higher cost of
liquidity.''), Antitrust Division of the DOJ Letter at 5, and XTX
Markets Letter at 3.
\1227\ See supra note 993 for a discussion on the relationship
between pennying and trading speed. See supra note 1202 and
accompanying text for discussions regarding the amendments to Rule
612 and pennying.
---------------------------------------------------------------------------
In the context of the proposal, one commenter stated that a smaller
tick size would be expected to increase the frequency of sniping
because smaller ticks generate faster and more frequent price
changes.\1228\ While there will be more prices at which to trade, the
underlying information is not changing (prices may change more rapidly,
but the information of each price change is smaller). That is, sniping
may become more frequent, but the profits per each individual snipe
attempt would decline. However, and as stated above, the Commission
does agree that a large number of ticks within the spread can make
pennying more prevalent, and to the extent that profits are linked to
speed, can increase the rents to speed. The adopted amendments imply
fewer ticks intra-spread than the proposing amendments, reducing this
effect. Thus, the Commission does not expect slower traders to be
disadvantaged by the adopted amendments.
---------------------------------------------------------------------------
\1228\ See Virtu Letter II at 23. Sniping pertains to the
ability to ``pick off'', by executing against a stale quote in
response to new information before it can be updated.
---------------------------------------------------------------------------
Effect on thinly traded securities: One commenter stated that
narrower quoted spreads due to a smaller tick would be harmful for
liquidity, particularly for smaller and medium-sized companies and for
thinly-traded securities.\1229\ This is because narrower quoted spreads
would discourage some liquidity providers from entering the market. The
Commission disagrees with this
[[Page 81707]]
characterization. The academic research on the Tick Size Pilot (TSP),
which increased the tick size for some smaller stocks from $0.01 to
$0.05 between 2016 and 2018, suggests that for many stocks affected by
the TSP, particularly those with narrower quoted spreads, the TSP led
to worse market quality.\1230\
---------------------------------------------------------------------------
\1229\ See STA Letter at 5.
\1230\ See infra section VII.D.1.b.ii and Barardehi et al.,
supra note 231 for additional discussion of the tick size
literature.
---------------------------------------------------------------------------
Additionally, lowering excess rents and the oversupply of liquidity
caused by tick size induced pricing distortions is likely to reduce
aggregate depth across all price levels. However, this reduction is
unlikely to be harmful to overall market quality, even for smaller or
thinly traded securities, as it would relieve a distortion resulting in
an oversupply of liquidity. As the amount of liquidity provision comes
closer to equilibrium levels, quoted spreads narrow and queue lengths
shorten, lowering transaction costs and increasing the likelihood that
relatively slower fundamental and/or retail traders could interact with
each other. This will reduce total transaction costs for these traders
because one side would be earning the quoted spread on the
transaction.\1231\
---------------------------------------------------------------------------
\1231\ See Retirement Coalition Letter at 2, Pragma Letter at
10.
---------------------------------------------------------------------------
ii. Empirical Analysis
This section presents the Commission's empirical analysis, as well
as a discussion of commenter analysis and views concerning the effect
of a tick size reduction on various aspects of market quality. Based on
these analyses, the Commission concludes that on average, stocks that
receive the smaller $0.005 tick size will experience improved market
quality--implying that, for these affected stocks, the predominant
market quality effect of the smaller tick size will be an increase in
pricing efficiency.\1232\
---------------------------------------------------------------------------
\1232\ See supra note 1199 and surrounding text for additional
discussion of the tick size tradeoff.
---------------------------------------------------------------------------
The academic literature examining the effect of tick sizes on
financial markets largely studies two events: decimalization, which
occurred in 2001 \1233\ and reduced the tick from \1/16\th of a dollar
($0.0625) to $0.01; and the TSP, which ran from October 2016 to October
2018 and temporarily increased the minimum tick increment from $0.01 to
$0.05 for a sample of small cap stocks.\1234\ Most of the literature
surrounding decimalization suggests that, on average, decimalization
was associated with a decline in quoted spreads consistent with the
notion that lowering the tick size relieved distortions related to
having a tick size that is too wide.\1235\
---------------------------------------------------------------------------
\1233\ See, e.g., Order Directing the Exchanges and the National
Association of Securities Dealers, Inc. to Submit a Phase-in Plan to
Implement Decimal Pricing in Equity Securities and Options; Pursuant
to Section 11A(a)(3)(B) of the Securities Exchange Act of 1934,
Securities Exchange Act Release No. 42914 (June 8, 2000), 65 FR
38010 (June 19, 2000); Commission Notice: Decimals Implementation
Plan for the Equities and Options Markets, SEC (July 24, 2000),
available at https://www.sec.gov/rules/other/decimalp.htm.
\1234\ See Edwin Hu, et al. (2018), supra note 1002, for
additional details about the Tick Size Pilot.
\1235\ See Hendrick Bessembinder, Trade Execution Costs and
Market Quality After Decimalization, 38. J. Fin. & Quantitative
Analysis 747 (2003). See also Michael A. Goldstein & Kenneth A.
Kavajecz, Eighths, Sixteenths and Market Depth: Changes in Tick Size
and Liquidity Provision on the NYSE, 56 J. Fin. Econ. 125 (2000) and
Charles M. Jones & Marc L. Lipson, Sixteenths: Direct Evidence on
Institutional Execution Costs, 59 J. Fin. Econ. 253 (2001), both
examining the earlier tick size change from \1/8\ to \1/16\ of a
dollar. See also Sugato Chakravarty, Venkatesh Panchapagesan &
Robert A. Wood, Did Decimalization Hurt Institutional Investors?, 8
J. Fin. Mkts. 400 (Nov. 2005) and Sugato Chakravarty, Bonnie F. Van
Ness, & Robert A. Van Ness, The Effect of Decimalization on Trade
Size and Adverse Selection Costs, 32 J. Bus. Fin. & Acc. 1063 (June/
July 2005), both suggesting that large institutional trades may have
become more costly following decimalization.
---------------------------------------------------------------------------
In the Proposing Release, the Commission supplemented existing
research with its own analysis on the TSP.\1236\ As stated in the
Proposing Release, market dynamics have changed dramatically in the
more than two decades since decimalization. Most notably over that
period, electronic, algorithmic, and high-frequency trading have come
to dominate the trading landscape, whereas they were much less
prominent in 2001. These changes diminish the relevance of evidence
from these prior periods, making it desirable to supplement existing
studies with evidence that is closer in time.
---------------------------------------------------------------------------
\1236\ See Proposing Release, supra note 11, at 80318-80322.
---------------------------------------------------------------------------
Some commenters questioned using the TSP to estimate the effects of
a reduced minimum pricing impact because the TSP affected only a subset
of small cap stocks, did not contain ETPs, and did not affect access
fee caps.\1238\ One of those commenters suggested that the TSP analysis
was not applicable because it focused on stocks with quoted spreads
much wider than the few cent quoted spreads contemplated by the
amendments.\1239\ The same commenter suggested that the TSP was not
applicable because it applied to a 5 to 1 tick size change, which is
different from the tick size change in the amendments.\1240\ Some
commenters went further and questioned whether anything could be
learned from the TSP because it did not involve sub-penny tick
sizes.\1241\
---------------------------------------------------------------------------
\1238\ See, e.g., CCMR Letter at 27, Virtu Letter II at 64.
Lewis Letter attached to Virtu Letter II at 34.
\1239\ See CCMR Letter at 27.
\1240\ Id.
\1241\ See Citadel Letter I at 12 (stating that the TSP
``provides no information on what would be expected to occur if
minimum quoting increments were further reduced to levels that have
never before been tested''); and Virtu Letter II at 3 (``The TSP
studied the impact of a widened minimum quoting and trading
increment for certain small capitalization stocks, and offered no
analysis, data, or conclusions on the potential impact that a
narrowed, sub-penny tick regime would have on the marketplace, the
investor experience, or issuers. It is an apples-to-oranges
comparison and is irrelevant as a basis for support'').
---------------------------------------------------------------------------
As explained in the following discussion, the Commission continues
to believe that the TSP provides a meaningful environment to study the
potential effects of a tick size change for the reasons articulated
below, even as the TSP has limitations for determining the exact effect
of the amendments to Rule 612.
First, the economics of being tick-constrained do not depend on the
absolute size of the tick in question. Rather, they depend on the
relationship between the economic spread \1242\ implied by the
economics of the stock and the quoted spread that is possible given the
tick size, regardless of the specific tick size. Specifically, when the
economic spread is narrower than a single tick, the negative effects of
being tick-constrained are expected to emerge for the reasons discussed
in section VII.D.1.b.i above.\1243\ Those reasons are independent of
the absolute size of the tick. They instead depend on the ratio of the
market price of liquidity to the tick, i.e., the lowest quoted spread
permitted by the tick size. In this context, the TSP analysis has
merit, even as it includes some stocks with quoted spreads wider than
1.5 cents (the cutoff for the amendments), because its purpose is to
gain insight into how stocks with various numbers of tick increments
intra-spread react to changing the tick. Addressing this question
necessitates considering stocks with wider quoted spreads.
---------------------------------------------------------------------------
\1242\ See supra note 990 and surrounding text for discussion of
the term economic spread id.
\1243\ See also id. for a discussion of the concept of economic
spread.
---------------------------------------------------------------------------
Second, as discussed above, a key factor in the economics of being
tick constrained is the implied quoted spread relative to the tick
size, not the market capitalization or any other the qualifying factors
for the TSP. Because the economics of being tick-constrained do not
depend on market capitalization, the findings derived from the TSP can
be usefully applied to a broader section of the market. Also, one
study,\1244\ referenced in the Proposing Release, specifically examined
only the most
[[Page 81708]]
liquid TSP stocks and removed stocks with very low prices.\1245\ The
authors' results indicate that, in this subset of the most liquid TSP
stocks, all key findings of the TSP not only hold, but that the
patterns of the results of the TSP on market outcomes tend to
strengthen.
---------------------------------------------------------------------------
\1244\ See Barardehi et al., supra note 231.
\1245\ See Proposing Release, supra note 11, at 80273 n.85.
---------------------------------------------------------------------------
Third, while the Commission acknowledges the difference between the
TSP and the amendments with regard to tick size splits--the TSP was a
1:5 tick size change while the amendments provide a 1:2 tick size
change for some stocks--the TSP provides meaningful information about
the likely direction of the effects due to a tick size change: that is,
whether market quality improves or declines when the tick size is
changed. The actual effect of a 1:2 split may differ from that observed
from the TSP's 1:5 split, but it is unlikely to go in the opposite
direction if the TSP were to indicate a market quality improvement when
the tick size is reduced. This is because the potential negative
effects of too many ticks intra-spread would be stronger for the 1:5
split associated with the TSP than with the 1:2 split associated with
the amendments.\1246\ Thus, it is unlikely that, were the TSP to show
an improvement in market quality associated with a 1:5 tick size split
for certain stocks, that there would have been an opposite effect with
a 1:2 tick size split.
---------------------------------------------------------------------------
\1246\ For example, the potential negative effects from sub-
pennying would be higher from a 1:5 split compared to a 1:2 splits
because the cost of gaining priority over other liquidity providers,
by updating the best price by a single tick, is lower with a smaller
tick. The risk liquidity fragmenting across price levels would also
be higher with a 1:5 split as compared to a 1:2 split. See supra
section VII.D.1.b.i for further discussion.
---------------------------------------------------------------------------
Some commenters stated that reducing the tick size below $0.01 was
opposed to the conclusions and analysis provided by the Commission when
adopting Rule 612 in 2005, and that the Commission did not provide
analysis explaining why it was reversing its opinion.\1247\ The
Commission disagrees that the analysis and conclusions associated with
the initial proposal and adoption of Rule 612 are inconsistent with the
analysis provided in the Proposing Release and repeated here. When
initially proposing and adopting Rule 612, the Commission acknowledged
that lowering the tick size from fractions to $0.01 improved the
trading environment.\1248\ It also expressed concern, as stated by
commenters, that further reducing the tick size for all stocks could
harm market quality via pennying and reduced liquidity at the top of
the book.\1249\ When originally proposing Rule 612, the Commission also
provided an analysis of sub-penny trading and quoting and stated that
there was, at the time, no industry standard for trading and quoting
increments.\1250\ The Commission's sub-penny analysis suggested that,
at the time, sub-penny trading was primarily used to facilitate
pennying because sub-penny trades congregated at $0.001 and $0.009
rather than having a uniform distribution or clustering midpoint prices
(i.e., in $0.005 increments), justifying the use of some minimum
pricing increment.\1251\
---------------------------------------------------------------------------
\1247\ See, e.g., Equity Market Structure Citadel Letter at 17,
Craig Louis Letter attached to Virtu Letter II at 32-33, and Virtu
Letter II at 16. See supra section VII.C.1.b for a discussion of the
Commission analysis referred to by commenters.
\1248\ See 2004 Regulation NMS Proposing Release, supra note 31,
at 11170.
\1249\ Id.; see also Proposing Release, supra note 11, at 80280
(``Minimum pricing increments that are too small can also add to
complexity in trading and increase the risk of stepping ahead'');
see also supra note 994 for the definition and discussion of
pennying.
\1250\ Id. at 11171. Although Nasdaq and the exchanges permitted
quoting in single penny increments, these markets allowed trades to
be printed in increments below a penny. Although certain online
brokers only accepted orders priced in one-cent increments, ECNs and
Nasdaq market makers accepted orders and executed trades in sub-
penny increments. While market makers quoted through Nasdaq only in
penny increments, they could display orders in ECNs in sub-pennies.
Exchanges, where the majority of trading volume occurred, were bound
by the Decimals Implementation Plan, which was ordered by the
Commission, and which ultimately established $0.01 as the tick size
for exchange quotes. Other market participants, however, were not so
bound leading to non-standard quoting increments across various
venues such as ECNs.
\1251\ Id. at 11169.
---------------------------------------------------------------------------
When adopting Rule 612, the Commission did not empirically analyze
whether a minimum pricing increment of $0.005 would have harmed or
helped market quality for some stocks, and specifically did not opine
on a tiered tick structure such as is being adopted. The analysis
provided therein was in the context of a uniform tick size applicable
to all stocks. Within this context, the Commission concluded that ``the
marginal benefits of a further reduction in the minimum pricing
increment [below $0.01 for all stocks] are not likely to justify the
costs to be incurred by such a move'' \1252\ The analysis provided in
this release does not disagree with that assessment. Applying a tick
size lower than $0.01 for all stocks could cause harm to stocks with
wide quoted spreads due to pennying concerns and fragmenting liquidity.
---------------------------------------------------------------------------
\1252\ Id. at 11170.
---------------------------------------------------------------------------
Additionally, the need to address tick-constrained stocks has
increased substantially in the subsequent nearly two decades as tick
constraints have become more pervasive over time. Table 3 indicates
that in 2023, about 74% of share volume was associated with securities
trading with quoted spreads at or below $0.015; following the same
methodology, in 2005 the figure was about 54%.\1253\ This statistic
understates the true increase in trading in tick-constrained securities
because overall average daily trading volume has more than doubled over
the same period of time.\1254\ Thus, precisely because average quoted
spreads have been coming down, the benefits of alleviating the tick
constraint have increased substantially since 2005. Additionally, as
discussed throughout this section, there has been considerable research
since the implementation of Rule 612 in 2005 by the Commission,
industry members, and academics surrounding tick sizes that did not
exist when Rule 612 was adopted. This research supports the notion that
a tick size below $0.01 will likely improve market quality for some
stocks.
---------------------------------------------------------------------------
\1253\ These patterns are not driven by a change in sub-dollar
trading (which may benefit from a narrower tick size); the patterns
are not materially changed when symbol-days with average prices
below $2 or $5 are dropped from the sample.
\1254\ This statistic is computed by comparing the average daily
share volume in all securities covered by WRDS Intra-day indicators
in 2005 and 2023. Additionally, total trading volume has also more
than doubled over that same time period. Thus, there is more trading
volume and more of it is trading in a tick-constrained environment.
---------------------------------------------------------------------------
One commenter illustrated its disagreement with the Commission's
use of the TSP analysis by presenting a hypothetical TSP in which the
tick size is increased from $0.01 to $0.15.\1255\ The commenter stated
that such a change ``would have negatively impacted a greater range of
stocks . . . and predictably liquidity conditions in those stocks would
have meaningfully improved at the end of the pilot when the changes
were reversed.'' \1256\ The commenter proceeded to state that ``this
experiment would not suggest that regulators should always reduce the
minimum quoting increment for tick-constrained symbols by a factor of
fifteen.'' \1257\ The commenter further stated that the TSP ``merely
reverted to the status quo after a failed experiment'', and this
``reversion provides no information on what would be expected to occur
if minimum quoting increments were further
[[Page 81709]]
reduced to levels that have never before been tested.'' \1258\
---------------------------------------------------------------------------
\1255\ See Citadel Letter I at 12.
\1256\ Id.
\1257\ Id. (emphasis in original).
\1258\ Id.
---------------------------------------------------------------------------
The Commission disagrees with this assessment in several respects
and continues to believe that analysis of the TSP provides meaningful
information for the effects of the amendments. The TSP enables analysis
that empirically tests whether market quality depends on being tick-
constrained. The TSP provides two events that can be used for this
test: one at the start of the TSP when tick sizes were increased for
certain stocks, and one at the end of the TSP where tick sizes for
those same stocks were decreased. Academic research shows that the
effects of both of these events are consistent with the theory that
stocks with few ticks intra-spread have worse market quality.\1259\ The
Commission provided its own analysis of the end of the TSP; the end of
the TSP involved a reduction in tick size, which directionally
corresponds to what will happen under the adopted rule. This analysis
found evidence to support the theory that stocks with too few ticks
intra-spread have worse market quality. The Commission therefore
disagrees that the TSP analysis ``provides no information'' as to the
effects of the adopted rules.
---------------------------------------------------------------------------
\1259\ See, e.g., Barardehi et al., supra note 231.
---------------------------------------------------------------------------
The commenter states that ``the end of the Tick Size Pilot provides
no basis for suggesting that regulators should always reduce the
minimum quoting increment for tick-constrained symbols by a factor of
five.'' The Commission does not reach the conclusion that regulators
should always reduce the minimum quoting increment by a factor of five,
and indeed the Commission is not adopting such a rule. As discussed
earlier in this section, TSP analysis indicates that for stocks with 1-
2 ticks intra-spread, reducing the tick size improves market quality on
average. While the Commission is reducing the tick size by a factor of
two rather than five for some stocks, the direction is likely to be the
same as what was observed in the TSP, though the magnitude may be
different.\1260\
---------------------------------------------------------------------------
\1260\ See supra note 1246 and surrounding text for additional
discussion.
---------------------------------------------------------------------------
Furthermore, the commenter assumes in the hypothetical experiment
of a bigger increase in the tick size, that this increase would have
``negatively impacted'' stocks. However, the fact that causing stocks
to become tick-constrained worsens their market quality is an
assumption made by the commenter. Absent evidence, such as the evidence
provided by the TSP, it is unclear upon what the commenter bases this
assumption. The ability to make this inference, that being tick-
constrained worsens market quality, is precisely why analyzing the TSP
is valuable because it provides the empirical result which permits one
to employ with confidence the commenter's assumption in its
hypothetical.\1261\
---------------------------------------------------------------------------
\1261\ More specifically, the commenter's hypothetical assumes
that the start of a large tick size increase would worsen liquidity,
then concludes this means that evidence from the end of the TSP is
uninformative because it simply reverses the effect. But, this
conclusion is incorrect because the TSP results from both the
imposition and conclusion of the TSP are what make the first
assumption credible.
---------------------------------------------------------------------------
With regard to the commenter's statement that the TSP conclusion
``provides no information on what would be expected to occur if minimum
quoting increments were further reduced to levels that have never
before been tested,'' \1262\ while it is true that the tick sizes in
the adopted amendments are not among the tick sizes implemented in the
TSP, this fact does not render the TSP analysis uninformative. The
theory that stocks that are tick-constrained will trade better with
more ticks intra-spread, successful as it was in predicting the market
quality effects of the end of the TSP, can be reasonably relied upon to
help determine the effects of the amendments. What matters is not the
magnitude of the spread, or the size of the tick, but the number of
ticks intra-spread.\1263\
---------------------------------------------------------------------------
\1262\ See Citadel Letter I at 12.
\1263\ See infra section VII.F.1 for a discussion of reasonable
alternative tick sizes.
---------------------------------------------------------------------------
One commenter stated that the analysis in the Proposing Release
should have accounted for both fixed effects and volatility, as well as
other variables.\1264\ Barardehi et al. (2022) provide estimates that
account for fixed effects and a number of control variables including
volatility.\1265\ This paper shows that the results shown in the
Proposing Release and repeated below are robust to these effects.
---------------------------------------------------------------------------
\1264\ See Virtu Letter II at 15.
\1265\ See Barardehi et al., supra note 231. The authors control
for market capitalization, dollar volume, average quoted spread, and
return volatility, see analysis associated with their table 11.
---------------------------------------------------------------------------
Commission Empirical Analysis: The Proposing Release provided a
review of the existing TSP empirical academic research.\1266\ This
research consistently found that stocks that became tick-constrained by
the TSP, on average, traded better across many market quality
dimensions when their tick size was reduced from $0.05 to $0.01. Some
analysis also showed that some stocks with wide spreads traded better
with the $0.05 tick than with the $0.01 tick. The empirical analysis in
the Proposing Release sought to identify the thresholds where the TSP
tick size change transitioned from harmful, to benign, to beneficial.
Specifically, table 8 provides analysis that examines the impact of the
end of the TSP on a wider range of quoted spread profiles than simply
tick-constrained or not. This analysis focuses on the end of the TSP,
when the tick size was reduced from $0.05 back to $0.01, because that
event more closely matches the amendments, which reduce the tick size.
---------------------------------------------------------------------------
\1266\ See Proposing Release, supra note 11, section V.D.1.
---------------------------------------------------------------------------
The analysis presented in table 8 uses a difference-in-differences
methodology to study the effect of lowering the tick size from $0.05 to
$0.01 on TSP stocks at the end of the TSP.\1267\ TSP treated and
control stocks are assigned near the end of the TSP into one of four
bins ranging from the most tick-constrained in the first bin to the
least constrained in the fourth bin.\1268\ Key variables such as quoted
depth and spreads were measured before and after the tick size was
lowered, and difference-in-differences estimation methods were
[[Page 81710]]
used to examine how these variables reacted to the tick size change.
The analysis uses ordinary least squares \1269\ and quantile (median)
regressions \1270\ to estimate the following regression model: \1271\
---------------------------------------------------------------------------
\1267\ Difference-in-differences is a statistical technique in
which the effect that a treatment has on some response variable is
estimated by comparing the average change in the response over time
in the treatment group to the average change in the control group.
\1268\ Bin assignments are calculated according to the stock's
average quoted spreads for May and June of 2018, near the end of the
TSP. Specifically, we use WRDS Intra-day indicators to collect the
time weighted quoted spread for all TSP and control stocks for each
trading day in May and June 2018. Then for each stock we calculate
the equally-weighted average quoted spread across all trading days.
Based on this average, TSP and control stocks are sorted into one of
four bins. The first bin is for stocks with quoted spreads ($0.00,
$0.06). Empirically, for stocks in the TSP, this bin includes stocks
that nearly always traded at the minimum quoting increment of $0.05
during the TSP. The second bin is for stocks with quoted spreads in
the range ($0.06, $0.09). For stocks in the TSP, this bin is said to
include those stocks with one to two ticks intra-spread during the
TSP. The third bin is for stocks that had quoted spreads of ($0.09,
$0.15) or approximately 2-3 ticks intra at a $0.05 tick increment.
The fourth bin is for stocks with quoted spreads greater than $0.15.
The TSP had three test groups: the first group applied the $0.05
tick only to quoting, the second group applied the $0.05 tick to
quoting and trading (with exceptions for benchmark and midpoint
trades and for certain retail price improvement trades), and the
third group applied the $0.05 tick to trades and quotes the same as
the second group but also had a trade at rule applied. Barardehi et
al., supra note 231 provide similar analysis, and also expand the
analysis in many dimensions. Their analysis finds evidence that all
key results presented here are robust along many dimensions
including the test group analyzed and to many other factors
including fixed effects and volatility--factors that one commenter
suggested that the Commission should consider in their TSP analysis,
see Virtu Letter II at 17.
\1269\ Ordinary least squares (OLS) regression refers to a
statistical technique for estimating the linear relationship between
an independent variable and dependent variables by minimizing the
sum of squared errors between the estimate and the observed
independent variable. The use of OLS and quantile regressions is
common in the literature on the TSP pilot.
\1270\ The primary advantage to quantile regressions is that
they are less sensitive to outliers that can affect mean inference
in OLS. Thus, median regressions provide additional robustness to
the analysis and ensure that results are not driven by outliers.
\1271\ In this equation the variable Y denotes the response
variable of interest such as quoted spread and depth. The subscripts
j and t serve to index stocks and days respectively.
[alpha]0, [alpha]p, [alpha]e, and
[beta] are coefficients (to be estimated), and uj,t is
the error term. Pilotj is an indicator variable that
equals 1 if stock j was in the treatment group, or 0 if stock j was
in the control group. Eventt is an indicator variable
which is equal to 1 if the day t was post the treatment event and
equals 0 otherwise. Table 8 reports the difference-in-differences
estimator of [beta] for a different response variable Y across the
different quoted spread bins. One commenter criticized this model
for failing to include fixed effects and not controlling for other
criteria such as volatility. See Virtu Letter II at 15. A very
similar analysis, which did consider fixed effects and a host of
control variables including volatility, is included in Barardehi et
al., supra note 231. Their analysis showed that all key results were
economically unchanged when considering fixed effects and a host of
control variables.
Yj,t = [alpha]0 + [alpha]pPilotj + [alpha]EEventt + [beta](Pilotj x
---------------------------------------------------------------------------
Eventt) + microj,t
where the quantile regression optimizes: \1272\
---------------------------------------------------------------------------
\1272\ In this equation uj,t is the error term from
the previous regression specification equation, supra note 1271, and
the loss function is defined as: [rho][tau](u) = [tau]
max(u,0) + (1-[tau]) max(-u,0) ; where 0 < [tau] < 1.
[GRAPHIC] [TIFF OMITTED] TR08OC24.002
Table 8--Effects of a Reduction in Tick Size on Quoting and Trading Outcomes \a\
--------------------------------------------------------------------------------------------------------------------------------------------------------
OLS Quantile (median) regression
-------------------------------------------------------------------------------------------------------
Spread bin # Quoted spread ($) May & June 2018 Quoted spread ($) May & June 2018
-------------------------------------------------------------------------------------------------------
1st 2nd 3rd 4th 1st 2nd 3rd 4th
--------------------------------------------------------------------------------------------------------------------------------------------------------
Depth (100 shares).............................. *** -22.5 *** -5.30 *** -1.55 -0.51 *** -11.8 *** -3.16 *** -0.96 *** -0.21
[-12.02] [-7.09] [-4.40] [-1.30] [-16.99] [-23.52] [-17.81] [-4.30]
Depth ($1,000).................................. *** -16.7 *** -8.41 *** -4.67 *** -2.06 *** -11.2 *** -7.27 *** -3.96 *** -1.48
[-14.58] [-10.94] [-7.82] [-3.66] [-22.04] [-20.70] [-12.58] [-4.14]
Quoted Spread ($)............................... *** -0.033 *** -0.027 *** 0.023 *** 0.12 *** -0.034 *** -0.031 ** 0.012 *** 0.12
[-18.71] [-6.46] [2.99] [5.51] [-35.41] [-10.31] [2.03] [6.80]
Relative quoted Spread.......................... *** -0.0049 * -0.00097 0.00034 *** 0.0046 *** -0.0041 *** -0.0014 0.00021 *** 0.0034
[-9.59] [-1.80] [0.53] [3.30] [-8.54] [-6.89] [0.74] [4.66]
Effective spread ($)............................ *** -0.027 -0.026 *** 0.029 ** 0.038 *** -0.026 *** -0.021 -0.0018 *** 0.051
[-4.97] [-1.43] [5.17] [2.16] [-58.10] [-12.81] [-0.63] [4.81]
Relative eff. spread............................ *** -0.0039 0.00043 0.0055 *** 0.0028 *** -0.0030 *** -0.0010 -0.00013 *** 0.0016
[-3.12] [0.17] [1.36] [4.42] [-10.78] [-9.58] [-1.09] [3.23]
Cancel-to-trade................................. *** 5.10 *** 6.69 *** 7.56 *** 18.8 *** 4.56 *** 5.49 *** 6.87 *** 12.3
[5.99] [6.38] [6.79] [8.84] [7.75] [7.79] [10.44] [10.61]
Odd-lot rate (%)................................ *** 4.89 *** 5.61 *** 2.85 ** 1.49 *** 5.59 *** 6.39 *** 3.29 ** 1.85
[9.62] [8.04] [4.35] [2.15] [8.02] [8.99] [4.72] [2.51]
Realized spread ($)............................. *** -0.014 *** -.0099 .00037 *** 0.040 *** -0.014 *** -0.013 *** -0.0068 *** 0.038
[-27.94] [-7.43] [0.12] [4.45] [-48.36] [-17.96] [-5.13] [5.64]
Relative real. spread........................... *** -.0024 -.00032 -.00039 ** .0014 *** -.0014 *** -.00054 *** -.00013 *** .0012
[-11.82] [-1.36] [-1.25] [2.37] [-14.08] [-12.52] [-2.77] [3.65]
Volume (1,000 shares)........................... 26.5 ** 30.3 12.5 -5.41 19.1 3.35 0.20 ** -3.25
[1.30] [2.13] [1.32] [-1.07] [1.42] [0.40] [0.04] [-2.44]
Cum Depth 10c from mdpt......................... *** -0.17 *** -0.26 ** -0.27 ** -0.34 *** -0.49 *** -0.54 *** -0.45 ** -0.63
[-3.93] [-5.00] [-2.59] [-2.37] [-5.51] [-6.29] [-4.91] [-3.15]
Cum Depth -10c from mdpt........................ *** -0.22 *** -0.19 *** -0.37 ** -0.45 *** -0.49 *** -0.42 *** -0.50 ** -0.79
[-5.28] [-3.74] [-3.44] [-2.83] [-6.33] [-5.21] [-5.68] [-2.75]
CRT 10 round lots............................... *** -0.026 -0.001 *** 0.035 *** 0.14 *** -0.037 *** 0.085 *** 0.035 ** 0.075
[-19.56] [-0.19] [3.99] [1.03] [-2.72] [2.75] [4.20] [2.37]
--------------------------------------------------------------------------------------------------------------------------------------------------------
\a\ This table presents the effects of a reduction in minimum tick size from $0.05 to $0.01 cent on various quoting and trading outcome variables. The
first bin is for stocks with quoted spreads ($0.00, $0.06). The second bin is for stocks with quoted spreads in the range ($0.06, $0.09). The third
bin is for stocks that had quoted spreads of ($0.09, $0.15). The fourth bin is for stocks with quoted spreads greater than $0.15. A difference-in-
differences regression with no control variables is estimated using data covering Control, Test Group 2, and Test Group 3 TSP stocks from 08/01/2018-
11/30/2018. All observations are at the stock day level. For each outcome variable Yjt, listed in the left-hand side column, the table presents only
the difference-in-differences coefficient estimates that indicate the effect of the TSP on the dependent variable. Estimates are performed by past
quoted spread subsamples that decompose the sample based on average quoted spreads during May-June of 2018. Among the outcomes' variables, the quoted
spread refers to the distance between the NBBO midpoint and the NBBO quote. The effective spread is the distance between the NBBO midpoint and the
realized trade price; the realized spread is the distance between a future NBBO midpoint (5-minutes ahead) and the trade price. Relative spread
measures are calculated as the spread scaled by the NBBO midpoint. The cancel-to-trade ratio is the daily number of order cancellations divided by the
number of trades, for displayed orders. The odd-lot rate is the percentage of trades in a day which executed against an odd-lot quote. CRT 10, or the
cost of a round-trip trade of 10 round lots, measures the cumulative transaction costs from buying and then immediately selling 10 round lots. The CRT
assumes that an order that is larger than the displayed depth at the best price will not execute in full at that price. Instead, the assumed unfilled
portion will execute at worse prices until completely filled with displayed depth. All data are Winsorized at the 1% and 99% level. The numbers in the
[ ] brackets reflect t-statistics that are based on two-way stock-and-date clustered standard errors. Symbols *, **, and *** reflect statistical
significance at 10%, 5%, and 1% type-1 error levels.
As discussed in the Proposing Release, this analysis provides
evidence of a fundamental tradeoff between accurate pricing on one hand
and incentives for liquidity provision on the other. Across all
specifications, the end of the TSP was associated with a decrease in
depth at the NBBO, when the tick size was reduced from $0.05 to $0.01,
as signified by the negative and, in most cases, statistically
significant coefficients reported. The reduction in shares available at
the NBBO was the greatest for stocks with tighter quoted spreads and
smaller for stocks with wider quoted spreads. The finding that tighter
quoted spread stocks experience the greatest decline in depth at the
[[Page 81711]]
NBBO is consistent with the idea that, for these stocks, the $0.05 tick
was the most constraining, and so liquidity that would have naturally
spread out within the quoted spread given a smaller tick, bunched at
the wider tick increments, and that once the tick-constraint was
relaxed this liquidity naturally spread out over the additional price
levels. For less tick-constrained stocks, the bunching was less severe
since liquidity already had some room to spread out.
One commenter stated that the Commission did not provide any
analysis of the effect of the proposal on displayed liquidity and
liquidity deeper in the book, including with respect to less liquid
securities and during times of market stress.\1273\ The Proposing
Release did examine the effect of a tick reduction on displayed
liquidity and cited academic literature for further analysis.\1274\ The
same commenter stated that reducing the tick as presented in the
proposal would reduce depth at the NBBO by more than 82%.\1275\ The
Commission acknowledges that, given the magnitude of the reduction in
the proposal, it is conceivable that such a reduction could have
occurred for some stocks. Barardehi et al. (2022) document that depth
at the NBBO was 50% lower with the $0.01 tick compared to the $0.05
tick. However, this was only true for tick-constrained TSP stocks. For
stocks with wide quoted spreads, depth was only about 16% lower with
the smaller tick size. The TSP was associated with a 1:5 tick size
split, and the proposal that the commenter was commenting on would have
created a 1:10 split for some stocks relative to the baseline. In
contrast, the adopted amendments create a smaller split than either a
1:5 or a 1:10 split. The Commission does expect depth at the NBBO to
decrease for stocks receiving the smaller tick size with the TSP
analysis providing a likely higher end estimate of the magnitude of the
decrease since the TSP was a bigger change to the baseline than the
adopted amendments.
---------------------------------------------------------------------------
\1273\ See Citadel Letter I at 9.
\1274\ See Proposing Release, supra note 11, section V.D.1.
\1275\ See Citadel Letter I at 5.
---------------------------------------------------------------------------
For stocks in the first or second bins, table 8 demonstrates that
lowering the tick to $0.01 leads to significantly lower quoted spreads.
These stocks went from having approximately 1-2 ticks inside the quoted
spread, with a $0.05 tick, to having 1-10 ticks inside the quoted
spread, with a $0.01 tick. This finding is consistent with the idea
that for stocks that are tick-constrained the effect of decreasing the
tick size will narrow quoted spreads by improving competition. For the
stocks in the third and fourth bins, the story is different, as the
reduction in the tick size leads to wider quoted spreads. These stocks
went from having more than two or more ticks within the quoted spread,
with a $0.05 tick, to having more than 10 ticks within the quoted
spread, with a $0.01 tick. This result is consistent with the idea that
for wider quoted spread stocks, the prevailing effect of reducing the
tick size was to increase transaction costs and widen spreads by
fragmenting liquidity and increasing the risk of pennying which made
trading more costly leading to wider quoted spreads. This pattern of
results--namely narrower spreads for the first and second bins and
wider spreads for the fourth--holds regardless of whether dollar
spreads, relative spreads, OLS, or quantile regressions are used,
suggesting this is a robust outcome of the end of the TSP.
The pattern for effective spreads is similar to that observed for
quoted spreads. Effective spreads measure the average realized
transaction cost for trades as it measures the absolute distance
between the realized trade price and the NBBO midpoint at the time of
the trade. Effective spreads do not always equal quoted spreads because
trades can execute inside the NBBO for numerous reasons, such as odd-
lot trades, midpoint trades, and hidden orders. For stocks in bin one--
i.e., stocks for which the $0.05 tick was the most restrictive--all
specifications suggest that reducing the tick size was associated with
a decrease in realized transaction costs as measured by effective
spreads. For stocks in bin four, those with the widest quoted spreads
prior to the tick size reduction, all specifications suggest that the
reduction in the tick size leads to an increase in transaction costs,
measured by effective spreads. For stocks in between these extremes in
bins two and three, the results are not as uniform. For stocks in bin
two, the sign of the coefficients for all estimates (dollar effective
spreads, relative effective spreads, OLS, and quartile regressions)
suggests lowering the tick size decreased effective spreads, although
not all specifications agree as to statistical significance. The OLS
regressions suggest that the effect was statistically insignificant,
while the quantile regressions found a statistically significant effect
and suggest that effective spreads decreased. For stocks in the third
bin, the analysis did not find a consistent, statistically significant
change in effective spreads, or in other words, moving from roughly two
to three ticks within the spread to ten to fifteen ticks did not appear
to reliably help or harm transaction costs as measured by effective
spreads.
These results, like the results for quoted spread, suggest that for
stocks for which the narrowing of the tick size meant that the stock
went from having less than 2 ticks within the quoted spread to 1-10
ticks within the quoted spread, the effect of reducing the tick was
beneficial in terms of reducing transaction costs. For stocks with very
wide quoted spreads, reducing the tick size appeared to harm liquidity,
which is consistent with fragmentation and pennying being the
prevailing effect.
The theoretical discussion above suggests that executing a larger
order may become more complex with a smaller tick size--meaning it may
take visiting more venues as well as executing across more price levels
to execute an order with a smaller tick size. This potential outcome is
explored using the ``cancel-to-trade'' ratio. A higher ratio indicates
more frequent canceling of orders per the amount of trading volume, and
it is an indication that market participants are more active in
managing their quotes and their order strategies. In this analysis,
both the OLS and the quantile regressions confirm that a smaller tick
resulted in a statistically significant increase in the cancel-to-trade
ratio, suggesting more complexity. Additionally, the magnitude of the
effect is increasing in the quoted spread, with wider quoted spreads
having larger coefficients, suggesting a larger effect in the cancel-
to-trade ratio for stocks with wider spreads. This pattern is
consistent with pennying and increased complexity having a greater
impact on stocks with wider quoted spreads. These stocks are unlikely
to receive the smaller tick size under the adopted amendments.
The analysis also looks at the effect of lowering the tick size at
the end of the TSP on the usage of odd-lot orders. Across all quoted
spread bins, the usage of odd-lot orders increases when the tick size
decreases. This finding is consistent with the notion that liquidity
will be spread out over more levels leading to an increased use of odd-
lot orders to allow liquidity providers to offer smaller levels of
liquidity at finer price increments.\1276\ This result also suggests
that a lower tick size increases the need for market participants to
have ready access to odd-lot information given that the lower tick size
can be expected to increase the usage of odd-lot quotes.
---------------------------------------------------------------------------
\1276\ See also Virtu Letter II at 6.
---------------------------------------------------------------------------
[[Page 81712]]
Effective spreads provide a measure of liquidity providers' revenue
for the immediate execution of an incoming order and the contrasting
economic effects also have implications for how liquidity providers'
revenue will be affected by a lower tick. The effective spread captures
the liquidity premium, paid by those submitting orders for immediate
execution, and can theoretically be decomposed into two components:
Effective Spread = Realized Spread + Price Impact.\1277\ One component
of the effective spread is the price impact or adverse selection
component. It is the change in the NBBO midpoint at the time of trade
to some point in the future. This component of the effective spread
captures the portion of the effective spread liquidity providers lose
from trading with investors who are more informed than they are and is
also referred to as the adverse selection component of the bid-ask
spread. The remainder of the effective spread, after removing the
adverse selection component, is the realized spread. This portion of
the effective spread acts as a proxy \1278\ for the compensation to the
liquidity provider for its non-adverse selection costs. If a smaller
tick decreases revenue for liquidity providers, by allowing bid and ask
prices to more accurately reflect supply and demand, then this effect
should manifest as a decrease in realized spreads for liquidity
providers. However, if increased order book fragmentation and pennying
risk increase the cost of providing liquidity, then liquidity providers
will need to be compensated for these costs in order to provide
liquidity and, thus, realized spreads will increase. To the extent that
the two effects offset one another, realized spreads might not change.
---------------------------------------------------------------------------
\1277\ Effective spreads can be interpreted as what liquidity
providers expect to earn from providing liquidity, assuming that
prices do not change before the liquidity provider is able to unwind
its position and realize its profit. Under this interpretation,
realized spreads would proxy for what liquidity providers actually
earn, taking into account that the market price may have moved
against the liquidity provider before it could unwind its position.
Effective Spread = Realized Spread + Price Impact. For a full
mathematical decomposition of effective spreads into realized spread
and price impact components see Peter N. Dixon, Why Do Short Selling
Bans Increase Adverse Selection and Decrease Price Efficiency, 11
Rev. Asset Pricing Stud. 122 (2021) app. at 165.
\1278\ Realized spreads do not measure the actual trading
profits that market makers earn from supplying liquidity. In order
to estimate the trading profits that market makers earn, we would
need to know at what times and prices the market maker executed the
off-setting position for a trade in which it supplied liquidity
(e.g., the price at which the market maker later sold shares that it
bought when it was supplying liquidity). If market makers offset
their positions at a price and time that is different from the NBBO
midpoint at the time lag used to compute the realized spread measure
(Rule 605 realized spread statistics are measured against the NBBO
midpoint 5 minutes after the execution takes place), then the
realized spread measure is an imprecise proxy for the profits market
makers earn supplying liquidity.
---------------------------------------------------------------------------
For tick-constrained stocks in bin one, the analysis indicates a
decrease in realized spreads across all specifications, and when using
dollar or relative realized spreads when the tick size was reduced from
$0.05 to $0.01. This result is consistent with the notion that
liquidity providers' non-adverse selection revenues will decrease due
to bid and ask prices being more reflective of supply and demand with a
smaller tick. The opposite occurs for stocks with wide quoted spreads
in bin four, where realized spreads increase significantly--consistent
with liquidity providers needing to be compensated for the increased
cost and complexity associated with trading a wide quoted spread stock
in a small tick environment. For stocks in the middle two bins, the
effect of lowering the tick size on realized spreads is unclear, as
about half of the specifications indicate no change in realized spreads
while the other half indicate lower effective spreads.
One commenter states that the Proposing Release did not address how
the rule would affect institutional transaction costs, given that
institutional traders frequently trade, and are concerned with sourcing
larger quantities.\1279\ However, as considered here, and in the
Proposing Release, the Commission considered multiple measures of depth
beyond the NBBO, and Barardehi et al. (2022) also considers more. This
depth beyond the top of the book analysis uses MIDAS data to study how
the tick size change affected liquidity deeper in the book. Analyzing
liquidity deeper in the book is valuable because it gives an indication
of how trading larger orders, which must go deeper in the book to be
fulfilled may be affected by a change in the tick size. Specifically
this analysis calculates the daily average cumulative shares available
at $0.10 above and below the midpoint for control and treated stocks,
and uses the same difference-in-differences analysis to examine the
effect of reducing the tick size on cumulative depth.\1280\ Our
analysis suggests that reducing the tick size also reduced the total
depth available deeper in the book with the coefficient for bin 4--
i.e., those with the widest quoted spreads--being the largest in
magnitude. This finding is consistent with a smaller tick discouraging
the posting of displayed liquidity due to pennying concerns for stocks
with wide quoted spreads.
---------------------------------------------------------------------------
\1279\ See MEMX Letter at 10.
\1280\ In the regressions we take the natural log of shares
available. This conversion helps standardize shares available for
stocks with different prices by making the interpretation in terms
of percentage changes. See also e.g., STA Letter at 6 and CCMR
Letter at 24 suggesting that a smaller tick size could affect depth
deeper in the book.
---------------------------------------------------------------------------
These depth of book findings do not directly imply that trading
deeper in the book became more expensive for two reasons. First,
research suggests the use of non-displayed quotations increases
significantly when the tick size is reduced.\1281\ Thus the decline in
liquidity that we document is only a decline in displayed liquidity.
Second, quotes tend to congregate at the price just worse than the
quoter's desired price so that the quoter does not lose money on a
transaction. When a wider tick is tightened, quotes that were
previously congregated at the wide tick will spread out at prices
better than the previous tick allowed. Thus, a market participant
taking liquidity from multiple price layers in the order book to
fulfill an order will have some shares that transact at superior prices
than it would have with the wider tick.\1282\
---------------------------------------------------------------------------
\1281\ See analysis presented in Nasdaq Intelligent Tick
Proposal, supra note 150; see also Justin Cox, et al., Increasing
the Tick: Examining the Impact of the Tick Size Change on Maker-
Taker and Taker-Maker Market Models, 54 Fin. Rev. 417 (2019); Amy K.
Edwards, et al., The Effect of Hidden Liquidity: Evidence from an
Exogenous Shock (working paper Mar. 1, 2021), available at https://ssrn.com/abstract=3766512 (2021) (retrieved from SSRN Elsevier
database).
\1282\ Consider a numeric example. A market with a $0.05 tick is
quoting asks of 500 shares at $10.05 and 500 shares at $10.10. An
investor wishing to purchase 700 shares would purchase 500 at $10.05
and 200 at $10.10 for a total price of $7,045. If the tick shrinks
to $0.01 and cumulative shares posted decline by 20%--for example--
but those shares are spread evenly over the finer grid then there
would be 80 shares at each price level from $10.01 to $10.10. An
investor wishing to buy 700 shares would need to purchase 80 shares
at each price level from $10.01 to $10.08 and 60 shares at $10.09
for a total purchase price of $7,034. So even though total depth
declined, the cost to execute a 500-share trade would decrease due
to more efficiently spreading liquidity across more price levels.
---------------------------------------------------------------------------
Table 8 also presents the effect of the TSP conclusion on the
round-trip cost to transact a trade for 10 round lots (1,000
shares).\1283\ This analysis suggests mixed results for the effect of
the tick size reduction on the cost of
[[Page 81713]]
executing a 10-round lot trade. For bin 1 stocks, the total round-trip
cost of a 10-round lot trade decreased when the tick size was lowered--
suggesting an improvement in liquidity deeper in the book. For stocks
in bin 2 (i.e., less tick-constrained stocks), the effect was not
clear. The OLS regressions suggested no effect, while the quantile
regressions suggested an increase in trading cost. For stocks in bins 3
and 4 (i.e., those that were not tick-constrained by the $0.05 tick),
the effect of lowering the tick size was to increase transaction costs
for larger trades. These results cohere with the idea that when stocks
are tick-constrained the pricing efficiency made possible by a smaller
tick improves liquidity, and for stocks with wider quoted spreads a
smaller tick harms liquidity by making individuals less willing to post
displayed liquidity due to complexity and the risk of pennying.
---------------------------------------------------------------------------
\1283\ A round-trip trade refers to executing an order to buy or
sell the stock and immediately reversing the position with an equal
countervailing order. We compute the cost of a roundtrip trade
following the methodology laid out in Griffith and Roseman (2019),
supra note 1002 and Chung, et al., supra note 1209. The methodology
uses MIDAS data to take snapshots of the order book at 15-minute
increments throughout the trading day and calculates the transaction
costs associated with walking the book up 5 or 25 round lots to
execute a large trade.
---------------------------------------------------------------------------
In conclusion, the analysis provided here suggests that, for stocks
that were limited to just 1-2 ticks intra-spread by the $0.05 tick, the
reduction to a $0.01 tick provided an improved trading environment.
Thus, trading in an approximate 1-10 tick range intra-spread provided a
superior environment to trading in a 1-2 ticks intra-spread range. One
caveat here is that the analysis highlights a key tradeoff with a
smaller tick for stocks with narrow quoted spreads. They tend to have
less depth at the NBBO, but narrower quoted spreads. Thus, the total
effect of this tradeoff on execution costs is largely a function of the
size of the trade being implemented with smaller trades receiving
improved terms while sufficiently large trades get worse terms with a
narrower quoted spread. However, as discussed in greater detail in
Barardehi et al. (2022), for tick-constrained stocks the point in terms
of trade size at which a tick size reduction harms execution quality is
quite large, around 50 round lots.\1284\ Additionally, for stocks with
quoted spreads greater than $0.15, where a $0.01 tick implied more than
15 ticks intra-spread, a $0.05 tick where there were only 3 ticks
intra-spread, appeared to provide a superior trading environment. For
stocks with quoted spread between $0.10 and $0.15, it is not clear
which tick size provided a superior trading environment.
---------------------------------------------------------------------------
\1284\ See Barardehi et al. (2022) Table 4. See also Citadel
Letter I at 11 requesting an analysis of the joint impact on depth
and quoted spread. See also Virtu Letter II at 18 stating that lower
quoted spreads would harm markets by reducing the incentive to post
liquidity.
---------------------------------------------------------------------------
In figure 2, data analysis shows how the quoted spread of a stock
during the TSP (``pre-shock dollar quoted spread'') correlates with how
investor transaction costs, as captured by effective spreads, changed
when the TSP ended. For stocks with an average of fewer than two ticks
intra-spread (i.e., those with pre-shock quoted spreads of $0.10 or
less), a reduction in tick size from 5 cents to 1 cent significantly
reduces effective spreads.\1285\ For stocks with an average of more
than three ticks intra-spread (i.e., those stocks with pre-shock quoted
spreads greater than $0.15), a narrower tick size increases effective
spreads. These results are broadly consistent with the findings
reported in table 8.
---------------------------------------------------------------------------
\1285\ See Proposing Release, supra note 11, at 80322-80323,
including figure 2, which is also in Barardehi et al., supra note
231.
---------------------------------------------------------------------------
[[Page 81714]]
[GRAPHIC] [TIFF OMITTED] TR08OC24.003
The Commission's results in table 8 provide useful information for
predicting how the tick size reduction associated with the amendments
may affect market quality for stocks priced at, or greater than, $1.00
per share and that receive the $0.005 tick size compared to the current
baseline. For stocks with prevailing quoted spreads less than $0.015
there would generally be at most 1.5 ticks intra-spread with a $0.01
tick, or 3 ticks intra-spread with a $0.005 tick. The analysis in table
8 for bin one stocks suggests that 1-5 ticks intra-spread provides a
better trading environment than does just one tick intra-spread.\1286\
Additionally, the results for bin 2 stocks suggest that moving from 1-2
ticks intra-spread to 5-10 ticks also generally improves market quality
across most measures. In short, the TSP analysis suggests that stocks
with fewer than 2 ticks intra-spread on average benefited from a tick
size reduction. Consequently, this analysis provides support for the
belief that reducing the tick size for stocks that generally have at
most 1.5 ticks intra-spread is likely to improve market quality for
these stocks. One concern discussed earlier in this section is that a
reduction that is too aggressive could harm market quality by providing
too many ticks intra-spread. However, the analysis provided in table 8
does not support this outcome since the Tick Size Pilot stocks in bin 1
and bin 2 still saw market quality improvements, implying that narrow
tick concerns didn't yet dominate the effect on these stocks, it is
unlikely that the smaller tick size reduction associated with this Rule
would lead to small tick problems diminishing market quality.
---------------------------------------------------------------------------
\1286\ One academic theoretical paper suggests that having a
two-tick quoted spread is optimal. See Sida Li & Mao Ye, Discrete
Prices, Discrete Quantities, and the Optimal Price of a Stock
(working paper Mar. 8, 2021, revised Jul. 7, 2023), available at
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3763516
(retrieved from SSRN Elsevier database). The paper suggests that
stocks reach their optimal price whenever the quoted spread is two
ticks wide. While the paper advocates for a lower tick size,
particularly for tick-constrained stocks, the two-tick quoted spread
conclusion is the result of a highly stylized trading model which
does not take into account pertinent factors from outside the model
which likely affect quoted spreads such as considerations of time
priority and pennying concerns. Conditional on there being non-
infinitesimal tick and round-lot sizes, their model suggests that a
two-tick wide quoted spread is optimal. Otherwise, their model
suggests an optimal policy choice of infinitesimal tick and round-
lot sizes.
---------------------------------------------------------------------------
Additional Sources of Information: Commenters also suggested
additional settings to identify when stocks are trading with the
optimal number of ticks intra-spread. One commenter suggested
[[Page 81715]]
that the Commission should have considered the European Union's tick
size approach, associated with MiFID II, which assigns one of over 20
variable tick sizes based on trading price and number of transactions
per day, and Japan's tick size approach.\1287\
---------------------------------------------------------------------------
\1287\ See CCMR Letter at 27.
---------------------------------------------------------------------------
One paper on the European experience cited by commenters studied
the effects of a new tick regime introduced by MiFID II on 511 stocks
listed on Euronext Paris and found results similar to the TSP analysis
in table 8.\1288\ In this study the authors point out that MiFID II led
to tick size increases for 339 stocks and decreases for 82 stocks. Tick
increases were followed by a widening of the quoted spread, an increase
in depth near the top of the book, and a reduction in message traffic.
Like the TSP analysis in table 8, tick decreases were followed by a
narrowing of quoted spreads, reductions in depth at the top of the
book, and an increase in message traffic; for a relatively wide price
interval, depths remained unchanged. Table 8 finds a reduction in the
cost of a round lot trade when low quoted spread securities experience
a reduction in the tick; similarly, the analysis of MiFID II documents
a reduction in transaction costs when the tick is reduced and the size
of the trade is held constant. In sum, the effects of a change in tick
size on Euronext Paris largely mirror the effects documented with the
TSP in table 8, indicating that the Commission's analysis is
documenting a generalizable phenomenon.
---------------------------------------------------------------------------
\1288\ Autorit[egrave] Des March[egrave]s Financiers (AMF),
MiFID II: Impact of the New Tick Size Regime (March 2018), available
at https://www.amf-france.org/sites/institutionnel/files/contenu_simple/lettre_ou_cahier/risques_tendances/MiFID%20II%20Impact%20of%20the%20New%20Tick%20Size%20Regime.pdf.
This paper was cited in Nasdaq Letter I at 10; Tradeweb Markets
Letter at 4; Robinhood Letter at 50-51; and IEX Letter I at 13.
---------------------------------------------------------------------------
However, there are several limitations to using studies of Europe's
tick size regime. Research based on experience with the E.U. regime is
difficult to apply to this Rule because the criteria in this Rule for
determining the tick size is the TWAQS, which is not a factor in the
European setting. When tick sizes change in Europe, it is due to
changes in price or trading volume which can simultaneously affect
quoted spreads. The Commission is unaware of research using the tick
sizes associated with MiFID II to identify the thresholds, in terms of
quoted spread, where a stock likely benefits or is harmed by a
modification of the tick size, and the commenter did not provide such
research.
In addition, it could be difficult to apply such analysis to the
U.S. setting due to a number of structural differences between the
European market and the U.S. markets. Key among these is that European
financial markets are not as integrated as the U.S. national market
system. For example, there is currently no consolidated tape or
requirement to route orders to the exchanges with the best prices in
the E.U.\1289\ This fact can affect inference in the context of
analyzing this Rule because it means that existing studies of the
effect of the European tick size regime are generally limited to one
exchange (e.g., the London Stock Exchange).\1290\ This can be a
significant limitation when trying to apply insights from European
studies to U.S. markets because, as has been found in existing
research, using data from one exchange as compared to across all
exchanges can lead to opposite market quality effects being
documented.\1291\
---------------------------------------------------------------------------
\1289\ See Rule 611 and supra note 226.
\1290\ See, e.g., Eros Favaretto et al., Impact of MiFID II
Tick-Size Regime on Equity Markets--Evidence From the LSE, 29 Eur.
Fin. Mgmt. 109, 109-149 (2023).
\1291\ For example, Chung, et al., supra note 1209, which uses
data from all US exchanges, and Griffith and Roseman (2019), supra
note 1002, who use data from just Nasdaq produce opposite findings
in some areas concerning the effect of the TSP on various dimensions
of market quality--i.e., the effect of the TSP on the cost to trade
very large orders.
---------------------------------------------------------------------------
As discussed above, a commenter also suggested that the Commission
should have considered the effects of tick size modifications in Japan
to inform the appropriate thresholds for the tick size change
considered in this Rule.\1292\ The Tokyo Stock Exchange (TSE) assigns
tick sizes ranging from 0.1 (JPY) to 100,000 (JPY) depending on the
price of the security in question. Stocks in the TOPIX 100 index have a
different tick size schedule than do other securities. Again, making
inferences from the TSE to this Rule is difficult due to the structural
differences between the way that the tick sizes operate. On the TSE,
tick sizes are price determined, under this Rule, tick sizes depend on
the prevailing quoted spread.\1293\ Additionally, trading in Japan is
largely consolidated on the TSE, whereas in the United States it is
considerably more fragmented suggesting the same issue relating to
inference as can occur with the European studies.
---------------------------------------------------------------------------
\1292\ See supra note 1287.
\1293\ See infra section VII.D.1.b.iii for additional discussion
of price as a determinate of tick-constrained securities and thus of
the tick size.
---------------------------------------------------------------------------
In considering the experience in the Japan markets, one study used
the median quoted spread to divide stocks on the TSE and examines
market quality.\1294\ These researchers find that a tick size reduction
reduces quoted spreads for both the narrower and wider quoted spread
stocks--although the effects are considerably larger for narrower
quoted spread stocks. The results of this study are consistent with
those in this release as it documents that for some stocks,
particularly those with narrower quoted spreads, reducing the tick size
can reduce quoted spreads. This study also documents that depth at the
best prices also tends to decrease with a smaller tick. However, this
study has limitations in the context of using it to determine the
optimal tick to quote size ratio because the authors do not separate
stocks using a nominal quoted spread threshold which would allow
inference about how stocks with quoted spreads above or below a certain
threshold were affected by the tick size. Rather, they simply bifurcate
the sample in half which doesn't apply a specific quoted spread
threshold and so it is difficult to discern from their study what the
optimal tick to quoted spread threshold will be.
---------------------------------------------------------------------------
\1294\ See Ingrid M. Werner et al., Tick Size, Trading
Strategies and Market Quality, 69 Mgmt. Sci. 3818 (2023).
---------------------------------------------------------------------------
One commenter presented two industry studies based on reverse stock
splits showing large reductions in percentage spreads and hence trading
costs following reverse splits.\1295\ When a stock undergoes a reverse
split, its share price rises. All else equal, one would expect quoted
spreads to widen in proportion so that the trading cost remains
constant as a percentage of price. If the stock is tick-constrained,
however, a reverse split causes the current penny tick to become lower
relative to the (higher) post-split price, relieving the tick
constraint and reducing trading costs as a percentage of the trade
amount.\1296\ One study presented evidence from GE's eight-for-one
reverse split, and showed that trading costs--as measured by the quoted
spread as a fraction of the stock price--fell 75%.\1297\ Another study
presented evidence from five tick-constrained ETPs that underwent a
five-for-one reverse split; the study found
[[Page 81716]]
that trading costs for these ETPs declined substantially.\1298\ These
studies are consistent with results in table 8 above--when the tick
constraint is relaxed, the quoted spread becomes smaller relative to
the share price, thereby reducing transaction costs.
---------------------------------------------------------------------------
\1295\ See Why GE's basis point spread was four times higher
before its reverse split attached to MEMX Letter at 43-63; see also
The Tick Size Debate Revisited attached to MEMX Letter at 64-70.
\1296\ See supra note 1064 for a numerical example showing the
effect that a reverse split has on transaction costs for a tick-
constrained stock.
\1297\ See Why GE's basis point spread was four times higher
before its reverse split attached to MEMX Letter at 43-63. Chart A
of the study indicates that the reverse split caused the average
quoted spread as a fraction of the share price to decline from
approximately 8 basis points to 2 basis points.
\1298\ See The Tick Size Debate Revisited attached to MEMX
Letter at 64-70. The appendix of the study indicates that the
reverse split caused transaction costs to fall by more than 40% for
tick-constrained ETPs, with some ETPs experiencing cost reductions
of 80%. The study further indicates at 66 that similar results,
``can be achieved by amending the tick regime to simply allow more
granular prices, without the need to change the price of the
security in question.''
---------------------------------------------------------------------------
Some commenters provided explicit specifications of what they
stated is the optimal tick to quoted spread range.\1299\ These
commenters presented evidence and views typically suggesting that 2 to
4 ticks intra-spread may be the optimal range. Given the analysis of
these commenters, analysis presented here, and additional
research,\1300\ the Commission believes that the amendments, which only
reduce the tick size when prevailing quoted spreads fall below $0.015,
are likely to improve market quality for these stocks on average and
are unlikely to lead to detrimental pennying or complexity concerns.
Specifically, while there will likely be less depth at the NBBO and at
each price level, quoted and effective spreads are likely to decline
such that the cost of executing small and medium size trades will
likely decline. Pennying is unlikely to predominate. This is because,
at most, the smaller tick size will result in 3 ticks intra-spread. The
analysis contained in this release, additional research,\1301\ and
commenters tend to agree that pennying is unlikely to be a dominant
effect at 3 ticks intra-spread.\1302\ In fact, 3-ticks intra-spread
falls in the middle of the 2 to 4 ticks intra-spread suggested as
potentially optimal by many commenters.\1303\
---------------------------------------------------------------------------
\1299\ Nasdaq Letter I at 8 provides some empirical analysis
suggesting that 2 to 3 or maybe 2 to 4 ticks intra-spread is
optimal. Pragma Letter at 1 uses data from stock splits to suggest
that 1.5 to 4 ticks is optimal and that more than 4 is too many. RBC
Letter at 3 cites research that 2 ticks intra-spread is optimal.
CCMR Letter at 23 cites academic research suggesting 2 ticks intra-
spread may be optimal. XTX Markets Letter at 4 suggests 2 to 4 ticks
intra-spread as optimal. MMI Letter at 5 suggests 3-9 as optimal.
IEX Letter I at 14 states for 2 to 4 ticks intra-spread. Budish
Letter at 5 indicating that two or fewer is too few and suggesting
that 2 to 4 may be reasonable. Harris Letter at 7 suggesting 2 ticks
intra-spread.
\1300\ See Barardehi et al., supra note 231.
\1301\ Id.
\1302\ One commenter stated that at 4 or more ticks intra-spread
hidden orders become more prevalent which can harm market quality.
See IEX Letter I at 12. However, the smaller tick size implemented
by the Rule will not result in 4 or more ticks for the associated
stocks and so, even if true, the effect suggested by the commenter
is unlikely to play a large role in market quality because the rule
will not result in 4 or more ticks intra-spread for stocks receiving
the $0.005 tick size.
\1303\ Other commenters expressing support for a $0.005 tick for
stocks with narrow quoted spreads include Schwab Letter II at 6, STA
Letter at 7, XTX Letter at 4.
---------------------------------------------------------------------------
The narrower quoted spreads from a smaller tick size may result in
fewer opportunities for price improvement by retail wholesalers, which
may cause execution quality for wholesalers as measured by price
improvement statistics to appear worse. However, the prices that retail
investors receive for trades in stocks receiving the lower tick size is
likely to improve overall relative to the baseline due to a narrower
quoted spread, even though the portion of their price labeled price
improvement may decline.
One commenter suggested generally that the Commission use
simulations to determine the optimal tick size without providing
details about how such a simulation could be structured.\1304\ However,
it is unclear how such simulations would be structured, and the
Commission is unaware of existing simulations or of existing frameworks
for simulations studying the effect of tick sizes on market quality.
Challenges associated with simulations include model accuracy and
complexity, data quality, computational limitations, uncertainty and
sensitivity, validation and verification, scalability, and the
challenges associated with applying simulations to human behavior which
is inherently unpredictable, and the challenges associated with
modeling dynamic and evolving systems like financial markets.
---------------------------------------------------------------------------
\1304\ See Mitre Corp. Letter at 5.
---------------------------------------------------------------------------
One commenter, while not opposing a half-cent tick for some stocks,
stated that volume on inverted exchanges implied that there may not be
a demand to trade with significantly narrower tick sizes.\1305\
Inverted venues, which as discussed in section VII.C.2.b, offer a
rebate to liquidity demanders and charge a fee to liquidity providers
effectively allow market participants to price orders within the tick
by the amount of the rebate to liquidity demanders. The commenter
states that if there were significant demand to trade within the tick
size, then we would expect to see inverted exchanges capture
significant market share among truly tick-constrained stocks.\1306\
---------------------------------------------------------------------------
\1305\ See IEX Letter I at 14.
\1306\ Id.
---------------------------------------------------------------------------
This argument, however, does not take into account the fact that
inverted exchanges are less likely to be at the NBBO than maker-taker
exchanges and thus the risk of an order not being able to be fulfilled
on an inverted exchange is higher.\1307\ If the inverted exchange
cannot fill an order it would generally re-route it or cancel it,
depending on the terms of the order. Both options are costly as re-
routing orders usually involves a re-routing fee charged by the routing
exchange and an access fee charged by the receiving exchange, which
would make the trade more expensive to transact relative to just
sending it first to a high volume maker-taker exchange. The transfer
would also take a small amount of time which could increase adverse
selection risk for the re-routed order. Failing to execute the order is
also costly as it could expose the trader to increased costs if the
market moves against the trader in the time it takes to submit a new
order. Consequently, market share on inverted exchanges may be low for
reasons other than the stock being tick-constrained. For these reasons,
low volume on inverted exchanges is not sufficient to identify demand
to trade within the quoted spread.
---------------------------------------------------------------------------
\1307\ Id.
---------------------------------------------------------------------------
iii. Alternative Definitions of Tick-Constrained
Some commenters supported using TWAQS to determine which stocks are
tick-constrained.\1308\ Other commenters stated that other criteria, in
addition to or in place of the quoted spread was needed to identify
truly tick-constrained securities.\1309\ One commenter stated that
adding additional criteria would decrease complexity as it would limit
the number of stocks receiving a smaller tick size, and would prevent
stocks from bouncing back and forth between tick regimes.\1310\ The
Commission agrees with commenters that the quoted spread is sufficient
to determine whether a stock is tick-constrained as a matter of
principle but acknowledges that not all stocks receiving the lower tick
size will react in the same manner to the tick size. The Commission has
considered commenters' views regarding variables other than quoted
spread, and has conducted additional, supplemental analysis, presented
in this section.
---------------------------------------------------------------------------
\1308\ See, e.g., Budish Letter at 4 and Harris Letter at 7.
\1309\ See, e.g., T. Rowe Price Letter at 4, NYSE Letter I at 3,
BlackRock Letter at 5, Citadel Letter I at 30. Optiver Letter at 1,
2. See also infra section VII.F.1.a for additional analysis of these
alternatives.
\1310\ See Citadel Letter II at 4. The Commission considers the
effect of stocks moving between tick size regimes in section
VII.D.1.d, infra.
---------------------------------------------------------------------------
Commenters' suggestions for additional variables fell into three
broad categories. The most common type of suggestion involved a depth
measure, with commenters questioning whether a
[[Page 81717]]
stock with a narrow-quoted spread was actually tick-constrained if the
sizes of the quotes were small.\1311\ Others suggested metrics based on
trading volume.\1312\ Another commenter suggested price as a measure of
tick constraint.\1313\
---------------------------------------------------------------------------
\1311\ See, e.g., Invesco Letter at 3 suggesting that we
consider whether a stock had multiple bids and offers at the NBBO.
BlackRock Letter at 5 suggesting we consider Average Quoted Size.
Citadel Letter I at 30 suggesting we consider average size at the
NBBO/Daily volume. Cboe, State Street, et al. Letter at 2 and
BlackRock Letter at 5 suggesting we consider quote size at the NBBO
to average trade size, NYSE Letter I at 3 suggesting a quote
stability measure of how long it takes the NBBO to return to pre-
trade levels after a trade takes liquidity at the NBBO. T. Rowe
Price Letter at 4 suggesting that depth could be measured by how
many exchanges are currently quoting the NBBO.
\1312\ See, e.g., BlackRock Letter at 5, Citadel Letter I at 30
(both suggesting trading volume as a measure of tick-constrained)
and T. Rowe Price Letter at 4 (suggesting a turnover-based measure
of tick-constrained).
\1313\ See BlackRock Letter at 5.
---------------------------------------------------------------------------
Some commenters stated that even stocks with the narrow quoted
spread may not benefit from a smaller tick unless other criteria were
met.\1314\ For example, commenters stated that reducing the tick size
for stocks that have narrow quoted spreads, but low depth, could lead
to pennying which could harm market quality.\1315\ Another commenter
stated that if there is insufficient depth, a narrower tick size could
harm market quality by fragmenting liquidity over multiple price
levels--even if quoted spreads are tight.\1316\ Another commenter
didn't articulate specific mechanism by which harms that might arise
due to failing to take into account other criteria, but encouraged the
Commission to take a ``pragmatic approach'' starting with a narrow set
of ``truly tick-constrained stocks'' as defined by multiple
criteria.\1317\
---------------------------------------------------------------------------
\1314\ See, e.g., Citadel Letter I at 5-6 and BlackRock Letter
at 5, Invesco Letter at 3, Virtu Letter II at 17.
\1315\ See, e.g., Citadel Letter I at 5-6 and BlackRock Letter
at 5.
\1316\ See Invesco Letter at 3.
\1317\ See Cboe Letter II at 2, 6.
---------------------------------------------------------------------------
As discussed above, quoted spread reflects supply and demand forces
in the market for liquidity provision. Many of the concerns of
commenters regarding a smaller tick would be expressed by wider quoted
spreads; for example, pennying could potentially reduce incentives for
liquidity provision, raising quoted spreads. This argues in favor of
quoted spread as the main criterion. Furthermore, it is unlikely that a
stock which trades near $0.01 nearly all the time does so at random and
would continue to do so even if unconstrained. It is more likely that a
stock trading with a quoted spread at the minimum pricing increment
nearly all the time does so because the stock would trade with a
narrower spread if that was possible, but the minimum quoting increment
constrains the spread. Finally, quoted spread is correlated with depth,
volume, and price.\1318\ For these reasons, it is likely that stocks
with narrower quoted spreads will also have relatively high depth, and
volume, and have lower prices. However, to fully address commenters'
concerns, the Commission has provided additional, supplemental analysis
that examines whether there is meaningful variation of the effect of a
reduction of tick size within those stocks that have lower quoted
spreads.
---------------------------------------------------------------------------
\1318\ Using WRDS Intra Day Indicators Data for all stocks
priced over $5.00 without sym_suffix, the correlation between daily
log quoted spread, log share volume, log price, and log average
depth at the NBBO is -.39, .35, and -.49 respectively. See also
Harris Letter at 7.
---------------------------------------------------------------------------
Specifically, the analysis presented in table 9 explores the
question of whether stocks with similarly narrow quoted spreads respond
differently to the TSP tick size change when conditioning on other
factors suggested by commenters, such as depth, volume, or price. In
sum, our analysis does not find evidence supporting the hypothesis that
a narrower tick size is likely to harm stocks when considering criteria
such as depth, volume, or price in addition to the quoted spread.
Instead, our analysis suggests that across most market quality metrics
stocks with high and low price, volume, or depth respond in a similar
magnitude and direction to the same tick size change. For some market
quality metrics, one subset of stocks may not have a statistically
significant response to the tick size change. However, in no cases do
we observe that stocks with high or low price, volume, or depth respond
in opposite directions with magnitudes that are statistically
significant. Thus, we find no evidence of a benefit from adding
criteria to the tick size determination that would compensate for the
considerably greater complexity that such an addition would cause.
The analysis presented in table 9 builds on the TSP analysis
considered above, with a few modifications.\1319\ First, for brevity,
we only report findings for bin 1 and bin 2 stocks--i.e., those stocks
with narrower quoted spreads--because the final Rule will similarly
only affect stocks with less than 2 ticks intra-spread.\1320\ We also
only present quantile regressions. We obtain daily regular hours
trading volume, value weighted average price, and depth at the NBBO
information from WRDS Intraday indicators for all trading days in May
and June 2018. We divide bin 1 and bin 2 stocks into quartiles based on
their average volume, price, or depth at the NBBO. The analysis uses
quantile (median) regressions \1321\ to estimate the following
difference-in-differences regression model separately for the high and
low quartile samples: \1322\
---------------------------------------------------------------------------
\1319\ See supra section VII.D.1.b.ii.
\1320\ The Proposing Release analysis divided stocks into four
bins based on their prevailing quoted spreads prior to the
conclusion of the TSP. The first bin is for stocks with quoted
spreads ($0.00, $0.06). The second bin is for stocks with quoted
spreads in the range ($0.06, $0.09). The third bin is for stocks
that had quoted spreads of ($0.09, $0.15) or approximately 2-3 ticks
intra at a $0.05 tick increment. The fourth bin is for stocks with
quoted spreads greater than $0.15.
\1321\ The primary advantage of quantile regressions is that
they are less sensitive to outliers that can affect mean inference
in OLS. Thus, median regressions provide additional robustness to
the analysis and ensure that results are not driven by outliers.
\1322\ In this equation the variable Y denotes the response
variable of interest such as quoted spread and depth. The subscripts
j and t serve to index stocks and days respectively.
[alpha]0, [alpha]p, [alpha]e, and
[beta] are coefficients (to be estimated), and [micro]j,t
is the error term. Pilotj is an indicator variable that
equals 1 if stock j was in the treatment group, or 0 if stock j was
in the control group. Eventt is an indicator variable
which is equal to 1 if the day t was post the treatment event and
equals 0 otherwise. Table 8 reports the difference-in-differences
estimator of [beta] for a different response variable Y across the
different quoted spread bins.
Yj,t = a0 + aPPilotj + aEEventt + b(PilotjxEventt) + mj,t
By checking to see if market quality measures are affected
differently for stocks with similarly narrow quoted spreads, but which
have different depth, or price, or volume profiles, the analysis can
show whether the effects of reducing the tick size depend on these
additional characteristics in a manner that a quoted spread measure
misses. The results are presented in table 9. Panel A presents the
analysis that considers the effect of trading volume on a stock's
response to a reduction in the tick size. The analysis sorts and
subdivides the treated and control stocks separately by volume into
quartiles, and it then performs regressions on the top and bottom
quartile of stocks separately. Columns one and two present the results
for stocks with TWAQS less than or equal to $0.011, whereas Columns
three and four present the results for stocks with TWAQS greater than
$0.011 but less than or equal to $0.020. Columns one and three present
the results for stocks in the bottom quartile of trading volume while
columns two and four present the results for stocks in the highest
quartile of trading volume. Panel B presents the
[[Page 81718]]
same analysis as panel A but where stocks are sorted by price, and
panel C does the same for depth at the NBBO. We consider the
differential effect of the TSP on depth, quoted spreads, effective
spreads, cancel-to-trade ratio, share of odd-lot volume, and realized
spread.
BILLING CODE 8011-01-P
[GRAPHIC] [TIFF OMITTED] TR08OC24.005
[[Page 81719]]
[GRAPHIC] [TIFF OMITTED] TR08OC24.006
[[Page 81720]]
[GRAPHIC] [TIFF OMITTED] TR08OC24.007
[[Page 81721]]
[GRAPHIC] [TIFF OMITTED] TR08OC24.008
BILLING CODE 8011-01-C
The results show that in most cases that there is relatively little
disagreement in terms of how low and high characteristic stocks respond
to the TSP. In panel A, which presents results sorted based on trading
volume, 10 out of the 20 pairs of regressions agree on both sign and
statistical significance. A pair refers to the high and low quartile
group for the same market quality outcome and the same quoted spread
group--e.g., columns one and two are a pair, and three and four are a
pair. This means that in 10 out of the 20 regression pairs run, high
and low trading volume stocks had market quality responses to the
change in the tick size that were in the same direction and were
statistically different from zero. A further 8 out of the 20 pairs
agree on sign, but not statistical magnitude. This means that while the
point estimators from the regressions indicate that both high and low
volume stocks had market quality responses in the same direction,
either one or both groups experienced an effect that could not be
measured to be statistically different from zero.\1323\ Lastly, only 2
out of 20 pairs disagree on sign, but there are no cases where both
effects reject the hypothesis of no effect. This means that while the
coefficients for the effect of the reduction in the tick size on high
and low volume stocks disagree in the direction of the effect, in at
least one case the effect is not statistically distinguishable from
zero.
---------------------------------------------------------------------------
\1323\ A result that is not statistically significant does not
directly imply that there is no effect. Rather it implies that the
test did not find enough evidence to overturn the hypothesis of no
effect. This could be because (1) there is actually no effect, or
(2) because the test did not have sufficient power to identify an
existing effect.
---------------------------------------------------------------------------
The story is similar for panel B which presents results for price.
11 out of 20 regression pairs agree on sign and statistical
significance. A further 7 out of 20 agree on the sign of the effect
though not necessarily statistical significance while only 2 out of 20
disagree on sign, but again in these cases neither result is
statistically different from zero. Lastly in panel C, which presents
results for the sort based on depth at the NBBO, 14 out of the 20
regression pairs agree on sign and statistical magnitude, 4 out of 20
agree on sign, but not statistical magnitude, and only 2 pairs disagree
on sign, but in these cases neither regression provides a statistically
significant result.
This analysis does not support the idea that failing to include
price, volume, or depth-based criteria when determining which stocks
receive a smaller tick size is likely to produce significant harm. In
most cases, the analysis suggests that market quality is likely to
respond similarly to a reduction in the tick size for stocks that are
both high or low volume, price, or depth. It is possible that in some
cases, some of the positive anticipated effects discussed above may be
mitigated for certain subgroups of stocks. We fail to find evidence
that certain subgroups of stocks will be significantly harmed by the
amendments when only considering the time weighted quoted spread when
assigning tick sizes.
c. Effect on Message Traffic and Market Data
One commenter stated that, since ``reducing the minimum pricing
increments for quotations'' will result in ``significantly more
quotation
[[Page 81722]]
increments within a penny-wide spread for a given security,'' these
amendments will ``significantly increase the capacity requirements for
market participants to process these additional quotation messages.''
\1324\ Specifically the commenters stated that the additional message
traffic could impact CAT costs which they state was not analyzed by the
Commission.\1325\
---------------------------------------------------------------------------
\1324\ See SIFMA Letter II at 21, stating that it is concerned
about the impact of the amendments on ``consolidated market data,
capacity costs and CAT data requirements.''
\1325\ See SIFMA Letter II at 21, 43; see also Citadel Letter II
at 4-5 and FIF Letter at 10-12.
---------------------------------------------------------------------------
Commenters also stated that increased message traffic can increase
market participants' technological needs in terms of computing and
processing requirements.\1326\ Some commenters stated that more message
traffic could slow down markets as it would take more time for market
participants to process the increased data, this could in turn disrupt
trading strategies.\1327\ Another commenter stated that increased
message traffic would lead to increased technology costs in terms of
server, processor, and storage requirements needed to deal with
increased message traffic.\1328\ One commenter stated that significant
increases in message traffic can cause technological difficulties for
the exchanges.\1329\ Another commenter stated that increased message
traffic would give an advantage to sophisticated algorithmic traders
who have a greater capacity to manage the cost of the data and could
better apply the data itself.\1330\
---------------------------------------------------------------------------
\1326\ See, e.g., BlackRock Letter at 5, GTS Letter at 5, FIF
Letter at 1, 6-8, RBC Letter at 4, IBKR Letter at 5, Fidelity Letter
at 17.
\1327\ See Tradeweb Letter at 2.
\1328\ See STA Letter at 6.
\1329\ See Budish Letter at 5.
\1330\ See, e.g., Goldman Sachs Letter at 8 and RBC Letter at 4.
---------------------------------------------------------------------------
The Commission recognizes the potential for these costs articulated
by the commenters, and considers the information provided. However, the
Commission expects these effects, including the effect on CAT costs, to
be mild. This conclusion stems from experience in options markets,
which, as discussed above, experience considerably more message traffic
than equity markets without significant adverse effects.\1331\ It also
stems from research provided by commenters that gives a framework for
estimating the increase in message traffic and the costs of such an
increase which will now be discussed suggesting an estimated overall
increase in technology costs to market participants of approximately
1%, and an estimated increase in CAT costs of approximately $4.1
million.
---------------------------------------------------------------------------
\1331\ See discussion surrounding supra notes 239 to 250.
---------------------------------------------------------------------------
The amendments add one additional tick intra-spread for some
stocks. The Commission expects the increase in message traffic from
this increase to be mild. One commenter referenced research performed
by one of the exchanges suggesting that each additional tick intra-
spread adds 20% to message traffic.\1332\ This research suggests
message traffic could increase 20% for stocks receiving the smaller
tick size. The Commission expects 74% of share volume to receive the
smaller tick, and the remaining share volume to be unchanged by the
amendments. Multiplying 74% by 20% suggests that, based on the
information provided by commenter, it is reasonable to believe total
message traffic may increase by about 15% due to the amendments.
---------------------------------------------------------------------------
\1332\ See FIF Letter at 7 referencing Phil Mackintosh, More
Ticks, More Messages (Oct. 27, 2022), available at https://www.nasdaq.com/articles/more-ticks-more-messages. The research
provided by the exchange provided three data points, the baseline
case with no increase in ticks, a 1:5 increase and a 1:10 increase
which are associated with 0%,100% and 500% increase in message
traffic. If message traffic is the variable on the vertical axis and
the tick size change the horizontal axis, then all three of these
data points fall on the line . Substituting in the number 1, to
indicate the number of ticks added intra-spread yields an increase
in message traffic of 1.2, or a 20% increase.
---------------------------------------------------------------------------
Accordingly, the costs associated with this increased message
traffic on most market participants are expected to be relatively
small. One commenter estimated that the proposed rule would lead to an
increase in total infrastructure costs of 10% due to increased message
traffic.\1333\ Using the above methodology we estimate that if the
proposal would have led to an increase in message traffic related costs
of 10%, as suggested by the commenter, and if the costs of message
traffic are proportional to the increase in message traffic, then by
this reasoning, the adopted amendments, which create fewer ticks, will
be associated with an increased infrastructure costs of approximately
1%.\1334\ Additionally, another exchange, commenting on the Proposing
Release which had more and smaller ticks than the adopted amendments,
estimated that the increase in message traffic due to the proposal
would be relatively small compared to historical variation.\1335\
---------------------------------------------------------------------------
\1333\ See Fidelity Letter at 17.
\1334\ The proposal would have added 9 ticks for the stocks
receiving the $0.001 tick size, 4 ticks for the stocks receiving the
$0.002 tick size, and 1 additional tick for the stocks receiving the
0.005 tick. Using the methodology provided by the commenters
suggesting that each additional tick would increase message traffic
20% relative to the baseline suggests that stocks receiving the
$0.001 tick would have an estimated message traffic increase of
180%, for stocks receiving the $0.002 tick the estimated increase in
message traffic would be 80%, and for stocks receiving that $0.005
tick the increase in message traffic would be 20%. Using the same
methodology as used to create table 7 to estimate the amount of
trading volume that would have been associated with each of the
proposed tick sizes based on 2023 data suggests that approximately
50% of trading volume would have received the $0.001 tick,
approximately 25% of trading volume would have been associated with
the $0.002. Combining these figures suggests that message traffic
associated with the proposal would have increased approximately
125%. Thus, if a 125% increase in message traffic is expected to be
associated with a 10% increase in infrastructure costs, then
assuming that costs are proportional to message traffic, then a 15%
increase in message traffic will be associated with just over a 1%
increase in infrastructure costs associated with the amended Rule.
\1335\ See NYSE Letter II at 10-11.
---------------------------------------------------------------------------
Commenters asked the Commission to estimate the effect of the tick
size reduction on CAT costs via the mechanism of increased message
traffic.\1336\ The fourth quarter 2023 CAT cloud hosting services costs
were $30.47 million,\1337\ or $121.88 million if annualized. Further, a
recent order approving an amendment to the NMS plan governing CAT
stated that equity message traffic accounted for 23% of total CAT
message traffic (with options traffic accounting for the
remainder).\1338\ Thus, estimated CAT costs associated with equity
message traffic equal $121.88 million * 23% = $28.03 million. Under the
assumption that CAT costs are linear in message traffic,\1339\ an
[[Page 81723]]
increase in equity message traffic of 15% would correspond to an
increase in CAT costs according to the following formula: $28.03
million * 15% (expected increase in message traffic) [ap]$4.1 million.
Based on this data, we estimate CAT costs would increase by an
estimated $4.1 million per year due to increased equity message traffic
associated with the smaller tick size. The CAT NMS Plan requires
participants and industry members to fund the CAT. The funding model
for the CAT NMS Plan allocates the costs to fund CAT among participants
and the members of a national securities exchange or a member of a
national securities association.\1340\
---------------------------------------------------------------------------
\1336\ See FIF Letter at 10-12, SIFMA Letter II at 43.
\1337\ See Consolidated Audit Trail, LLC, 2023 Financial and
Operating Budget, available at https://www.catnmsplan.com/sites/default/files/2024-01/01.17.24-CAT-Q4-2023-Budget-vs-Actual.pdf.
\1338\ See SEC, Joint Industry Plan; Order Approving an
Amendment to the National Market System Plan Governing the
Consolidated Audit Trail; Notice (Joint Industry Plan), Securities
Exchange Act Release No. 98290 (Sep. 6, 2023), 88 FR 62628 (Sep. 12,
2023) at 62680 & n.1077.
\1339\ The relation between total CAT cost and volume is complex
and driven by both volume and the complexity of processing the
additional information and may not always be linear. See Letter from
Brandon Becker, Chair, CAT NMS Plan Operating Comm., to Vanessa
Countryman, Sec'y, SEC at 3 n.11 (Mar. 27, 2024), available at
https://catnmsplan.com/sites/default/files/2024-03/03.27.24-Proposed-CAT-NMS-Plan-Amendment-Cost-Savings-Amendment.pdf. However,
one commenter speaking generically about server and technology cost
associated with message traffic stated that these costs were
``proportional to the increase in message traffic.'' See FIF Letter
at 9. Consequently, there is uncertainty regarding the exact
relation between CAT costs and equity message traffic. However, to
facilitate estimates of the effect of the amendments on CAT costs,
the Commission makes the assumption that CAT costs are linear in
message traffic--knowing that there is uncertainty regarding the
exact relation. However, without some assumption about the relation
between CAT costs and message traffic, no inference could be made.
To the extent that CAT costs are convex or concave--rather than
linear--in message traffic then the actual costs associated with the
amendments would be higher or lower that what is estimated here.
\1340\ See Joint Industry Plan, supra note 1338.
---------------------------------------------------------------------------
Commenters stated that a smaller tick size will fragment liquidity
in the order book and reduce the displayed liquidity at the NBBO, which
in turn reduces the information about liquidity available in the market
for market participants who do not receive depth of book information
from proprietary data feeds.\1341\ These commenters stated that this
could increase reliance on and the subsequent cost of proprietary data
products due to the increased need for the data and simply due to
additional data points.
---------------------------------------------------------------------------
\1341\ See Citadel Letter I at 8, Virtu Letter II at 8, 10,
Citigroup Letter at 4, SIFMA Letter II at 41-42, and Lewis Letter
attached to Virtu Letter II at p 35. See also Lewis Letter attached
to Virtu Letter II at 33 (pointing out that the need to manage
increased message traffic could lead proprietary data feed to become
more expensive independent of market participant reliance such data
products). SRO fees are subject to the rule filing process under
Exchange Act section 19(b) and Rule 19b-4. 15 U.S.C. 78s(b); 17 CFR
240.19b-4.
---------------------------------------------------------------------------
The Commission acknowledges that, in addition to lowering
transaction costs, the amendments will likely also spread depth across
more price points in the limit order book and reduce depth at the NBBO
in some stocks with a 0.5 cent tick, which could increase the
complexity and cost associated with sourcing liquidity for larger
orders.\1342\ However, the inclusion of odd-lot information in
consolidated market data is anticipated to help to mitigate this
effect, and the eventual inclusion of depth of book information in
consolidated market data due to the implementation of the MDI Rules
should also help mitigate this effect.\1343\ Fragmentation of liquidity
in the limit order book will increase the precision and usefulness of
information about depth of book quotes and odd-lot quotes inside the
NBBO for some stocks with a 0.5 cent tick because orders in these
stocks will be displayed at half-penny increments rather than at penny
increments, which provides more precise information.\1344\ This will
increase demand for depth of book data, which could cause more market
participants that currently only subscribe to SIP data to purchase
exchange proprietary depth of book feeds. It could also cause some
market participants that only subscribe to SIP data and who previously
would not have purchased MDI depth of book data or odd-lot information
to decide to purchase one or both of these MDI data elements when they
become available. The Commission acknowledges that the amendments could
increase the cost associated with producing exchange proprietary data
feeds and MDI data, but the cost increases may not be
significant.\1345\ Any changes in the prices of exchange proprietary
data feeds would be subject to the SRO fee filing process.\1346\
---------------------------------------------------------------------------
\1342\ See supra section VII.D.1.b.i (discussing the smaller
tick size fragmenting depth across more price levels).
\1343\ After the MDI Rules are implemented, consolidated market
data is expected to contain depth information for five price levels
beyond the NBBO, which will help mitigate the effects of a reduction
in displayed depth at the NBBO from a reduction in tick size. See
MDI Adopting Release, supra note 10, at 18625-30 and 18755-59 for
further discussions on the content and effects of MDI depth of book
data.
\1344\ This could increase the usefulness of both exchange
proprietary depth of book feeds and MDI odd-lot information and
depth of book data relative to SIP data for stocks with a 0.5 cent
tick. However, as discussed below, the usefulness of MDI depth of
book data in comparison to exchange proprietary depth of book data
may decrease for these stocks.
\1345\ See supra note 1334 and accompanying text (estimating
increased infrastructure costs of approximately 1% due to an
increase in message traffic).
\1346\ SRO fees are subject to the rule filing process under
Section 19(b) and Rule 19b-4. 15 U.S.C. 78s(b); 17 CFR 240.19b-4.
---------------------------------------------------------------------------
After the MDI Rules are fully implemented, commenters stated that
the amendments could decrease the benefits of MDI data, which will
include depth of book data for prices with quotes that are five levels
outside of the NBBO.\1347\ The tick reduction will likely alter the
information in MDI data for some stocks with a 0.5 cent tick, making
MDI depth of book data less informative while increasing the
informativeness of MDI odd-lot information. The tick reduction may
generally result in MDI depth of book data showing information on fewer
shares in the order book for some stocks with a 0.5 cent tick. More
specifically, it may no longer include some information on orders that
are $0.03 to $0.05 away from the NBBO in some stocks with a 0.5 cent
tick.\1348\ However, the effects of the loss of this information may
not be significant because the MDI depth of book data would still show
five round lot price levels of depth and many market participants may
not need more than five levels of depth of book information.\1349\
Additionally, for stocks with a 0.5 cent tick, information about depth
within $0.025 of the NBBO may be more valuable than information about
depth that is deeper in the book (e.g., between $0.03 and $0.05 of the
NBBO), because only a relatively small portion of marketable orders
execute at prices outside the NBBO.\1350\ Further, for stocks that have
the $0.005 tick, MDI odd-lot information about odd-lots inside the NBBO
will likely be more valuable because it will show these odd-lots on a
finer grid--as long as the NBBO spread is $0.01 or more, the tick
reduction allows investors using MDI NMS data to see odd-lots at finer
half-penny increments. For stocks that retain the one cent tick size,
the tick size amendments are unlikely to affect the informativeness of
MDI depth of book data or odd-lot information because the amendments
are unlikely to change the trading environment, including odd-lot
quotes inside the NBBO and depth in the limit order book.\1351\
---------------------------------------------------------------------------
\1347\ SIFMA Letter II at 41-42 stated that the Rule, by making
NMS data under MDI less competitive with proprietary feeds, will
offset some of the benefits of the MDI Rules.
\1348\ The MDI NMS data includes information on five round lot
price levels outside the NBBO. The tick reduction implies that the
levels may be $0.005 apart from each other; without the tick
reduction, the levels would be at least $0.01 apart. Therefore, the
tick reduction may result in five levels that cover orders that are
$0.025 outside the NBBO, whereas without the tick reduction the five
levels would cover orders that are $0.05 outside the NBBO. It is
possible that MDI NMS data will include depth that is greater than
$0.025 outside the NBBO for stocks receiving a tick reduction--this
is because each displayed level is required to comprise at least one
round lot worth of shares, and odd-lot orders will be aggregated
with other orders to form a level with a sufficient number of shares
(with the displayed price equaling the least competitive price among
the aggregated orders). That is, if there is not enough liquidity at
a given price point within $0.025 of the NBBO, then a price point
further outside the NBBO will be displayed.
\1349\ See MDI Adopting Release, supra note 10, at 18625 nn.387,
388 (discussing levels of depth used in order routing and analysis
finding that a significant percentage of the total notional value of
all depth of book quotations for both liquid and illiquid stocks
falls within the first five price levels).
\1350\ See MDI Adopting Release, supra note 10, at 18731
(discussing analysis showing that only a small percentage of orders
execute outside the NBBO).
\1351\ See supra section VII.D.1.
---------------------------------------------------------------------------
Commenters stated that reducing the tick size would make MDI data
less competitive with exchange proprietary data feeds.\1352\ The tick
reduction may
[[Page 81724]]
cause some market participants that currently utilize exchange
proprietary depth of book feeds, and would have replaced them with MDI
depth of book data and odd-lot information if it was available at a
lower cost,\1353\ to no longer substitute MDI depth of book data and
odd-lot information for proprietary feeds.\1354\ Proprietary feeds will
likely gain certain advantages over MDI depth of book data as a result
of the tick reduction. First, self-aggregators and competing
consolidators relying on MDI depth of book data may not be able to
provide information on depth for orders that are between $0.03 and
$0.05 away from the NBBO for some stocks that have been assigned a
$0.005 tick size because the MDI depth of book data only provides data
on depth at 5 round lot price levels. Second, in cases where there is
insufficient liquidity at a given price level to form a round lot due
to the reduction in the tick spreading liquidity out across more
levels, self-aggregators and competing consolidators relying on MDI
depth of book data will not be able to provide information on depth at
each price point beyond the NBBO, but will rather only have information
on liquidity that is aggregated over multiple price points to compose a
round lot; proprietary feeds, on the other hand, can offer information
on available liquidity at each price point. These two advantages may
lead some market participants that would have substituted MDI depth of
book data for exchange proprietary depth of book data, if it was
available at a lower cost once it was implemented, not to do so due to
the tick reduction. To the extent this occurs, it would result in a
transfer from competing consolidators and the market participants who
would have substituted MDI depth of book data for their proprietary
feeds to exchanges, because these market participants will continue
paying, and exchanges will continue to receive revenue, for their
proprietary data feeds.\1355\ On the other hand, to the extent that the
tick size reduction increases the usefulness and demand for depth of
book data--as discussed above--some market participants that currently
rely on SIP data may be able to satisfy their increased data needs by
purchasing MDI odd-lot information and depth of book data from a
competing consolidator once the MDI rules are implemented.\1356\
---------------------------------------------------------------------------
\1352\ See, e.g., SIFMA Letter II at 41-42 and Citadel Letter I
at 8.
\1353\ In the MDI Adopting Release, the Commission stated that
it believes that the total fees for the equivalent of consolidated
market data are likely to decline because of the MDI Rules, but
recognizes uncertainty about how the effective national market
system plan(s) will set the fees for data content underlying
consolidated market data offerings and how SROs will set the fees
for connectivity necessary to receive the data content underlying
consolidated market data as well as how the competing consolidators
will price their services. See MDI Adopting Release, supra note 10,
at 18772.
\1354\ This would apply to both market participants that would
have self-aggregated the MDI depth of book data and odd-lot
information or purchased it from a competing consolidator. See MDI
Adopting Release, supra note 10, at 18793-95 for discussions on
market participants substituting consolidated market data for
exchange proprietary feeds. However, as discussed above, demand for
MDI depth of book data could increase for market participants that
currently rely on SIP data.
\1355\ See MDI Adopting Release, supra note 10, at 18793-95 for
discussions on market participants substituting consolidated market
data for exchange proprietary feeds and related transfers.
\1356\ See infra section VII.D.4.b for additional discussions of
why low latency traders may not substitute odd-lot information from
the exclusive SIPs for exchange proprietary feeds.
---------------------------------------------------------------------------
d. Analysis of the Length of Evaluation and Operative Periods
The adopted amendments specify three-time periods that govern tick
assignment.\1357\ First are the periods that a new tick size is
operative. The adopted rules update the tick size twice a year, so they
set this period at six months. Second, the rules set the ticks based on
two backward-looking three-month Evaluation Periods over which NMS
stocks' TWAQS is measured; if the TWAQS is at or below $0.015 during
this Evaluation Period, then the stock will be assigned a half-penny
tick. Otherwise, it will receive a penny tick. Finally, there is a one-
month gap between the Evaluation Period and the initiation of the
subsequent tick size. The January through March Evaluation Period will
determine the tick for the May through October operative period.
Likewise, the July through September Evaluation Period will determine
the tick for November through April of the subsequent calendar year.
---------------------------------------------------------------------------
\1357\ See 17 CFR 242.612(a)(1), defining the evaluation period
to assign a tick size as ``(i) the three months from January through
March of a calendar year and (ii) the three months from July through
September of a calendar year during which the Time Weighted Average
Quoted Spread of an NMS stock shall be measured by the primary
listing exchange to determine the minimum pricing increment for each
NMS stock,'' and 17 CFR 242.612(b)(1), and specifying the pricing
increment will be operative ``(i) the first business day of May for
the Evaluation Period from January through March and continue
through the last business day of October of the calendar year, and
(ii) the first business day of November for the Evaluation Period
from July through September and continue through the last business
day of April of the next calendar year.''
---------------------------------------------------------------------------
Commenters highlighted a tradeoff with respect to the proposed
backward-looking evaluation period. On one hand, a short period ``uses
the period immediately closest in time to measure securities' behavior
to ensure that tick-constrained names are selected using the most
recent and relevant basis . . .'' \1358\ On the other hand, a short
period may place too much emphasis on idiosyncratic and
unrepresentative events.\1359\
---------------------------------------------------------------------------
\1358\ See NYSE Letter I at 4.
\1359\ See FIA PTG Letter II at 2, Optiver Letter at 2, and MMI
Letter at 6.
---------------------------------------------------------------------------
Commenters also highlighted a tradeoff regarding the proposed
length of the operative period. Frequent updating will allow tick
assignment to more quickly react to changes in the ``market environment
and individual stock behavior . . .''; \1360\ quicker updating would
thereby ``reduce how often and for how long a stock's tick size stays
outside the optimal range.'' \1361\ However, more frequent updating may
impose costs in terms of adjustments to algorithms, operations, trading
models and systems,\1362\ customer complaints,\1363\ and tick size
oscillation.\1364\
---------------------------------------------------------------------------
\1360\ See NYSE Letter I at 4.
\1361\ See Pragma Letter at 6.
\1362\ See FIA PTG Letter II at 2, UBS Letter at 13, Citigroup
Letter at 2, BlackRock Letter at 9, Hudson River Letter at 3,
JPMorgan Letter at 4, Morgan Stanley Letter at 3, State Street
Letter at 3, and MMI Letter at 6.
\1363\ See TradeStation Letter at 5.
\1364\ See Mitre Corp. Letter at 5 and MEMX Letter at 16.
---------------------------------------------------------------------------
Commenters also discussed the importance of a gap between the
proposed evaluation period and the beginning of the subsequent tick
size to give industry time to update systems and avoid
disruptions.\1365\ The proposal did not include a gap between the
evaluation period and the subsequent tick size, so the Commission has
evaluated the tradeoffs in response to commenters' concerns.
---------------------------------------------------------------------------
\1365\ See RBC Letter at 3, T. Rowe Price Letter at 4, and IEX
Letter I at 7. T. Rowe Price and IEX both suggested a 1-month gap.
RBC suggested ``an appropriate amount of time to be informed of any
changes in order to implement them, and to minimize errors as much
as possible.''
---------------------------------------------------------------------------
The Commission acknowledges, as it did in the Proposing
Release,\1366\ that quoted spreads are not static from day to day. It
is possible that a stock could have a narrow quoted spread during an
evaluation period, and thus be assigned a $0.005 tick, and then during
the following operative period it could experience points in time where
the quoted spread is much wider.\1367\
[[Page 81725]]
Likewise, a stock could have a wide quoted spread during an evaluation
period and therefore be assigned a $0.01 tick, yet subsequently trade
with a narrower spread such that a $0.005 tick would be beneficial. The
extent to which these outcomes occur will depend on the length of the
evaluation period, the operative period, and the lag between the two.
---------------------------------------------------------------------------
\1366\ See Proposing Release, supra note 11, at 80324.
\1367\ Likewise, two stocks with equal average quoted spreads
may not be equally tick-constrained. For example, one stock with a
$0.02 average quoted spread could have a $0.01 quoted spread 40% of
the time while another has a $0.01 quoted spread 10% of the time.
The effect of the Rule on market quality could differ in much the
same way as the effects described in this paragraph. Additionally,
some commenters inquired about how stock splits and reverse splits
would be handled; see SIFMA Letter II at 42 and Virtu Letter II at
18. Stock splits and reverse splits can mechanically affect the bid-
ask quoted spread and are not considered in the evaluation periods,
thus it is possible that a stock could temporarily be misassigned to
a tick size due to a split or reverse split. These effects would be
temporary and would rectify in the next evaluation period. See supra
section III.C.8 for further discussions of how tick sizes are
assigned following a stock split or reverse split.
---------------------------------------------------------------------------
The Commission evaluates the tradeoffs mentioned above by computing
diagnostic statistics for many different combinations of evaluation and
operative periods. The four panels of table 10 each present different
diagnostics: panel A estimates the fraction of aggregate share volume
that is misassigned to a tick of $0.01, panel B estimates the fraction
of aggregate share volume that is misassigned to a tick of $0.005,
panel C estimates the rates of false positives, and panel D estimates
the rates of false negatives. Table 10 is similar to table 10 of the
Proposing Release,\1368\ which estimated in the context of the proposal
the fraction of aggregate volume that would receive a tick reduction
yet trade with too many intra-spread ticks during the subsequent three
months. Table 10 herein extends this analysis by: computing a wider
range of diagnostic statistics, computing the statistics on a rolling
basis for a 4.5-year sample, computing the statistics for a wide
variety of period lengths, and incorporating a one-month lag between
the evaluation period and the implementation of the tick updates as
provided for in the adopted amendments.
---------------------------------------------------------------------------
\1368\ See supra note 11, at 80324.
---------------------------------------------------------------------------
Each diagnostic in table 10 is computed for sixteen combinations of
evaluation and operative periods--four evaluation period lengths, and
four operative period lengths. The evaluation period lengths include:
one month, three months (as suggested in the proposal and finalized in
the adopted amendments), five months, and twelve months.\1369\ The
operative period lengths include: one month, three months (as suggested
in the proposal), six months (as in the adopted amendments), and twelve
months.\1370\ The calculations incorporate a one-month lag between the
evaluation period and the implementation of the updated tick size--
which was not part of the proposal but is part of the adopted
amendments. Further, each diagnostic is computed for every month from
January of 2018 to June of 2022. That is, the Commission simulates the
tick assignment procedure on a rolling basis for every month of the
4.5-year sample and computes the diagnostics at every month.\1371\ Each
diagnostic is therefore computed under each of the sixteen combinations
of periods and for each of the 54 months in the sample, for a total of
864 calculations per diagnostic. The panels in table 10 summarize the
diagnostics over the 54 months and present a single number for each of
the sixteen combinations of periods.
---------------------------------------------------------------------------
\1369\ The evaluation periods in this analysis were chosen to
cover a range of time periods for both the evaluation period and the
operative effective period suggested by commenters who suggested
time horizons ranging from monthly updating to annual updating, see,
e.g., Pragma Letter at 6, BlackRock Letter at 9, FIA PTG at 2, UBS
at 13. Although commenters suggested 6-month evaluation periods, the
analysis here considers five-month evaluation period specifically so
that the evaluation period would not encompass two tick regimes when
combined with a six-month operative period and a one month lagged
implementation. For example, consider the six-month operative period
spanning May to October (which includes a one month lagged
implementation); the next six-month evaluation period would span
April to September, thus including two distinct tick regimes. The
five-month evaluation period would only span May to September, which
is wholly contained in the most recent operative period. An
evaluation period that encompasses two tick regimes may be less
informative of the appropriateness of the current tick assignment;
therefore, table 10 uses a five-month evaluation period instead of a
six-month evaluation period.
\1370\ The operative periods were chosen to cover a range of
intervals, and to fit evenly into a twelve-month calendar year. In
this way, any changes to the tick size would occur in the same
month(s) of each year.
\1371\ For example, suppose the month is April 2018. To compute
the statistics for an evaluation length of three months and an
operative length of six months, the tick size for May 2018 through
August 2018 is set by the symbol's TWAQS from January 2018 through
March 2018. The statistics in the table are determined by symbols'
trading during the May 2018 through August 2018 period (the
operative period). This process is repeated for every month from
January 2018 to June 2022, and the table summarizes results across
all months.
---------------------------------------------------------------------------
The diagnostics computed in table 10 are subject to the following
caveat: they are estimated from historical data in which the access fee
cap was set at 30 mils, and stocks were constrained by the $0.01 tick.
The amendments will reduce the access fee cap to 10 mils for all stocks
and will reduce the tick to $0.005 for stocks that maintain a TWAQS at
or below $0.015 during the evaluation period. These changes are likely
to have two opposing effects on quoted spreads: the reduction in the
access fee cap will put upward pressure on quoted spreads, while the
reduction in the tick will allow quoted spreads to fall below $0.01 for
some NMS stocks.\1372\
---------------------------------------------------------------------------
\1372\ See supra section VII.D.1.b.ii for a discussion of the
effect of the reduction in the tick on quoted spreads, and section
VII.D.2 for a discussion of the effect of the reduction in the
access fee cap on quoted spreads.
---------------------------------------------------------------------------
This caveat implies that the estimates in table 10 will be
systematically different from the same statistics calculated with
realized data--i.e., data after the rule is implemented. However, the
purpose of the analysis in table 10 is to detect patterns in how the
diagnostics vary across combinations of evaluation and operative
periods. These patterns are likely to be robust to the aforementioned
caveat. For example, suppose the reduction in the access fee cap causes
a stock's quoted spread to widen by 20 mils; this effect would occur
whether the Evaluation Period is three months or twelve months, and
whether the operative period is one month or six months, etc.
Therefore, table 10 can still inform the choice of evaluation and
operative period. The subsequent discussion of table 10 will further
highlight when a diagnostic may be over- or under-estimated.
Panel A of table 10 estimates the fraction of aggregate share
volume that is misassigned to a tick size of $0.01. A stock's volume is
misassigned to a penny tick if its average TWAQS is above $0.015 during
the most recent evaluation period, yet trades with a TWAQS below $0.015
in the operative period. This trading volume would have benefited from
a lower tick of $0.005 in those subsequent months and is therefore
considered a false negative. The fraction of aggregate share volume
that is a false negative is reported in panel A. Further, this fraction
is reported for every combination of evaluation period and operative
period.
Looking at the false negatives in panel A, the table demonstrates
that shorter periods tend to reduce the percent of volume that is
misassigned to a $0.01 tick. One can pick any evaluation length shown
in the table (1, 3, 5, or 12). For that evaluation length, the fraction
of aggregate share volume that occurs with a $0.01 tick and maintains a
TWAQS at or below $0.015 is increasing in the operative period. In
other words, more frequent updating provides greater benefits, on
average, for any choice of evaluation period. For example, an
evaluation period of 3 months with a 1-month operative period will
misassign 9.4% of trading volume to a tick of $0.01; if the same
evaluation period is used but the tick is updated annually, then the
fraction of misassigned trading increases to 12.9%. Similarly, one can
[[Page 81726]]
pick any column corresponding to the operative period (1, 3, 6, or 12)
and observe that the prevalence of false negatives is increasing with
the length of the evaluation period. These patterns are consistent with
commenters' views that more recent (and relevant) data tend to increase
the benefits of the amendments.
A tick size can be misassigned in a second way: a stock may be
assigned a tick of $0.005 yet end up trading with an average quoted
spread over $0.015. This trading volume may not fully benefit from the
lower tick and is therefore considered a false positive. If the quoted
spread widens sufficiently, relative to the quote, then the stock could
trade in a range of ticks intra-spread that may harm market
quality.\1373\ The fraction of aggregate volume that is a false
positive is reported in panel B of table 10 for every combination of
evaluation period and operative period. This is analogous to table 10
of the Proposing Release, but with a quoted spread threshold of $0.015
instead of 10 or 15 ticks to align with the adopted amendments.
---------------------------------------------------------------------------
\1373\ The empirical analysis in section VII.D.1.b.ii,
suggesting that a lower tick size benefits tick-constrained stocks,
is an ``on average'' result. While the Commission expects that a
lower tick would on average decrease transaction costs for tick-
constrained stocks, the Commission cannot rule out the possibility
that for some of these stocks, a smaller tick could lead to wider
quoted spreads. For these stocks, if quoted spreads increase to a
sufficient degree, then the stock could be re-assigned a wider tick
after the next evaluation period.
---------------------------------------------------------------------------
Panel B shows that, for the adopted rule choices of a 3-month
Evaluation Period and a 6-month operative period, 3.4% of share volume
will receive a $0.005 tick yet trade in an environment with an TWAQS
over $0.015. It is possible that the reduction in the tick size could
cause a worse trading environment for some of this fraction of trading
volume, compared to what the trading environment could have been had
the stock retained a $0.01 tick. This effect will not be indefinite
because, if a stock's quoted spread remains elevated, then at the end
of the next evaluation period the stock will be assigned a wider tick--
mitigating the negative consequences of having a tick size that is too
narrow relative the quoted spread.
The false positive statistics in panel B of table 10 further show
that less frequent tick updating tends to result in more volume trading
with a smaller tick yet a relatively wide quoted spread. This can be
seen by the increased prevalence of false positives as one moves left-
to-right along a row--for any chosen evaluation period, a longer
operative period results in more volume trading at a low-tick yet
relatively wide quoted spread. This pattern is consistent with
commenters' views that more frequent updating is better at adapting to
changing market trends and stock behavior. Similarly, moving down any
given column shows that increasing the evaluation period tends to
reduce the amount of volume in low-tick stocks with wide quoted
spreads; this pattern is consistent with commenters' views that short
evaluation periods may, in some circumstances, assign stocks to the
$0.005 tick on the basis of transient events, and that these stocks may
not be able to sustain the tick reduction.
The estimates of false positives in panel B of table 10 are likely
to be over-estimates for two reasons. First, the threshold for tick
misassignment is chosen at a quoted spread of $0.015, which corresponds
to three intra-spread ticks. The above analysis on the Tick Size Pilot
indicates that the trading environment does not deteriorate until the
number of intra-spread ticks is well above three. Hence, the chosen
threshold is conservative. Second, and as previously discussed, the
estimates in panel B are constructed from historical data in which
stocks were constrained by the $0.01 tick. Once these amendments are
implemented the stocks assigned a $0.005 tick will be able to trade at
quoted spreads below $0.01; this will have a mechanical effect of
lowering the stocks' TWAQS and thereby keeping more volume under the
$0.015 quoted spread threshold.
By comparing the magnitudes of the false negatives and false
positives in panels A and B of table 10, we can assess the relative
importance of the two ways in which a tick may be misassigned. The
false negatives are generally substantially larger and exhibit greater
variation within the table.\1374\ This implies that the incremental
effect of the choice of period length will be greater for the share of
volume that is misassigned to a $0.01 tick than for volume that is
misassigned to a $0.005 tick. For example, the difference between the
highest and lowest fraction of false negatives is 12.3%--this is the
increase in aggregate share volume at a misassigned $0.01 tick when
moving from 1-month periods to 12-month periods--while the difference
between the highest and lowest fraction of false positives is only 1.8%
of aggregate share volume.\1375\
---------------------------------------------------------------------------
\1374\ If false positives are over-estimated, as discussed in
the previous paragraph, then the relative importance of false
negatives becomes greater still.
\1375\ The greater prevalence of false negatives may be due to
the existing $0.01 tick size effectively censoring the observed
quoted spreads at the penny tick. For symbols near the $0.015
threshold, the prevalence of false positives and negatives should be
approximately equal--i.e., a symbol with a quoted spread of $0.0149
is as likely to cross over the $0.015 threshold (and be a false
positive) as a symbol with a quoted spread of $0.0151 is likely to
cross under the $0.015 threshold (and be a false negative). When we
move away from the threshold, however, the low-tick symbols are more
likely to be constrained by the penny tick--if we see a TWAQS of
$0.012, the true market-clearing quoted spread--i.e.,unconstrained
by the minimum tick--is likely lower than $0.012 because the TWAQS
is censored at $0.01; if we see a TWAQS of $0.018, however, this
censoring is less important. Therefore, the likelihood that a symbol
with a TWAQS of $0.012 crosses over the $0.015 threshold (and is
recorded as a false positive) is lower than the likelihood that a
symbol with a TWAQS of $0.018 crosses under the $0.015 threshold
(and is recorded as a false negative).
---------------------------------------------------------------------------
An advantage of panels A and B of table 10 is that they estimate
the prevalence of false negatives and positives as fractions of
aggregate share volume; these panels therefore show the amount of
aggregate trading that is misassigned to a tick. A disadvantage of
panels A and B is that they do not condition on the aggregate level of
quoted spreads. Some combinations of evaluation and operative periods
may do relatively well when aggregate quoted spreads are high, and some
combinations may do better when aggregate quoted spreads are low.\1376\
---------------------------------------------------------------------------
\1376\ To take an extreme example, suppose every stock's quoted
spread increases by $0.10 immediately after new ticks are assigned.
This implies that no stock should have a tick of $0.005. In this
case, the evaluation and operative periods that assign the least
amount of trading to the $0.005 would do best. Conversely, suppose
that quoted spreads fall by half immediately after new ticks are
assigned; in this case, the periods that assign more trading to the
$0.005 tick would generally do better.
---------------------------------------------------------------------------
Panel C and D of table 10 address the aforementioned disadvantage
of panels A and B by estimating rates of false negatives and positives.
The false negative rate conditions on the amount of low- quoted spread
volume, while the false positive rate conditions on the amount of high-
quoted spread volume.\1377\ Specifically, the rate of false negatives
is measured as the amount of share volume that occurs with a tick of
$0.01 and a TWAQS under $0.015, divided by the total amount of share
volume that occurs with a TWAQS under $0.015. Analogously, the rate of
false positives is measured as the amount of share volume that occurs
with a tick of $0.005 and a TWAQS over $0.015, divided by the total
amount of share volume that
[[Page 81727]]
occurs with a TWAQS over $0.015. In the context of the evaluation
period, a low false negative rate signifies effective identification of
stocks that would benefit from a tick reduction, while a low false
positive rate suggests effective avoidance of assigning a reduced tick
to stocks that would not benefit from it. In contrast to panels A and
B, patterns in panels C and D of table 10 are likely to be more robust
to market-wide changes in quoted spreads.
---------------------------------------------------------------------------
\1377\ In the statistics and medicine literature, the false
positive rate is related to a test's specificity, while a false
negative rate is related to a test's sensitivity.
---------------------------------------------------------------------------
Panel C of table 10 presents results for the false negative rates
across 16 combinations of evaluation periods and operative periods. The
pattern for false negative rates is similar to the fraction of share
volume inappropriately assigned a $0.01 tick in panel A--as the period
lengths shorten, the false negative rates decrease. For example, an
evaluation period of 3 months with monthly updating typically fails to
assign a tick reduction to 12.8% of volume that would benefit from it.
If the same evaluation period is used but the tick is updated annually,
then the rule would fail to assign a tick reduction to 19.3% of volume
that would benefit from it. The pattern holds moving down columns and
moving across rows from shorter to longer periods. This lends support
to commenters' views that more recent data--from shorter evaluation
periods and more frequent updating--tends to do a better job at
assigning a low tick to stocks that will benefit from it in the
operative period.
False positive rates are reported in panel D. The pattern for false
positive rates is consistent with results in panel B on the fraction of
volume inappropriately assigned a $0.005 tick--shorter evaluation
periods and longer operative periods tend to have higher false positive
rates, indicating that more trading is potentially harmed from having
too narrow of a tick. For every column, the highest false positive rate
occurs with an evaluation length of 1 month, and the lowest false
positive rate occurs with an evaluation length of 12 months. This
indicates that short evaluation periods may put more weight on
transient events, as some commenters stated. Similarly, for every row
the false positive rate is highest with an operative period of 12
months, and the rate is lowest with an operative period of 1 month.
This indicates that infrequent updating increases the risk that a stock
is stuck at an inappropriately low tick for many months.
The contrasting patterns in the rates of false positives and
negatives illustrates a tradeoff highlighted by commenters (and
discussed at the beginning of this section). A short evaluation period
uses the most recent and relevant information, which on average reduces
the false negative rate; however, a short evaluation period also places
more emphasis on short-term and volatile events, which raises the false
positive rate.
Table 10--Effect of the Evaluation Period on Inappropriate Tick Assignment
----------------------------------------------------------------------------------------------------------------
Length of operative period in months
-------------------------------------------------------------------
1 (%) 3 (%) 6 (%) 12 (%)
----------------------------------------------------------------------------------------------------------------
Panel A: Fraction of aggregate share volume assigned a $0.01 tick with a subsequent TWAQS below $0.015 (i.e.,
false negatives)
----------------------------------------------------------------------------------------------------------------
Length of evaluation period in months:
1....................................... 8.5 9.4 10.7 12.9
3....................................... 11.4 12.4 13.7 15.8
5....................................... 13.1 14.1 15.3 17.3
12...................................... 17.0 17.8 18.9 20.8
----------------------------------------------------------------------------------------------------------------
Panel B: Fraction of aggregate share volume assigned a $0.005 tick with a subsequent TWAQS above $0.015 (i.e.,
false positives)
----------------------------------------------------------------------------------------------------------------
Length of operative period in months:
1....................................... 3.1 3.3 3.8 4.3
3....................................... 2.6 2.9 3.4 3.8
5....................................... 2.6 2.9 3.2 3.6
12...................................... 2.5 2.7 2.9 3.2
----------------------------------------------------------------------------------------------------------------
Panel C: False negative rates
----------------------------------------------------------------------------------------------------------------
Length of operative period in months:
1....................................... 12.8 14.2 16.2 19.3
3....................................... 17.4 18.8 20.6 23.6
5....................................... 19.9 21.2 23.0 25.9
12...................................... 25.7 26.9 28.6 31.2
----------------------------------------------------------------------------------------------------------------
Panel D: False positive rates
----------------------------------------------------------------------------------------------------------------
Length of operative period in months:
1....................................... 8.5 9.4 10.9 12.6
3....................................... 7.2 8.2 9.6 11.2
5....................................... 7.1 8.1 9.3 10.6
12...................................... 6.8 7.5 8.4 9.5
----------------------------------------------------------------------------------------------------------------
\a\ For every month from January 2018 to June 2022, the Commission simulates the tick assignment procedure under
16 combinations of evaluation and operative period lengths. The evaluation period determines the number of
prior months to use when averaging each stock's quoted spread; a TWAQS of $0.015 or below during the
evaluation period causes the stock to receive a tick of $0.005 during the subsequent tick assignment interval.
The operative period determines the length of each tick assignment.
All the statistics in the tables are computed using data beginning one month after the evaluation period. For
example, suppose the month is April 2018. To compute the statistics for an evaluation length of 3 months and
an operative length of 6 months, the tick size for May 2018 through August 2018 is set by the stock's TWAQS
from January 2018 through March 2018. The statistics in the table are determined by stocks' trading during the
May 2018 through August 2018 period (the operative period). This process is repeated on a rolling basis for
every month from January 2018 to June 2022, and the table summarizes results across all months.
[[Page 81728]]
TWAQS is determined by computing the time weighted quoted spread during regular trading hours as computed by the
WRDS intra-day indicators for every sym_root and sym_suffix combination in the dataset. When calculating a
stock's TWAQS during an evaluation period, the stock's daily TWAQS is averaged across all trading days in the
evaluation period.
When assigning volume to a TWAQS bucket in an operative period, the TWAQS on a given day for a particular stock
is used. That is, if a stock trades with a TWAQS of $0.011 on Monday but the same stock has a TWAQS of $0.016
on Tuesday, then its volume on Monday is assigned to the sub-$0.015 category while its Tuesday volume is
assigned to the over-$0.015 category in the operative period.
The universe of securities in the WRDS intra-day indicators dataset is used.
Panel A computes the fraction of total aggregate share volume that occurs in stocks that would have been
assigned a $0.01 tick yet subsequently trade at a TWAQS of under $0.015 in the operative period. These stocks
would benefit from a $0.005 tick instead of a $0.01 tick.
Panel B computes the fraction of total aggregate share volume that occurs in stocks that would have been
assigned a $0.005 tick yet subsequently trade at a TWAQS of over $0.015 in the operative period. These stocks
may not benefit from the $0.005 tick.
Panel C computes false negative rates. The false negative rate is the fraction of share volume that is assigned
a tick of $0.01 among the share volume that trades with a TWAQS under $0.015 in the operative period.
Panel D computes the false positive rates. The false positive rate is the fraction of share volume that is
assigned a tick of $0.005 among the share volume that trades with a TWAQS above $0.015 in the operative
period.
Table 11 further explores commenters' discussions about the risk of
placing too much emphasis on transient market conditions. In
particular, panels A and B of table 11 computes rates of false
negatives and positives--similar to panels C and D of table 10--but
does so using only the evaluation period that ends with March of 2020.
The one-month evaluation period includes only March of 2020; the three-
month evaluation period covers January to March of 2020; the five-month
evaluation period covers November of 2019 to March of 2020; the twelve-
month evaluation period covers April of 2019 to March of 2020. In this
way, panels A and B of table 11 demonstrate what may happen if an
unusual market event causes an inappropriate tick assignment. These
panels show that a one-month evaluation period performs particularly
poorly when that month is March of 2020. Specifically, the one-month
evaluation period exhibits a substantially higher false negative
rate.\1378\ A three-month evaluation period consistently has the lowest
false negative rate--it has the benefit of recent data without an over-
reliance on short-term fluctuations. Finally, the false positive rates
in panel B approximately double when moving from a six-month operative
period to a twelve-month operative period, indicating that infrequent
tick updating can lead to stocks getting stuck with an inappropriately
low tick for an extended period.
---------------------------------------------------------------------------
\1378\ The false positive rates are generally lower than usual
(e.g., below those in Panel B), likely because quoted spreads
narrowed after March of 2020.
---------------------------------------------------------------------------
Panels A and B of table 11 add to our understanding of the tradeoff
presented by the evaluation period. Table 10 suggests that short
evaluation periods tend to reduce false negatives but increase false
positives. Given that false negatives (vs. false positives) tend to be
more prevalent and vary more across evaluation periods, table 10
suggests that a one-month evaluation period may be best. However, table
11 shows that a one-month evaluation period may substantially increase
false negatives when the evaluation period includes a period of unusual
market stress.
While panels A and B of table 11 includes March of 2020 in the
evaluation period, one commenter also provided, in the context of the
Proposing Release, analysis using March of 2020 in the operative
period. This commenter stated that during times of market stress,
``many symbols would be trading with far more price levels intra-spread
than contemplated under the Proposal, thereby further increasing the
liquidity-related harms . . .'' The commenter further showed that, were
the proposed rule in effect, most symbols receiving a tick reduction
would have experienced over ten intra-spread ticks during March of
2020, and many stocks would have experienced over twenty intra-spread
ticks.\1379\
---------------------------------------------------------------------------
\1379\ See Citadel Letter I at 6-7.
---------------------------------------------------------------------------
The Commission acknowledges that quoted spreads generally widen
during periods of market stress, and this may result in stocks trading
with more intra-spread ticks than is desirable; the rise in intra-
spread ticks may then compound the market stress. Relative to the
proposal, the amendments reduce the severity of such an outcome by
reducing the number of stocks that receive a tick reduction, and by
reducing the size of the tick reduction. To further examine this issue,
the Commission conducts its own analysis with March of 2020 as the
operative period. In particular, the Commission simulates tick
assignment in February of 2020 under the parameters of the amendments,
and then examines the outcome during March of 2020. The simulation is
done for a range of evaluation periods--a one-month evaluation period
uses January of 2020 to determine which stocks receive the $0.005 tick
in March (allowing for a one-month lag), a three-month evaluation
period uses November of 2019 to January 2020 to assign ticks, a five-
month evaluation period uses September of 2019 to January 2020, and a
twelve-month evaluation period uses February of 2019 to January 2020.
The Commission then examines the fraction of aggregate share volume
that would have received a $0.005 tick under the amendments yet traded
with a wide quoted spread during an operative period of March 2020.
Results are presented in panel C of table 11. Each row corresponds
to an evaluation length of 1, 3, 5, or 12 months. Each column
corresponds to a quoted spread threshold of $0.015, $0.02, or $0.05.
The upper-left number--21.4%--indicates that 21.4% of aggregate share
volume in March of 2020 would have occurred with a $0.005 tick and a
TWAQS over $0.015. With a three-month evaluation period, this fraction
halves to 10.6%, further reinforcing the conclusion of panel A that a
one-month evaluation period may do particularly poorly when market
conditions suddenly worsen. The Commission further reiterates that
stocks receiving a $0.005 tick are unlikely to be harmed when trading
at a quoted spread of $0.015, so it is unlikely that 10.6% of share
volume would have been harmed in March of 2020 were the amendments in
place. To further explore this issue, the Commission performs similar
calculations with wider quoted spread thresholds of $0.02 and $0.05.
With a three-month evaluation period, the fraction of aggregate share
trading that receives a $0.005 tick and trades at a quoted spread over
$0.02 is 6.4%; this reduction (from 10.6% trading at a quoted spread
over $0.015) indicates that a substantial proportion of the false
positive trading during March of 2020 is occurring with 3-4 intra-
spread ticks, which is generally in line with commenters' views on the
optimal number of intra-spread ticks.\1380\
---------------------------------------------------------------------------
\1380\ See supra note 1299.
---------------------------------------------------------------------------
[[Page 81729]]
Finally, with a three-month evaluation period, the fraction of
aggregate trading that receives a $0.005 tick and trades at a quoted
spread over $0.05 is 1.2%. Analysis in the Proposing Release and herein
suggests that this 1.2% of volume is at an increased risk of a
reduction in market quality due to having over ten intra-spread
ticks.\1381\
---------------------------------------------------------------------------
\1381\ See supra section VII.D.1.b.ii, particularly figure 2 and
surrounding discussion.
Table 11--Effect of the Evaluation Period on Inappropriate Tick Assignment Around March 2020
----------------------------------------------------------------------------------------------------------------
Length of operative period in months
-------------------------------------------------------------------
1 (%) 3 (%) 6 (%) 12 (%)
----------------------------------------------------------------------------------------------------------------
Panel A: False negative rates using March of 2020 in the evaluation period
----------------------------------------------------------------------------------------------------------------
Length of evaluation period in months:
1....................................... 36.2 35.4 37.4 40.1
3....................................... 21.7 22.5 23.7 23.9
5....................................... 34.7 34.9 36.1 37.4
12...................................... 30.0 31.7 32.0 33.9
----------------------------------------------------------------------------------------------------------------
Panel B: False positive rates using March of 2020 in the evaluation period
----------------------------------------------------------------------------------------------------------------
Length of operative period in months:
1....................................... 1.7 2.6 3.2 8.0
3....................................... 3.0 4.6 6.0 12.0
5....................................... 2.6 4.1 5.4 10.2
12...................................... 3.6 5.0 6.5 11.0
----------------------------------------------------------------------------------------------------------------
Panel C: Fraction of March 2020 share volume with a $0.005 tick and wide quoted spreads
----------------------------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------------------------
Quoted spread threshold
--------------------------------------------------
AQS > $0.015
(%) AQS > $0.02 (%) AQS > $0.05 (%)
----------------------------------------------------------------------------------------------------------------
Length of evaluation period in months:
1........................................................ 21.4 14.4 3.5
3........................................................ 10.6 6.4 1.2
5........................................................ 12.4 7.6 1.5
12....................................................... 14.4 9.2 2.0
----------------------------------------------------------------------------------------------------------------
\a\ The Commission simulates the tick assignment procedure under 16 combinations of evaluation and operative
period lengths. The methodology and data used for this simulation is described in the note to table 10.
In contrast to table 10, which performed the simulation on a rolling basis for every month from January 2018 to
June 2022, this table only simulates tick assignment and outcomes around March of 2020.
Panel A repeats the false positive calculations of table 10 as if the tick is assigned in April 2020 (to ensure
that March 2020 is in the evaluation period). For example, to compute the statistic for an evaluation length
of 3 months and an operative length of 6 months, the tick size for May 2020 through August 2020 is set by the
stock's TWAQS from January 2020 through March 2020. April 2020 is the hypothetical month in which the tick
assignment is calculated and acts as the gap between the evaluation period and the operative period. Panel B
similarly repeats the false negative calculations of table 10 as if April 2020 is the month in which the tick
assignment is calculated.
Panel C instead uses March 2020 as the operative period. For example, to compute the statistic for an evaluation
length of 3 months, the tick size for March 2020 is assigned by the stock's TWAQS from November 2019 to
January 2020, with February of 2020 being the hypothetical month in which the tick assignment is calculated.
Panel C computes the fraction of total aggregate share volume that occurs in stocks assigned to a $0.005 tick,
yet trades at a high TWAQS during March of 2020. The fraction is calculated using a range of TWAQS thresholds
for trading during March of 2020--the columns correspond to trading with a TWAQS over $0.015, $0.02, and
$0.05.
The analysis indicates that a shorter operative period almost
always results in fewer errors. By updating the tick more frequently,
the tick better reflects changing market conditions--this is shown by
increasing rates of both false negatives and positives as the operative
period widens.\1382\ However, commenters stated that more frequent
updates may impose costs in terms of adjustments to algorithms,
operations, systems, and an increase in customer complaints. One
commenter requested that the Commission ``investigate historical rates
of change in the TWAS for a variety of trading symbols. This would
avoid imposing undue costs on market participants that may be caused by
tick size updates that are too frequent but should mitigate the
possibility that tick sizes do not update frequently enough and lead to
worse trading outcomes.'' \1383\
---------------------------------------------------------------------------
\1382\ In panels A-D of table 10 and panels A-B of table 11,
moving left-to-right along any row results in a monotonic increase
in both false positives and false negatives. The only exception to
this pattern is in Panel A of table 11: with an evaluation period of
one-month, the false negative rate falls by 0.8% when the operative
period increases from one to three months.
\1383\ See Mitre Corp. Letter at 5.
---------------------------------------------------------------------------
[[Page 81730]]
The Commission examines the frequency of tick size updates in table
12. This table shows the typical number of tick-changes that occur in a
12-month period under each combination of evaluation and operative
period length.\1384\ Shorter operative periods present a tradeoff, they
better tailor the tick size to current conditions for the stock which
can improve market quality for the stock, but they impose two costs:
first, a shorter operative period implies ticks are updated more
frequently during a year, which increases the number of discrete system
changes that market participants need to make. Second, a shorter
operative period increases the number of tick-changes that stocks
experience during a typical twelve-month period--this is seen in table
12 by the increase in tick changes as one moves right-to-left in any
row. More tick changes may result in result in more extensive changes
to trading algorithms and may increase customer confusion and
complaints. The risk of customer confusion is mitigated by the adoption
of a tick size indicator in the regulatory data. This indicator can be
directly incorporated into trading algorithms helping to automate the
process adjusting trading strategies to different tick sizes for
algorithmic traders.\1385\
---------------------------------------------------------------------------
\1384\ These estimates of tick changes are likely to be over-
estimates because the estimates are constructed from historical data
in which stocks were constrained by the $0.01 tick. Once the adopted
tick amendment is implemented, the stocks assigned a $0.005 tick
will be able to trade at quoted spreads below $0.01; this is
expected to lower the stocks' TWAQS, thereby reducing the
probability that a low-tick stock will experience a tick change by
crossing over the $0.015 TWAQS threshold. It is possible, however,
that the amended rule's reduction in the access fee cap may cause
quoted spreads to widen, though transaction costs are not expected
to go up, and thus shift the distribution of quoted spreads toward
the $0.015 threshold; if more stocks trade near the $0.015
threshold, then there may be more switching as stocks move across
the threshold more frequently.
\1385\ See infra section VII.D.5 for additional discussion of
the expected costs associated with the amendments to Rule 612.
Table 12--Effect of the Evaluation Period on the Median Number of Tick Changes Over a 12-Month Period \a\
----------------------------------------------------------------------------------------------------------------
Length of operative period in months
---------------------------------------------------------------
1 3 6 12
----------------------------------------------------------------------------------------------------------------
Length of evaluation period in months:
1........................................... 4,722 2,317 1,356 665
3........................................... 2,632 1,969 1,253 653
5........................................... 1,724 1,435 1,153 622
12.......................................... 772 680 591 488
----------------------------------------------------------------------------------------------------------------
\a\ The Commission simulates the tick assignment procedure under 16 combinations of evaluation and operative
period lengths. The simulation is performed on a rolling basis for every month from January 2018 to June 2022.
The methodology and data used for this simulation is described in the note to table 10. This table computes
the number of tick changes that occur over a median 12-month horizon for each combination of evaluation and
operative period length.
To summarize the results of this subsection, tables 10, 11, and 12
illustrate the tradeoffs inherent in the choice of evaluation and
operative periods. With respect to the evaluation period, table 10
implies that shorter period lengths reduce false negatives but increase
false positives; the higher prevalence of false negatives tilts the
scale toward a short evaluation period. Table 11, though, shows that a
one-month evaluation period may unduly increase the influence of
aberrant events, while a three-month evaluation period continues to
perform well in unusual market conditions.
With respect to the operative period, tables 10 and 11 imply that
shorter period lengths reduce both false negatives and positives. Table
11 shows a particularly large reduction in false positives when using a
six-month operative period instead of a twelve-month period. Focusing
on an evaluation period of three-months, tables 10 and 11 show similar
error rates with operative period of three and six months. However, a
three-month operative period requires market participants to update
their systems twice as often as a six-month period, and table 12 shows
that the number of annual tick changes increases 57% when moving from
the six-month to the three-month period.\1386\ The amendments, which
adopt a three-month Evaluation Period and a six-month operative period,
reflect these considerations.\1387\
---------------------------------------------------------------------------
\1386\ With a three-month evaluation period, a six-month
operative period results in 1,253 annual tick changes, while a
three-month operative period results in 1,969. The increase is
therefore calculated as: (1969 - 1253)/1253 = 57%.
\1387\ See 17 CFR 242.612(a)(1), (b)(1) for rule text relating
to these periods.
---------------------------------------------------------------------------
Finally, the Commission examined the effect of a one-month lag
between the end of the evaluation period and the subsequent tick
assignment. All the results in tables 10, 11, and 12 include this one-
month lag; the Commission separately calculated table 10 statistics
without the lag. If the lag is removed, then the error rates in panels
A and B of table 10 fall by a small amount across all combinations of
evaluation and operative periods. The lag effectively makes the
evaluation period less informative about the operative period, and this
effect is stronger when updates are made every month. With respect to
magnitudes, removing the lag causes the fraction of trading that is a
false negative to decrease by an average of 0.9% with a maximum
decrease of 1.2% for the top left cell of panel A; the false positives
in panel B likewise drop by an average of 0.2% when the lag is removed.
On the other side of the scale, the lag may reduce operational burdens
on market participants by allowing them time to update their systems
before a new tick assignment occurs, and may likewise provide brokers
time to give advance notice to customers about which stocks will
experience tick changes.\1388\ The amendments, which provide a one-
month lag, reflect a conclusion that the cost of increased error rates
due to the one-month lag is small relative to the benefits that the lag
provides for industry testing and adjustment.
---------------------------------------------------------------------------
\1388\ The adopted rule additionally syncs the dates of tick
assignment with the dates of new round lot assignments. This further
reduces operational burdens on market participants. See discussion
in section VII.D.5.a.
---------------------------------------------------------------------------
e. Additional Effects of a Half-Penny Tick
Some commenters suggested, in the context of the proposal, that a
tiered tick size could create confusion in the markets particularly for
retail traders.\1389\ These commenters stated
[[Page 81731]]
that retail traders may be at a relative disadvantage because they may
get confused by the sub-penny increments. One commenter presented data
suggesting that retail traders tend not to use sub-penny non-marketable
limit orders even when they can, for instance for orders priced below
$1.00, and that fill rates for such trades tend to go down for retail
traders.\1390\ The commenter ascribes these findings to retail
investors getting confused and being unfamiliar with sub-penny trading
increments leading to a competitive disadvantage for these traders with
respect to more sophisticated traders that are more familiar with sub-
penny increments and thus retail traders using non-marketable limit
orders would be more likely to be undercut with a smaller tick.\1391\
Another commenter suggested that retail investor confusion would lead
retail brokers to need to hire additional staff to manage customer
confusion and education concerning multiple tick sizes, or to handle
phone trades by retail customers confused by the markets.\1392\
---------------------------------------------------------------------------
\1389\ Fidelity Letter at 3; Tastytrade Letter at 18-19.
\1390\ See Fidelity Letter at 3.
\1391\ See also O'Brien Letter at 4 and Tastytrade Letter at 20
making similar comments.
\1392\ See Tastytrade Letter at 18-19.
---------------------------------------------------------------------------
The Commission acknowledges some potential for confusion but does
not expect the amendments to disadvantage retail traders. First, the
comments are in response to the proposal; the adopted amendments have
fewer minimum quoting increments than the proposed amendments; indeed,
there are only two as opposed to four. Second, any confusion or
disadvantage must be evaluated relative to the baseline. Currently the
penny tick creates a price floor, which, especially for tick-
constrained stocks, advantages faster and generally more sophisticated
traders. It also creates incentives for more complex strategies such as
those involving alternative venues, again, contributing to complexity
and putting less sophisticated investors at a disadvantage. Third, the
one-month period of time between measuring the TWAQS and the assigned
tick size becoming operational will allow time for broker-dealers to
educate customers about tick sizes.\1393\ For these reasons, the half-
penny tick is not expected to disadvantage retail traders.
---------------------------------------------------------------------------
\1393\ See supra note 307 and surrounding text.
---------------------------------------------------------------------------
One commenter requested that the Commission consider how a smaller
tick size could affect stock splits.\1394\ The commenter cited academic
research suggesting that stock splits can be used to affect the bid-ask
spread.\1395\ To the extent that issuers engage in stock splits to
manage their quoted spread, this behavior is likely to continue under
the amended rules when issuers believe that a stock split could improve
their liquidity. However, the amendments are expected to improve
liquidity on average for stocks subject to the smaller tick size, so it
may be less likely that an issuer may feel the need to manage liquidity
via stock splits going forward and so there could be fewer associated
stock splits for this reason. Furthermore, although stock splits and
reverse splits can mechanically affect the bid-ask quoted spread and
change the optimal tick, the next evaluation period would rectify any
misalignment in the stock's tick size assignment.\1396\
---------------------------------------------------------------------------
\1394\ See CCMR Letter at 28-29.
\1395\ See James J. Angel, Tick Size, Share Prices and Stock
Splits, 52 J. Fin. 655 (1997), and Sida Li & Mao Ye, supra note
1286.
\1396\ See supra section VII.C.1.c for additional discussion of
tick sizes, quoted spreads, and stock splits. See supra note 1367
and supra section III.C.8 for discussions of how tick sizes are
assigned following a stock split or reverse split.
---------------------------------------------------------------------------
Some commenters requested that the Commission provide an analysis
of the effect of the amendments on the use of ISOs (intermarket sweep
orders).\1397\ Some stated that spreading liquidity over more price
levels would increase the risk of information leakage with regards to a
large order being split over many smaller orders which would lead to an
increased complexity of implementing large orders and would thus lead
to an increased use of ISO orders.\1398\ Barardehi et al. (2022) \1399\
specifically examine the effect of tick sizes on ISO activity. Their
study finds that a narrower tick is associated with more ISO activity
which the authors attribute to the increased market complexity
associated with implementing trades in a narrower tick environment.
This is consistent with the effect considered by the commenters.
Consequently, ISO usage is likely to increase for stocks that receive
the narrower tick size. One commenter stated that increased use of ISO
orders may result in more locked and crossed markets.\1400\ Research on
the link between ISO usage and locked markets is scant. Increased ISO
usage could eventually lead to an increase in locked and crossed
markets, which could make transacting on exchanges more complicated,
however the magnitude of any effect is uncertain.
---------------------------------------------------------------------------
\1397\ See SIFMA Letter II at 34. See also 17 CFR 242.600(b)(47)
(defining an intermarket sweep order).
\1398\ See, e.g., Vanguard Letter at 5, Citigroup Letter at 4,
and Virtu Letter II at 17.
\1399\ See Barardehi et al., supra note 231.
\1400\ See RBC Letter at 3.
---------------------------------------------------------------------------
Another commenter specifically asked the Commission to consider the
effect of a lower tick size on ETFs as opposed to stocks.\1401\ The
commenter stated that ETFs tend to have larger trade sizes and also
that the creation and redemption process for ETFs is unique. However,
the commenter does not provide any analysis regarding how these
differences would lead an ETF to react differently than a common stock
to a reduction in the tick size.\1402\ The fundamental economics
regarding the tick size tradeoff discussed at the beginning of this
section applies to ETFs because the economics discussed in this section
rely on the mechanics of quoting and trading, not on the assets
underlying the stock or ETF. It is also unclear how the creation
redemption process would differ from the analysis provided above.
Additionally, the creation/redemption process also does not produce
unique economics in the context of these amendments. This is because
authorized participants purchasing the underlying shares to deliver in
exchange for shares of the ETF or delivering shares of the ETF in
exchange for the underlying assets would still need to purchase and
sell the underlying shares in the stock market, subjecting them to the
economics of supply and demand for liquidity provision for the stocks
in question. Consequently, ETFs with narrower quoted spreads likely
will experience an improvement in market quality with a $0.005 tick
size. To the extent that ETFs have larger average trade sizes the
benefits of the Rule may be smaller, consistent with the analysis
presented in Barardehi et al. (2022). But as the adopted amendment is
more conservative than the proposed rule in that it applies a tick size
of $0.005 to stocks with quoted spreads equal to or less than $0.015,
the Commission does not believe, based on the analysis above and
commenters evidence presented above, that the amendments are likely to
harm ETFs that receive a tick size reduction.
---------------------------------------------------------------------------
\1401\ See Tradeweb Letter at 3.
\1402\ Id.
---------------------------------------------------------------------------
One commenter stated that updated Rule 605 data would ``question
the validity of assumptions'' made in the tick size proposal.\1403\ The
commenter stated that Rule 605 execution quality data for stocks that
have quoted spreads wider than the tick would demonstrate that the
minimum quoting increment is not the driver of off-exchange retail
[[Page 81732]]
trading.\1404\ The Commission agrees that there are other factors
driving off-exchange retail trading, and adopts changes to Rule 612 for
reasons other than a significant change in retail order flow. For this
reason, additional information regarding retail execution quality that
will arise from amended Rule 605 is not needed prior to adopting
amendments to Rule 612.
---------------------------------------------------------------------------
\1403\ See Citadel Letter III at 1-2.
\1404\ Id.
---------------------------------------------------------------------------
One commenter stated that variable tick sizes ``raise[ ] concerns
about the ability to compare the execution quality for the stock across
multiple months,'' resulting in ``a significant possibility of investor
confusion when comparing Rule 605 reports across several months.''
\1405\ The Commission acknowledges that changes in the tick size may
result in changes to the levels of some measures of execution quality
that are sensitive to the tick size, such as price improvement, over
time. To the extent that this reduces the interpretability of Rule 605
reports, particularly for stocks that experience frequent changes in
the tick size, this could represent a cost of the amendments. However,
there are several factors that will mitigate this potential cost.
First, the adopted amendments include an operative period that limits
the frequency at which a tick assignment is updated to a minimum of six
months. The fact that a stock's tick assignment cannot vary more
frequently than every six months greatly reduces the number of
potential changes in execution quality levels in monthly Rule 605
reports that result from changes to a stock's tick size. Second, to the
extent that market participants will be able to combine Rule 605
information with information about a stock's historical tick
size,\1406\ this will allow them to control for this characteristic
when assessing a stock's execution quality data over time. In addition,
as acknowledged by the commenter, a change in the tick size ``may
impact market centers and broker-dealers reporting under 605 in the
same manner,'' \1407\ such that variations in the tick size (and the
resulting mechanical effects on execution quality levels) will not
impact the use of Rule 605 to compare execution quality across
reporting entities within a given month.
---------------------------------------------------------------------------
\1405\ See SIFMA Letter II at 20-21.
\1406\ While there is no requirement for the listing exchange to
disseminate historical information about the tick size, it is likely
that this information will be collected and disseminated by other
market participants, such as firms providing services related to
financial data and analysis because there would likely be demand for
such data because, for example, it would be needed to perform after
the fact transaction cost analysis.
\1407\ See SIFMA Letter II.
---------------------------------------------------------------------------
If FINRA chooses not to update Rule 5320 (the `Manning Rule'), the
lower tick size could make claiming the price improvement exception to
FINRA Rule 5320 harder for market participants since it would require a
two-tick price improvement for some stocks instead of a one tick price
improvement. One commenter stated that because the price improvement
exception in the Manning Rule is currently tied to $0.01 price
improvement, a tick size smaller than $0.01 would ``greatly increase
the cost and complexity of compliance and would likely disincentivize
(or eliminate) the handling of customer limit orders by wholesale
broker dealers.'' \1408\ The commenter made this statement specifically
referencing the proposed $0.001 tick increment. The adopted amendments
do not include the $0.001 tick size and so these concerns are
significantly mitigated. Nonetheless, requiring two tick price
improvement for some orders could increase the complexity associated
with complying with the Manning Rule, particularly in situations where
the quoted spread is only one tick wide because it would require more
than crossing the quoted spread in order to claim the exception. For
broker-dealers, such as wholesalers, whose business models center on
internalizing customer orders within the NBBO, the requirement to, in
some instances, more than cross the quoted spread in order to execute a
customer order could be a disincentive to handling some orders as it
could render such trades unprofitable.\1409\
---------------------------------------------------------------------------
\1408\ See Citadel Letter I at 8.
\1409\ See supra note 193 and surrounding text for further
discussion.
---------------------------------------------------------------------------
One commenter stated that a lower tick size could lead to
oscillation in some stocks between tick sizes.\1410\ The commenter
stated that a stock that falls just under the threshold and thus
receives a smaller tick may be subject to more undercutting with the
smaller tick size, which could cause quoted spreads to widen. Wider
spreads would make the stock revert to the wider tick size, which would
reduce undercutting so that quoted spreads would decline. A narrower
spread could lead to a smaller tick in the next round and so on.\1411\
The Commission believes that this outcome is unlikely given the
analysis provided in this section. Stocks receiving the smaller tick
size are expected to experience smaller quoted spreads due to the
smaller tick size allowing pricing that better reflects supply and
demand. This effect would reduce oscillation. Stocks receiving the
smaller tick size would likely experience tighter quoted spreads making
it less likely that they would revert to the $0.01 tick in the next
evaluation period.
---------------------------------------------------------------------------
\1410\ See Mitre Corp. Letter at 5.
\1411\ Id.
---------------------------------------------------------------------------
2. Lower Access Fee Cap
The amendments will lower the access fee cap from $0.003 per share
(30 mils) to $0.001 per share (10 mils) for NMS stocks priced at $1.00
or more, and from 0.3% to 0.1% of the share price for stocks with
prices less than $1.00. Lowering the access fee cap preserves price
coherence,\1412\ given changes to the tick size. Moreover, the
Commission expects that lowering the access fee cap will result in
lower transaction costs for investors. The Commission also expects that
a lower access fee cap will result in wider quoted spreads; however,
market quality will nonetheless improve. Lowering the access fee cap
will reduce exchange transaction revenue due to lower capture on sub-
$1.00 stocks. We describe these effects in more detail below.
---------------------------------------------------------------------------
\1412\ See infra section VII.D.2.a
---------------------------------------------------------------------------
Commenters, with few exceptions,\1413\ agreed on the need for
Commission action on access fees given the change in the tick
size.\1414\ One commenter stated that in light of the reduction in
ticks for some stocks to $0.005, leaving the access fee cap at 30 mils
would ``distort trading economics in a manner that undermines the
Commission's goals for competition and Best Execution.'' \1415\ Many
commenters
[[Page 81733]]
supported a 10 mils access fee.\1416\ Some commenters went further,
stating that the Commission should explore ``comprehensive access fee
reform'' or ban rebates entirely.\1417\ Some commenters suggested what
they viewed as a less extensive change, namely an alternative in which
some stocks had a fee of 15 mils whereas others had a fee of 30
mils.\1418\ In short, while commenters agreed on the need for
Commission action to lower the access fee cap, they disagreed regarding
the specifics.
---------------------------------------------------------------------------
\1413\ Exceptions include e.g. Citigroup Letter at 6, and World
Federation of Exchanges Letter at 4, Pragma Letter at 7, Hudson
River Letter at 4, and Budish Letter at 6-7. However, these
commenters did not present arguments suggesting that an access fee
greater than 50% of the tick size would not cause price coherence
problems. The Commission believes that retaining a 30 mil access fee
for stocks trading with a $0.005 tick would further separate the
price from the economics of the trade and disrupt the coherence
between nominal and net pricing as the access fee cap would be
greater than 50% of the tick size.
\1414\ See, e.g., Nasdaq Letter I at 19 (stating ``Nasdaq
recognizes that if Commission action successfully updates tick sizes
and narrows spreads for certain stocks, then existing exchange
access fees and rebates may no longer be appropriate.''), See Cboe,
State Street, et al. Letter at 3 (stating ``We acknowledge that a
reduction in quoting increments for tick constrained symbols could
make it advisable for market centers to reduce access fees for the
affected symbols to ensure a consistent equity market structure
framework.''), see Better Markets Letter II at 3-4 (stating ``A
reduction in the minimum tick size without reducing access fees
could permit fees to become a higher percentage of the minimum
pricing increment, which would almost certainly undermine price
transparency.''), and see RBC Letter at 4 (stating ``If the MPIs are
meaningfully reduced as noted in the Proposal, then access fees
would need to be lower as well.'').
\1415\ Nasdaq Letter I at 19.
\1416\ See, e.g., BlackRock Letter at 10-11, BMO Letter at 3-4,
Budish Letter, IEX Letters I-V, JPMorgan Letter at 6, NASAA Letter
at 9, Vanguard Letter at 2 and 6, XTX Letter at 5.
\1417\ See We The Investors Letter I at 3-4 (recommending
banning rebates); Harris Letter at 4 (recommending reverting to
traditional fees, thereby effectively eliminating rebates).
BlackRock Letter at 11 states ``Although we believe that the current
proposal may miss an opportunity to enact more holistic and lasting
access fee reform, we concede that, for highly liquid securities, a
10 mil access fee cap reasonably threads the needle between
countervailing adverse consequences. Accordingly, under a uniform
fee model, we would be supportive of setting the access fee cap at
10 mils.''
\1418\ See infra note 1805 for a list of commenters suggesting
this alternative, and a discussion of the costs and benefit of this
alternative.
---------------------------------------------------------------------------
a. Coherence Between Net and Quoted Prices
In the Proposing Release, the Commission discussed the need to
maintain an access fee cap that is less than half of the tick size due
to the need to maintain coherence between net and quoted prices.\1419\
Commenters, with few exceptions,\1420\ agreed.\1421\ Reducing the
access fee cap to 10 mils will satisfy this condition of coherence. As
explained in the Proposing Release, only the best posted price is
protected. Under the current regulatory framework, leaving the access
fee cap at 30 mils could preclude market participants from trading on
exchanges that have the best displayed price when fees and rebates are
included. Suppose a traditional exchange has a displayed protected bid
at $10.010, whereas an inverted exchange has a displayed protected bid
at $10.005. Order protection would require the exchange to have
policies and procedures in place reasonably designed to prevent trades
from occurring at a price worse than the protected quote,\1422\
effectively requiring the investor to go to the traditional exchange if
the investor wished to trade against a displayed, on exchange bid.
However, on the traditional exchange, the investor demanding liquidity
would take home $10.007, whereas on the inverted exchange, the investor
would take home $10.008.\1423\
---------------------------------------------------------------------------
\1419\ See Proposing Release, supra note 11, at 80348 n.712. Net
and quoted price rankings are coherent if sorting trading venues on
the competitiveness of their quoted prices yields the same ordering
as sorting on prices net of fees and rebates.
\1420\ Exceptions are Citigroup Letter at 6, and World
Federation of Exchanges Letter at 4. These commenters do not present
arguments that counter those others in the comment file. The
Commission believes that retaining a 30 mil access fee for stocks
trading with a $0.005 tick would further separate the price from the
economics of the trade and disrupt the coherence between nominal and
net pricing as the access fee cap would be greater than 50% of the
tick size.
\1421\ See, e.g., MEMX Letter at 22-24 and Pragma Letter at 7.
\1422\ See 17 CFR 242.611(a)(1); see also 17 CFR 242.611(b)
(exceptions).
\1423\ For the liquidity demander in this example, the net
proceeds of selling to the liquidity provider on the traditional
exchange would be the quoted price of $10.01 less the $0.003 access
fee, or $10.007. On the inverted venue the liquidity demander would
receive the quoted price of $10.005 plus a taker rebate of $0.003
from selling, or $10.008. Although the liquidity demander would
receive a better net price by selling at the inverted venue, because
the traditional exchange has the better quoted price the order
protection rule will prevent the trader from accessing the liquidity
on the inverted venue before first accessing the liquidity on the
traditional exchange.
---------------------------------------------------------------------------
Closely related to the lack of price coherence on exchanges is the
effect on price transparency. To illustrate, consider a situation with
an access fee and rebate of $0.003 and a tick size of $0.005. Consider
the effect on prices of a stock. Suppose one sees a trade executed at a
price of $10.005 followed by another executed at a price of $10.010.
Many investors would interpret this as a sign that a stock was
increasing in value. However, with an access fee of $0.003, the net
price of the first order if it represents a market order to buy is
$10.008 (the buyer pays $10.005 + $0.003), whereas the net price of the
second order if it is a market order to sell, is $10.007 (the seller
receives $10.010 and pays $0.003 in fees). The price has fallen, not
risen, an effect that only the most sophisticated market participants
would be able to discern. \1424\ Lowering the access fee cap to 10 mils
would solve both of these problems.
---------------------------------------------------------------------------
\1424\ See Budish Letter at 6.
---------------------------------------------------------------------------
b. Quantitative Net Capture Analysis
While the amendments do not directly dictate what rebates trading
venues can offer, trading venues generally finance rebates through
access fees, so in practice reducing the access fee cap will lower the
rebates offered.\1425\ If trading venues were to subsidize rebates by
taking a net loss per share transacted, they would be vulnerable to
experiencing extreme and unpredictable losses if volumes spike. Such a
trading venue could experience such losses if its non-transaction fee
sources of revenue do not increase enough with a spike in trading
volume to offset their negative net capture. Trading volumes can vary
significantly through time, with little ability for a trading venue to
predict the timing and magnitude of changes in trading volume. For
example, in January 2021 volume spiked dramatically for certain stocks
relative to pre-January 2021 levels.\1426\ Exchanges could face
financial hardship should rebates deviate substantially from fees; so
it is unlikely that exchanges would take this risk. For this reason,
rebates and the cap on access fees are tied together.
---------------------------------------------------------------------------
\1425\ See supra note 1077 and surrounding text discussing that
access fees fund transaction rebates and while trading centers could
subsidize rebates with non-fee revenues they do not do so in
practice.
\1426\ See Staff Report on Equity and Options Market Structure
Conditions in Early 2021 (Oct. 14, 2021) available at https://www.sec.gov/files/staff-report-equity-options-market-struction-conditions-early-2021.pdf.
---------------------------------------------------------------------------
As explained in section VII.C.2.b, the Commission understands that
the net capture for non-auction trading in stocks that have a price
equal to or greater than $1.00 is likely close to 2 mils for most
exchanges. An exchange net capture rate of approximately 2 mils is in
line with current pricing practices at most exchanges; it is reasonable
to estimate that exchanges would realize a similar net capture rate
because the current net capture rate will remain possible under the
adopted amendments. The Commission acknowledges uncertainty over
whether this 2 mils capture rate will persist or be different should
trading venues choose to alter their business model in response to the
change in access fees. The analysis that follows assumes that exchanges
will maintain the practice of financing rebates through access fees,
and thus for transactions in stocks priced $1.00 or more the Commission
expects the average access fee to be near the 10 mil access fee cap and
the average rebate to be approximately 2 mils lower.\1427\ The analysis
also assumes that the behavior of inverted exchanges and off-exchange
venues changes proportionally. Although the amendments would not
require proportional change on the part of inverted venues, there is
currently no restriction on the level of rebates for taking liquidity
or fees for posting; yet, as shown in table 4 in section VII.C.2,
inverted venues generally have fee and rebate levels similar to maker-
taker
[[Page 81734]]
venues and approximately a 2 mils capture rate.
---------------------------------------------------------------------------
\1427\ See section VII.C.2 for additional discussion about the
roughly 2 mil estimated net capture rate for exchanges. At certain
pricing tiers rebates may exceed the access fee cap. However,
because total overall fees exceed the total rebates paid out, the
average rebate would remain lower than the average access fee.
---------------------------------------------------------------------------
Table 13 uses volume estimates from table 5 to provide estimates of
the fees and rebates that would have been collected and disbursed in
2023 if the amended access fee cap was implemented.\1428\ Panel A shows
that under the current system with a 30 mil access fee cap for
quotations priced $1.00 or more and a 0.3% access fee cap for
transactions less than $1.00, the exchanges collected an estimated $3.4
billion in access fees and distributed $3.1 billion in rebates in 2023,
providing an estimated net capture of $337 million for the exchanges in
that time period.\1429\ In table 13 the Commission estimates that the
exchanges would have collected $1,188 million in access fees and
distributed $906 million in rebates in 2023 under the amendment to Rule
610, providing the exchanges a net capture of $282 million in that time
period. Thus, the Commission estimates that total access fees collected
would have declined by $2.23 billion and rebates distributed by $2.17
billion in 2023.\1430\ This amounts to an estimated decline in net
capture of $54.9 million across all exchanges. This decline is
conditional upon exchanges maintaining a 2 mil capture rate for stocks
trading at a price of $1.00 or higher. Any estimated changes in total
net capture across exchanges is due exclusively to the change in the
access fee cap for stocks trading below $1.00.
---------------------------------------------------------------------------
\1428\ This assumes that exchanges continue the practice of
funding rebates through access fees, that trading volumes are
unchanged relative to 2023, that the distribution of trading volume
across exchanges is unchanged, and that the distribution of trading
volume priced below $1.00 and at or above $1.00 remains unchanged.
\1429\ See table 6 for additional analysis on current estimates
of exchange net capture.
\1430\ Balancing out expected rebates paid on make-take,
inverted, and flat fee venues, the Commission estimates that
liquidity demanders will pay $1.93 billion per year less in access
fees netted across all venues under the Rule and liquidity providers
will receive $1.88 billion per year less in rebates netted across
all venues.
---------------------------------------------------------------------------
Panel B provides estimates of the effect of the amendments on
access fees paid and rebates received by liquidity demanders and
providers separately. The Commission estimates that, under the
amendments, liquidity demanders would have paid $1.93 billion less
\1431\ in access fees and liquidity providers would have received $1.88
billion less in rebates in 2023. Thus, the current estimated $2.6
billion in fee funded rebates in 2023 would have decreased by
approximately 70% under the amendments.
---------------------------------------------------------------------------
\1431\ The ultimate effect of this change will not result in
liquidity demanders saving the full $1.93 billion in transaction
costs, because the effect of reduced rebates will cause the quoted
spread to widen, offsetting this reduction in the access fee. See
supra section VII.B.3, infra section VII.D.2.c for a discussion of
these points.
\1432\ This $54 million estimate is lower than the estimated $89
million per loss year provided in the Proposing Release. The
difference comes because the adopted access fee cap for transactions
priced below $1.00 is higher than the proposal: 0.10% compared to
0.05% proposed. This reduces the loss on transactions priced below
$1.00. Additionally, as can be seen by comparing Panel B of table 5
in the Proposing Release and table 4 herein, multiple exchanges have
lowered access fees for transactions below $1.00 since the proposal
making the Rule's difference from the baseline smaller.
\1433\ As discussed in section VII.C.2, the Commission estimates
that most exchanges have a net capture of approximately 2 mils on
transactions priced greater than $1.00. For reasons discussed in
this section the Commission believes that it is reasonable to assume
that exchanges with a current 2 mil net capture would be able to
continue to earn a 2 mil net capture.
\1434\ See supra table 4. Most exchanges do not offer rebates
for stocks priced less than $1.00, or if they do the rebates are
quite small.
\1435\ See supra table 5.
\1436\ The benchmark model in section VII.B.3 implies that a
reduction in the access fee will cause the liquidity demand curve to
shift, resulting in a higher volume of trades at sub-dollar prices.
See also infra note 1462 and surrounding text for a case study on
the effect of a rebate instituted by MEMX for sub-dollar trades,
which resulted in a higher level of sub-dollar trades; if a
reduction in the access fee has a similar effect on equilibrium
trading as the institution of the rebate, then the volume of sub-
dollar trades will increase and the reduction in exchange revenue
will be mitigated.
Table 13--Estimated Access Fees and Rebates Collected--Current and Adopted 2023 \a\
----------------------------------------------------------------------------------------------------------------
Current Rule Difference
----------------------------------------------------------------------------------------------------------------
Panel A: Estimated Access Fees Collected and Rebates (in Millions of Dollars)
----------------------------------------------------------------------------------------------------------------
Fees Collected.................................................. 3,414.00 1,188.91 -2,225.21
Rebates Distributed............................................. -3,076.50 -906.16 2,170.30
Exchange Capture................................................ 337.66 282.75 -54.90
----------------------------------------------------------------------------------------------------------------
Panel B: Estimated Fees by Liquidity Type (in Millions of Dollars)
----------------------------------------------------------------------------------------------------------------
Liquidity Demander.............................................. 2,969.12 1,034.61 -1,934.51
Liquidity Provider.............................................. -2,631.47 -751.86 1,879.61
Exchange Capture................................................ 337.66 282.75 -54.90
----------------------------------------------------------------------------------------------------------------
\a\ This table takes trading volumes presented in table 5 to calculate aggregate fee and rebate estimates under
the Rule. Current estimates of fees collected and rebates distributed are taken from table 6. The analysis
presumes that exchanges with fees and rebates currently above 10 mils will decrease fees and rebates to a 10
mil fee and 8 mil rebate (the exceptions being IEX which charges 10 mils to takers and rebates 4 mils to
makers, NYSE Chicago which charges both sides 10 mils, and LTSE which does not charge fees). For trading in
securities priced less than $1.00, estimates of fees and rebates presume that all sub $1.00 fees from panel B
of table 4 which are over 0.10% are reduced to 0.10%, fees at or below 0.10% remain the same. Computations are
made per exchange and then aggregated as shown above.
Table 14 presents analysis showing an estimated total reduction of
approximately $55 million per year in net capture due to the reduction
in the access fee cap and how it might in turn affect the transaction
revenues of each of the various exchange families. This estimated
decline in transaction revenue comes exclusively from the reduction in
the access fee cap for transactions in securities below $1.00.\1432\
This is because, as previously explained, the Commission expects that
for transactions priced equal to or greater than $1.00 the exchanges
should be able to maintain their current net capture.\1433\ For
transactions priced below $1.00 most exchanges currently charge the
maximum 0.3% but typically offer no rebates.\1434\ Because very few
exchanges offer rebates on stocks priced below $1.00, the access fee
represents the exchange's net capture. Lowering the access fee from
0.3% to 0.1% on these transactions will represent a decrease in net
capture of 66% for many exchanges. This decrease may vary across
exchanges. Some exchanges do not charge any fees for trading in sub
$1.00 securities, while others charge a fee to both sides of a sub
$1.00 transactions. Additionally, the exchanges differ in the fraction
of sub
[[Page 81735]]
$1.00 trading volume that they handle.\1435\ Table 14 provides an
annual estimate of the effect on exchange transaction revenue of
lowering the access fee on exchanges' net capture given realized
volumes in 2023 for each exchange group. To the extent that the
reduction in the access fee causes more trading at sub-dollar prices,
table 14 overestimates the reduction in exchange transaction
revenue.\1436\
Table 14--Estimated Effect of Rule on 2023 Exchange Transaction Revenue
for Stocks Prices Below $1.00 \a\
------------------------------------------------------------------------
Transaction Transaction
revenues ($) revenues (%)
------------------------------------------------------------------------
Nasdaq.................................. -$18,593,052 -20
NYSE.................................... -18,750,074 -19
Cboe.................................... -12,375,769 -16
MEMX.................................... -4,517,207 -21
IEX..................................... 0 0
MIAX.................................... -667,838 -7
LTSE.................................... 0 0
-------------------------------
Total............................... -54,903,941
------------------------------------------------------------------------
\a\ The variable Transaction Revenue ($) provides an annualized
estimate of the effect of the amendment to Rule 612 on exchange net
capture. For all exchanges, other than LTSE which doesn't charge an
access fee and IEX which has an assumed net capture of 6 mils per
share traded above $1.00 (Panel A of table 4 shows that IEX charges a
fee of 10 mils coupled with a rebate of 4 mils), the net capture on
transaction priced equal to, or greater than, $1.00 per share is
expected to remain unaffected by the amendments at the assumed 2 mils
per share. The 2 mils per share assumption is further discussed in
section VII.C.2.c.Thus, the Commission does not expect any decrease in
overall exchange transaction revenue per share for shares priced above
$1.00. For transaction volume below $1.00 per share estimates for the
decline in transaction revenue is computed by assuming that under the
amendments all exchanges currently charging more than 0.10% for
transactions will lower the transaction fee to 0.10%. Exchanges
currently charging access fees less than or equal to 0.10% will
continue to charge their current rates. The list of current estimated
exchange sub $1.00 pricing comes from panel B of table 4. Sub $1.00
dollar volume estimates for each exchange are from table 5. The
estimated transaction revenue under the amendments is compared to the
estimated transaction revenue in the current environment that is
estimated using the sub $1.00 transaction fees/rebates for each
exchange presented in table 4 panel B and multiplying these fees by
volume estimates for each exchange from table 5. See section VIII.C.2
for tables 4 and 5. The difference is presented in the table 14 along
with the precent change in transaction revenue from the baseline.
Lastly, transaction fees in stocks priced less than $1.00 serve to
increase the net cost of accessing liquidity as they do not tend to
fund rebates to liquidity providers so there is no incentive that could
induce spreads to narrow and on average offset the fee.\1437\ Lower
transaction costs for these securities may improve liquidity for stocks
with prices less than $1.00. However, given the relatively low natural
trading interest, the Commission does not expect a significant
improvement in the trading environment for these securities.
---------------------------------------------------------------------------
\1437\ See supra section VII.B.3 for additional discussion of
how fee-funded rebates are largely off-set by changes in the quoted
spread to keep net-costs the same.
---------------------------------------------------------------------------
Given the low net capture rates, the Commission concludes that in
most cases, access fees are typically used to fund rebates and not used
exclusively to fund execution services. Multiple commenters stated that
current access fees and fee caps are not reflective of the current
actual costs of providing execution services.\1438\ One commenter
stated that the cost of processing and matching trades has dropped with
technological advances.\1439\ Another commenter stated that ``the fees
charged by exchanges are often far in excess of those necessary to
maintain operations of the exchange.'' \1440\ A commenter pointed to
significant reductions in spread and commissions since 2005, which have
resulted in the 30 mil access fee cap representing a more significant
economic factor in trading.\1441\
---------------------------------------------------------------------------
\1438\ See, e.g., Vanguard Letter at 6, Verret Letter I at 7,
and Retirement Coalition Letter at 1.
\1439\ See Better Markets Letter I at 15.
\1440\ See Healthy Markets Letter I at 22.
\1441\ Citing its own previous comment letters one commenter has
stated since at least 2014 that a reduction in the access fee cap is
warranted given the reduction in trade commissions and narrowing of
spreads relative to when the 30 mil access fee cap was first
established. See Citigroup Letter at 5.
---------------------------------------------------------------------------
One commenter stated that a 10 mil access fee cap would represent a
``fair pricing model based on the `cost plus reasonable return'
methodology'' by citing that ATSs charge 10 mil fees while employing
similar technologies as exchanges.\1442\ On this latter point, a
commenter stated that a uniform access fee cap of 10 mils would, ``have
the added benefit of aligning exchange fees with prevailing ATS fees,
and creating a more equitable competitive landscape across trading
venues.'' \1443\
---------------------------------------------------------------------------
\1442\ See Verret Letter I at 5-7. One commenter provided their
own review of form ATS-N and specifically looked at minimum and
maximum ATS fees; the commenter reported that the maximum ATS fee
often exceeds 10 mils by a considerable margin, see Nasdaq Letter
III at 3. The Commission does not dispute that maximum ATS fees can
exceed 10 mils, but the maximum ATS fee is not an appropriate
benchmark for exchange access fees because the maximum ATS fee can
be a function of particular services (e.g., block trades or special
order types) or of subscriber characteristics (e.g., subscriber
order flow might be segmented into specific categories), while an
exchange's access fee schedule applies to all members. Another
commenter presented analysis on the subset of ATSs that primarily
operate a ``continuous book'' market and are therefore most closely
comparable to exchanges. The commenter's analysis indicates that
seven such ATSs--representing 42% of all ATS volume--charge a
maximum of 10 mils. The commenter concluded that the standard rate
in the competitive ATS market is 10 mils, while rates substantially
above 10 mils are due to specialized services not available on
exchanges, see IEX Letter VI at 5; the Commission agrees.
\1443\ See BlackRock Letter at 11.
---------------------------------------------------------------------------
In contrast, one commenter stated that the Commission had not
established that the ``proposed reduced fee caps do, in fact, bear a
reasonable relationship to the actual costs to an exchange of a
trade.'' \1444\ The same commenter stated that technological costs are
not significant determinants of access fee levels, but rather that the
fees reflect the magnitude of risk associated with providing liquidity
as well as the value to the market that having access to those quotes
provides.\1445\ The commenter further stated that the current access
fee cap is not ``unreasonably high,'' because, among other things,
``exchange platform costs to market participants have remained
competitive over
[[Page 81736]]
time.'' \1446\ As the quantitative net capture analysis shows, the
access fee cap in the adopted amendments will still permit typical
exchange net captures. Thus, for stocks priced greater than $1.00,
lowering the access fee cap is not expected to affect the contribution
to revenue and thus to the platform. The following section discusses
the effects of a reduction of fees and therefore rebates on the
provision of liquidity.
---------------------------------------------------------------------------
\1444\ Nasdaq Letter I at 22.
\1445\ See Nasdaq Letter I at 21; Nasdaq Letter II at 4.
\1446\ Nasdaq Letter II at 2 and 5.
---------------------------------------------------------------------------
c. Effects on Liquidity and Transaction Costs
The main implications for liquidity from reducing the access fee
caps follow the basic principles laid out in sections VII.B.3, as well
as the empirical results in section VII.C.2. Most exchanges charge
close to the preexisting access fee cap due, in part, to the
disincentive to unilaterally reduce fees and rebates. For the same
reasons that exchanges charge close to the preexisting access fee cap,
the Commission believes that lowering the access fee cap will lead
exchanges to charge similarly close to the new cap. Some exchanges that
are currently charging less than the amended access fee cap may
continue to do so. Exchanges will most likely not alter their net
capture rates, implying that much of the access fee will continue to
fund rebates in stocks priced above $1.00. For reasons discussed
further below, the reduction in the access fee cap is likely to leave
the cost of accessing liquidity unaffected for some stocks, and to
reduce the cost of accessing liquidity for others.
For stocks priced less than $1.00, the reduction in the access fee
cap is also likely to reduce the cost of accessing liquidity. Unlike
for stocks priced $1.00 or more, for stocks priced less than $1.00 most
exchanges charge an access fee without providing a rebate.\1447\ Since
there is no rebate, which would serve to narrow spreads and offset the
cost of the access fee, the access fee only serves to increase the cost
of taking liquidity in these stocks. Therefore, a reduction in the
access fee from 0.3% to 0.1% for stocks priced less than $1.00 will
lower the cost to take liquidity. Some exchanges offer rebates in
transactions in stocks priced less than $1.00.\1448\ In these
instances, under the assumption that these exchanges will uniformly
reduce their fees and rebates to maintain the same net capture rate,
the reduction in the access fee cap is not expected to affect the cost
of taking liquidity.
---------------------------------------------------------------------------
\1447\ See supra table 4.
\1448\ For example, the Cboe EDGX and MEMX exchanges offer
rebates for sub-$1.00 stocks. See supra table 4 and surrounding
discussion noting that only a few exchanges offer rebates in
transactions for stocks priced $1.00 or less.
---------------------------------------------------------------------------
Additionally, reducing the access fee cap for stocks priced less
than $1.00 from 0.3% to 0.1% of the quote price will also ensure that
the cost of accessing liquidity is similar for stocks with one quote
below $1.00 and another quote equal to or greater than $1.00. Consider
a stock with a best bid quote at $0.99 and a best ask quote at $1.00.
Under the amendments, the maximum fee to access the bid quote is 9.9
mils and is roughly equal to the 10 mils maximum fee to access the ask
quote. Had the Commission lowered the access fee cap for stocks priced
$1.00 or more but left it unchanged at 0.3% for quotes priced less than
$1.00, the cost of accessing the sub-$1.00 quote would be relatively
more expensive than the cost of accessing the $1.00 or more quote.
Here, had the fee cap for quotes priced at or higher than $1.00 been
reduced to 10 mils but the fee cap for sub-$1.00 trades remained at
0.3%, the maximum allowable fee to access the $0.99 quote would be 29.7
mils, roughly 3 times greater than that of accessing the ask price.
Having a large differential between access fees on opposite sides of an
order book would inhibit the ability of markets to reach prices most
reflective of the underlying value.\1449\
---------------------------------------------------------------------------
\1449\ Reducing the access fee cap for trades priced at $1.00
per share or greater to 10 mils without a similar reduction in the
fee cap for those priced below $1.00 could distort markets by
introducing an incentive for market participants to exploit
differences in fees and rebates for stocks near the $1.00 threshold.
If the maker rebates available under the 0.3% fee cap for sub $1.00
stocks are greater than those for quotes priced at or greater than
$1.00 then market participants, then market makers may be
incentivized to push prices below $1.00 as they could capture higher
rebates by posting at bid and ask at $0.98 and $0.99 respectively as
opposed to quoting at $1.00 and $1.01.
---------------------------------------------------------------------------
Commenters on the proposed access fee cap reduction focused on
access fees for stocks priced above $1.00.\1450\ Several commenters
argued for an alternative in which stocks with a half-penny tick would
have an access fee of 15 mils, whereas stocks with a penny tick would
have an access fee of 30 mils (hereafter ``15 mils/30 mils
alternative'').\1451\ As discussed in sections VII.B.3 and VII.C.2,
there is a strong economic tie between the level of the access fee cap
and the ability to pay rebates. The discussion among commenters focused
on the effect on rebates, with some commenters who favored of the 15
mils/30 mils alternative naming the ability to pay rebates as the
primary reason for the higher access fee cap; \1452\ other commenters
specifically were in favor of a ban on rebates.\1453\ One of these
commenters stated that banning rebates (by requiring exchanges to
revert to a pricing model where both sides of a transaction were
charged a fee) would fix the problems associated with access fees and
rebates, but stated that a second best solution would be to impose a
uniform access fee on all exchanges.\1454\
---------------------------------------------------------------------------
\1450\ An exception is Cboe Letter IV and Letter II, discussed
further at the end of this subsection.
\1451\ See, e.g., NYSE, Schwab, and Citadel Letter at 2, Nasdaq
Letter I at 2, MMI Letter at 7, Robinhood Letter at 5, and MEMX
Letter at 23-24. See also Nasdaq Letter IV and NYSE Letter I.
\1452\ See, e.g., Nasdaq Letter I at 2.
\1453\ See We The Investors Letter I at 3-4 (recommending
banning rebates); see also Harris Letter at 4 (recommending
reverting to traditional fees, thereby effectively eliminating
rebates).
\1454\ See Harris Letter at 4.
---------------------------------------------------------------------------
Commenters in favor of the 15 mils/30 mils alternative expressed
the concern that a 10 mils access fee cap would reduce the flexibility
to offer rebates. The commenters assert that rebates are necessary to
compensate liquidity providers to post displayed (or ``lit'') quotes on
exchanges.\1455\ According to the commenters' logic, a lower access fee
cap translates into a lower rebate, which translates into fewer lit
quotes. These commenters also state that those quotes that are posted
are likely to be wider.\1456\ Wider and fewer posted quotes, according
to these commenters, signify lower market quality.
---------------------------------------------------------------------------
\1455\ See, e.g., Nasdaq Letter I at 21, Nasdaq Letter II at 5-
7, Interactive Brokers Group Letter at 5, Virtu Letter II at 10,
Citadel Letter I at 24, WFE Letter at 4, CCMR Letter at 27, and
State Street Letter at 4.
\1456\ See id. Some commenters specifically identified the NBBO
as a matter of concern (Nasdaq Letter IV at 7; Goldman Letter at 8;
Nasdaq Letter I at 22, Virtu Letter II at 10.). The NBBO reflects
lit quotes at a specific size and thus the arguments regarding the
NBBO (with an exception described in more detail below) are the same
as those for lit liquidity more generally.
---------------------------------------------------------------------------
The Commission agrees that lowering the access fee cap is also
likely to lower rebates because trading venues use access fees to fund
rebates.\1457\ The Commission also agrees that quoted spreads (spreads
that do not reflect rebates or access fees) on lit exchanges are likely
to be wider because liquidity providers would be expected to widen
spreads to compensate for the lower rebates \1458\--though the fact
that the tick size amendments will lower spreads means that the two
amendments combined may in fact lead to lower quoted spreads on some
stocks.
---------------------------------------------------------------------------
\1457\ See supra section VII.C.2 discussing why trading venues
fund rebates with access fees and why rebates are not funded by
other revenue sources.
\1458\ See section VII.B.3 discussing how spreads are expected
to widen in response to a reduction in fee-funded maker rebates so
to keep the net cost of liquidity constant.
---------------------------------------------------------------------------
[[Page 81737]]
The Commission, however, disagrees with the commenters' statements
that lower rebates from lower access fees will lower market quality and
increase transaction costs. The Commission draws on the economic
principles articulated in section VII.B.3.\1459\ Figure 1 shows how the
quoted spreads respond to an equal increase of an access fee and rebate
of 30 mils assuming a stock is not tick-constrained. The change
contemplated here is a shift of 20 mils because that is the difference
between the baseline fee cap of 30 mils and the amended fee cap of 10
mils. When the fees and rebates change together, supply and demand
intersect at the same quantity point (thus liquidity offered would be
unchanged) but at a different price point, leading to a wider quoted
spread. The net spread (the net cost of trading), which takes into
account the fees and rebates, would be unchanged.\1460\ Thus, the
Commission disagrees with commenters who argue on the basis of quoted
spread that the 10 mils access fee will lead to increased trading costs
and lower liquidity.
---------------------------------------------------------------------------
\1459\ The discussion in that section regarding neutrality of
fees and rebates does not depend on the access fee charged per share
being equal to the rebate, but rather on fees and rebates being
reduced or increased by the same amount. As the Commission does not
expect the net capture rate to change, the neutrality result
applies.
\1460\ As explained in section VII.B.3, any change in access
fees or rebates may be passed from brokers to customers either
directly or indirectly, such as through changes in commissions or
changes in the broker's services.
---------------------------------------------------------------------------
Crucially, the reasoning above applies only to a stock with an
economic spread of greater than the tick.\1461\ When the economic
spread is less than a tick, rebates funded by fees result in a pricing
distortion, as section VII.B.3.b explains. The price at which liquidity
providers would be willing to offer liquidity is less than one tick in
the presence of the rebate. However, the tick forms a binding price
floor, leading to an oversupply of liquidity. Specifically, the set
price of liquidity results in economic rents that accrue to some at the
expense of others, in this case to those able to get to the front of
the queue the fastest. For these stocks, lowering the access fee will
better equate supply and demand and lower transaction costs for
investors broadly.
---------------------------------------------------------------------------
\1461\ See section VII.B.1 for the definition of the economic
spread.
---------------------------------------------------------------------------
One commenter discussed the introduction of a rebate for sub-dollar
trades on MEMX and MIAX.\1462\ The rebate, when introduced, was
initially set to 0.3% of the dollar value of the trade,\1463\ and was
reduced to 0.05% several days later. The commenter presents empirical
results indicating that the effective spread fell from approximately
0.4% to 0.25% of dollar value after the introduction of the
rebate,\1464\ and almost completely reversed back to 0.4% days later
when the rebate was reduced to 0.05%.\1465\ In short, a rebate of 0.3%
of dollar value led to a reduction in effective spreads of 0.15% of
dollar value. The commenter's empirical result is consistent with the
model presented in figure 1. The model presented in figure 1 predicts
that a rebate will cause the liquidity supply curve to shift by the
amount of the rebate--liquidity suppliers are willing to offer
liquidity at a lower price on account of the rebate. In contrast to
panel B of figure 1, however, the commenter's example does not include
an increase in the access fee to fund the rebate; therefore, the
commenter's example can be modelled by recreating panel B without the
shift in the demand curve--i.e., the introduction of the rebate will
cause the equilibrium outcome to shift from the point where the dotted
lines intersect to the point where the solid supply curve intersects
with the dotted demand curve. The model therefore has multiple
empirical predictions for the commenter's example: when the rebate is
introduced, without a similar increase in access fees, the model
predicts that spreads will fall and the equilibrium amount of liquidity
transacted will rise; when the rebate is rolled back, the model
predicts that spreads will rise and the equilibrium amount of liquidity
transacted will fall. The model's predictions on spreads are borne out
by the commenter's data--spreads fell 0.15% when the rebate of 0.3% was
in place, and spreads reverted when the rebate was rolled back.\1466\
The model's prediction on the quantity of liquidity transacted are also
borne out by Commission analysis--when the 0.3% rebate was in place,
the dollar-volume of sub-dollar trades increased by a factor of
three.\1467\ In sum, the introduction of a rebate for sub-dollar trades
on MEMX and MIAX resulted in a market reaction that is directionally
consistent with the Commission's economic model presented in section
VII.B.3.a and figure 1. The large and abrupt tripling of trading volume
is also
[[Page 81738]]
consistent with concerns that rebates cause excessive
intermediation.\1468\
---------------------------------------------------------------------------
\1462\ See Nasdaq Letter II at 5-6 (stating ``spreads would
widen if access fees were to become inadequate to fund rebates to
market makers and other participants that provide displayed
liquidity to the markets. This widening would likely be significant,
as the data below suggests. It shows that in early December 2020,
when MIAX and MEMX first introduced rebates for sub-dollar stocks,
spreads for such stocks fell dramatically, but when MIAX and MEMX
then slashed rebates soon thereafter, spreads reverted to their
prior levels.'').
\1463\ That is, the rebate for sub-dollar trades was initially
set equal to the access fee cap for sub-dollar trades.
\1464\ See Nasdaq Letter II at 6. The effective spread is
calculated as the signed difference between the execution price of a
trade and the prevailing midpoint (i.e., the execution price minus
the midpoint for buy orders and the midpoint minus the execution
price for sell orders); the commenter then divides this by the
midpoint price to arrive at the effective spread as a percentage of
the price (mirroring the fact that the rebate is paid as a
percentage of the execution price). The effective spread differs
from the quoted half-spread because a trade may receive price
improvement--that is, the trade may execute at a better price than
the best quote, so that the effective spread is lower than the
quoted half-spread--or a large trade may execute against multiple
levels of the order book. Both the effective spread and the quoted
spread are measures of liquidity, but the quoted half-spread
measures the prospective cost of trading immediately at the best
available prices while the effective spread measures the ex-post
cost of trading immediately (accounting for hidden orders and other
sources of price improvement not known ex-ante, as well as order
size). Additionally, because the effective spread measures the ex-
post cost of trading immediately, it can only be calculated in the
presence of a trade. Therefore, the effective spread is typically
calculated by taking a weighted average of the effective spread
across transactions--the commenter, for example, weighted the
effective spread by the notional amount of each transaction. The
quoted spread can be averaged over time--as with the TWAQS--because
it is an ex-ante measure.
\1465\ The commenter's results can be exhibited with a numerical
example. Suppose in the absence of rebates a stock trades with a
best offer of $0.52 and a best bid of $0.48, yielding a midpoint of
$0.50. A liquidity supplier at the offer would therefore receive
proceeds of $0.52 when their offer is executed against. The
effective spread in the commenter's example would be calculated as
the distance between the execution price and the midpoint, divided
by the midpoint: ($0.52 - $0.50)/$0.50 = 4%. Now suppose that a
rebate of 0.3% is offered by the exchange. In the commenter's
analysis, this reduces the effective spread by 0.15% to 3.85% (from
4%). This implies that the offer price would shrink from $0.52 to
approximately $0.51925 (keeping the midpoint constant at $0.50 and
using the fact that the effective spread must equal the difference
in the ask and the midpoint, divided by the midpoint so that 3.85% =
($0.51925 - $0.50)/$0.50). The liquidity provider would therefore
earn $0.51925 plus the rebate of 0.30% for a total proceed of
$0.5208 ($0.51925 + 0.003 * $0.51925).
\1466\ The fact that spreads fell by less than the amount of the
rebate indicates that rebates do not generally lower trading costs
beyond the cost of funding the rebate; this is contrary to one
commenter's statement that, ``any cost savings non-retail investor
participants realize from a reduction in the access fee cap are
likely to be more than consumed by the rising frictional costs . . .
associated with wider spreads.'' See Cboe Letter IV at 5, and
further discussion surrounding infra note 1492.
\1467\ In the week of Nov. 23, 2020, there was daily trade
volume at sub-dollar execution prices of approximately $330 million;
the figure was $383 million in the week of Dec. 7. The intervening
week--the week of MEMX's 0.3% rebate for sub-dollar executions--saw
$1,025 million in daily trade volume at sub-dollar prices. The
calculations are constructed using all normal trades that execute
during normal trading hours from TAQ. Following the methodology in
Nasdaq Letter II at 6, the calculations for the week of Nov. 30
exclude Nov. 30 and Dec. 4.
\1468\ See supra note 1005, and see also the Proposing Release,
supra note 11, at 80292.
---------------------------------------------------------------------------
Some comments address the question of incentives for trading on
exchanges. These commenters state that, as a result of the Commission's
adoption of 10 mils versus 15 mils/30 mils alternative, the volume on
lit exchanges will decline.\1469\ Other commenters disagreed, stating
that lower access fees could lead volume on exchanges to
increase.\1470\ However, the above analysis indicates that liquidity
providers would not be deterred from quoting on exchange because they
could widen the quote, thereby receiving the same economic profit as
they received with the rebate. Liquidity demanders would not be worse
off because the reduction in access fee would offset, or, in the case
of stocks with an economic spread of less than a tick, more than
offset, the increase in spread.
---------------------------------------------------------------------------
\1469\ See, e.g., Cboe Letter II at 8-9, Cboe Letter IV at 3-5,
Nasdaq Letter I at 22-23, Nasdaq Letter II at 4, Nasdaq Letter IV at
9, and Nasdaq Letter V at 2, predicting that a reduction in rebates
will increase segmentation and may make ATSs and single-dealer
platforms more attractive. See also infra note 1761 and surrounding
text.
\1470\ See Better Markets Letter I at 15 stating that ``A
reduction in access fees will impose lower costs on investors,
removing a disincentive for trading on exchanges.'' Healthy Markets
Letter I at 22, stating ``Brokers' avoidance of these [access] fees
is a significant contributor for brokers often choosing to
internalize or first route to ATSs or OTC market makers, rather than
to exchanges'', IEX Letter I at 26 (stating ``A substantial
reduction in the access fees will be impactful for those investors
and is likely to increase their willingness to trade on exchanges. .
. . The result can be an increase in the use of displayed exchange
trading and an improvement in the price discovery function of the
market, with broad benefits extending beyond trading on exchanges
themselves''), See also IEX Letter IV at 18-19, BMO Letter at 3, and
Themis Letter at 7-8.
---------------------------------------------------------------------------
Commenters specifically stated that posted quotes on exchange face
the risk of adverse selection. They state that a premium is necessary
to compete with the off-exchange market, and that the rebate provides
that premium. As other commenters state, this does not take into
account the access fee, which (all else equal) discourages liquidity
takers from accessing exchanges. Moreover, a premium can come in the
form of the spread as opposed to a rebate. While the order protection
rule requires that trading centers enforce policies and procedures that
are reasonably designed to prevent trades from being executed at a
price worse than the protected quote, nothing prevents off-exchange
non-displayed liquidity being at a better price for the liquidity taker
and worse price for the maker, and indeed that happens under the
current fee/rebate structure. Rather than moving liquidity off-
exchange, liquidity providers could widen the difference between on-
and off-exchange quotes, leaving the underlying economic tradeoff the
same.
The above analysis shows that the same opportunities that are
available on-exchange in today's environment are still expected to be
available with the adoption of these amendments, even under the lower
access fee cap (indeed, these opportunities are expected to improve due
to the amendments to Rule 612). However, commenters state that the off-
exchange environment may change due to the amendments.\1471\ These
commenters raise concern regarding the amount of liquidity that is
displayed versus non-displayed. One commenter stated that wider quoted
spreads on exchange increase the range of prices at which trades
execute off-exchange.\1472\ Because Rule 611 generally requires that
off-exchange trades execute within the NBBO, as on-exchange spreads
widen, a liquidity provider, now facing a wider NBBO, would be able to
offer a wider spread off-exchange than that liquidity provider could do
now. The commenter appears concerned that this ability to offer a wider
spread off-exchange than previously will attract liquidity to off-
exchange, and more specifically, non-displayed venues.
---------------------------------------------------------------------------
\1471\ See supra note 1479 and surrounding discussion.
\1472\ See Cboe Letter III at 5: ``Wider spreads are likely to
most benefit wholesale broker-dealers, that may be able to offer
more levels of price improvement, but at the expense of increased
frictional costs for investors.''
---------------------------------------------------------------------------
However, while liquidity providers would have the ability to offer
wider spreads off-exchange than prior to the amendments, they would not
necessarily have the incentive to do so. For while wider spreads would
mean greater profits for the liquidity provider, that is only the case
if their orders are filled. As stated by a commenter, off-exchange
liquidity would still need to compete with on-exchange liquidity, and
that on-exchange liquidity is now less expensive to access due to a
lower access fee cap.\1473\ If the spread off-exchange were to widen,
non-displayed off-exchange quotes would be unlikely to attract
liquidity takers. Therefore, there is not an incentive for liquidity
providers to migrate off-exchange due to wider spreads on exchange. To
summarize, spreads may widen on-exchange increasing pricing flexibility
off-exchange, even so exchanges are not expected to lose volume due to
the reduction in the access fee cap through this mechanism.
---------------------------------------------------------------------------
\1473\ See IEX Letter IV at 23: ``The fact that exchanges use
rebates to draw orders from other exchanges says nothing about the
ability of exchanges to attract more orders that now go to off-
exchange venues by using lower access fees and offering better
execution quality.''
---------------------------------------------------------------------------
One commenter stated that volatility may increase due to the wider
quoted spread when the access fee reduction causes a reduction in
rebates.\1474\ The Commission acknowledges that wider spreads
definitionally imply a greater difference between the bid and the ask.
However, spreads that better reflect the true underlying cost of
liquidity are more efficient than spreads that mask this cost.
---------------------------------------------------------------------------
\1474\ See Goldman Sachs Letter at 8.
---------------------------------------------------------------------------
One commenter stated that a reduction in rebates will lead to more
off-exchange trading, which in turn will cause the NBBO to widen, and
result in worse execution for off-exchange trading, because of the way
some off-exchange trading uses the NBBO as ``a reference price for
benchmark pricing and other risk functions.'' \1475\ First, the
Commission describes above why the adopted amendments will not result
in a large amount of trading moving off-exchange. Furthermore, while
the Commission does expect the quoted spread, and therefore the NBBO,
to widen, we disagree that this will result in worsening off-exchange
executions. This commenter provided two examples of situations in which
off-exchange executions might worsen. The first is an ATS that provides
execution mechanisms based on the NBBO.\1476\ As explained in section
VII.B.3.a, the reduction in access fees and corresponding reduction in
rebates will not change the net spread on exchange. This means the cost
of liquidity will not materially change. There is no reason why ATSs
that base execution prices off the NBBO cannot alter their pricing
formulas to preserve the same execution prices (e.g., by executing
inside the NBBO by a pre-determined amount). Indeed, a typical example
of such matching mechanisms are mechanisms that match buy and sell
orders at the midpoint, and this will not be impacted at all by a wider
NBBO. The second example provided by the commenter was the case of
retail wholesalers. The commenter states that these wholesalers
[[Page 81739]]
``may be able to offer more levels of price improvement,'' but this
will come at the expense of increased costs of trading from wider
spreads.\1477\ The Commission again disagrees with this assertion.
Because the cost of liquidity will be largely unchanged, the price
improvement \1478\ acknowledged by the commenter will be capable of
offsetting the change in quoted spread.
---------------------------------------------------------------------------
\1475\ See Cboe Letter IV at 5: ``. . . the NBBO is utilized by
many market participants as a reference price for benchmark pricing
and other risk functions. In addition, if on-exchange liquidity
moves to off-exchange venues such as alternative trading systems,
these trading centers commonly use the NBBO as a reference price for
executing transactions, which will make transactions in off-exchange
venues more expensive as well. Wider spreads are likely to most
benefit wholesale broker-dealers, that may be able to offer more
levels of price improvement, but at the expense of increased
frictional costs for investors.''
\1476\ See Cboe Letter IV at 5.
\1477\ See Cboe Letter IV at 5.
\1478\ Retail wholesalers frequently offer ``price improvement''
on orders they receive, where they execute the order on a principal
basis at a price better than the NBBO.
---------------------------------------------------------------------------
One commenter stated that eliminating the access fee would cost
retail investors as much as $678 million per year.\1479\ The commenter
arrives at this estimate by using the BJZZ algorithm to identify retail
trades from TAQ data,\1480\ and computes the effective/quoted ratio (EQ
ratio) for each trade.\1481\ The effective to quoted spread ratio
computed as follows
---------------------------------------------------------------------------
\1479\ See Nasdaq Letter II at 6.
\1480\ ``BJZZ'' refers to the algorithm designed by Boehmer
Ekkehart, Charles M. Jones, Xiaoyan Zhang, and Xinran Zhang. See
Boehmer Ekkehart, et al., Tracking Retail Investor Activity, 76 J.
Fin, 2249 (2021).
\1481\ The EQ ratio measures how close to the NBBO or NBBO
midpoint a trade executes at. A trade executing at the midpoint
would have an EQ ratio of 0, while a trade that executes at the NBBO
would have an EQ ratio of 1.
[GRAPHIC] [TIFF OMITTED] TR08OC24.004
They assume that spreads will widen for all trades by 60 mils in the
absence of rebates. They then apply the observed EQ ratio to the
hypothetical 60 mil wider spreads in the absence of rebates to compute
hypothetical transaction costs for retail investors under a world
without rebates.
The Commission disagrees with the commenter's assertion that retail
traders will receive worse execution due to the reduced access fee cap.
As this section describes, quoted spreads for stocks trading with more
than one tick intra-spread are expected to widen on average by about 40
mils. However, the commenter's analysis relies on the assumption that
the EQ ratio for retail order executions will remain constant.\1482\
The commenter provided no evidence to support this assumption. This
assumption is important because economically what matters is not the
distance of the trade price from the NBBO, but rather the distance of
the trade price from the midpoint--the effective spread. The effective
spread covers the costs associated with providing liquidity as well as
provides the liquidity provider's profits. Assuming a constant EQ ratio
in the commenter's analysis implies that wholesalers internalize retail
orders at prices that are farther from the midpoint, and thus the
wholesaler will earn more money without providing any additional
benefit to retail traders or their broker-dealers.\1483\ Wholesalers
are subject to competitive forces that apply at the level of average
execution quality, and it is unlikely that market forces would allow
such excess profits to wholesalers for no additional benefit to
persist.\1484\ It also seems unlikely that the EQ ratio would change
mechanically with the NBBO because if the NBBO itself were the primary
determinate of the price level at which retail trades were
internalized, and wholesalers were free to choose any price level
within the NBBO, then wholesalers would routinely internalize orders at
or near the NBBO implying an EQ ratio for retail trades of near 1. This
is not the case. Wholesalers currently internalize retail orders at
prices that are significantly inside of the NBBO (i.e., EQ ratios
significantly less than 1) suggesting that other factors besides the
NBBO itself, such as distance from the midpoint, determine the
transaction price of retail orders that are internalized by
wholesalers.\1485\ These price levels will still be feasible for
wholesalers under the amendments, and so even with a wider NBBO,
wholesalers are likely to transact retail orders at similar price
levels under the amendments as they are today.
---------------------------------------------------------------------------
\1482\ The commenter's methodology is also flawed because the
BJZZ algorithm they employ to identify retail trades has been shown
in recent research as not being a very accurate measure of retail
trading volume, See Brad M. Barber, Xing Huang, Philippe Jorion,
Terrance Odean, & Christopher Schwarz, A(sub)penny For Your
Thoughts: Tracking Retail Investor Activity in TAQ (working paper,
Aug. 14, 2023), available at https://ssrn.com/abstract=4202874
(retrieved from SSRN Elsevier database. If the algorithm does not
reliably identify retail trades then it is unclear what can actually
be learned about retail trading volume from the exercise.
\1483\ The effective spread is defined as the signed difference
between an order's execution price and the midpoint of the quoted
spread; the larger the difference the less competitive the executed
price is relative to the midpoint. Because the EQ ratio is equal to
the effective spread divided by the quoted spread, the effective
spread would have to increase at the same scale by which the quoted
spread widens in order to keep the ratio constant.
\1484\ So long as there is some degree of competition, this
argument would hold. A market that is more competitive may have
retail effective spreads that would be lower, but in either case a
change in the NBBO, with no other changes to wholesaler costs or
competition would not be expected to change wholesaler profits.
\1485\ See Citadel Letter I at 33 showing EQ ratios ranging from
.27 for small retail orders to .88 for very large orders, CCMR
Letter at 35 showing average EQ ratios around .5 for the top three
wholesalers and Charles Schwab, U.S. Equity Market Structure: Order
Routing Practices, Considerations, and Opportunities. (2022)
(``Schwab 2022 Whitepaper'') at 9, 16, available at https://content.schwab.com/web/retail/public/about-schwab/Schwab-2022-order-routing-whitepaper.pdf (showing its EQ ratio of .33).
---------------------------------------------------------------------------
Put another way, economically there is no reason to assume that
relaxing a non-binding constraint, in this case widening the quoted
NBBO, would have an effect on existing equilibrium behavior. The NBBO
does not constitute a binding constraint for wholesale execution of
many retail trades.
In cases where the NBBO does constitute a binding constraint, there
are two important missing pieces from the commenter's analysis. The
first is that, for on-exchange execution, the wholesaler will pay a
lower access fee. Assuming (as the commenter's analysis implicitly
does) that the wholesaler does not pass on these lower fees at least in
part to some investors assumes a lack of competitive dynamics in the
retail execution market. Second, the commenter's methodology fails to
take into account the expected reduction in quoted spreads for some
stocks due to the reduction in the tick size. In cases where the NBBO
constitutes a binding constraint on wholesaler price improvement, then
wholesalers will offer better execution on these stocks. Moreover,
there are cases in which current wholesale execution may fall between
the spread under the new tick size, and the previous spread. In these
cases, the new tick size creates a new binding constraint, leading to
better execution for retail investors. So, if anything, the combined
effect of the amendments could improve retail execution quality on
average.
Additionally, as a general matter, some commenters stated that the
lower access fee on exchanges will make exchanges a more attractive
place to access liquidity.\1486\ The Commission believes that the cost
of accessing liquidity will decline for those stocks which continue to
trade with a one tick wide spread; the Commission, however, disagrees
that the cost of accessing liquidity will change on average for other
stocks.\1487\
---------------------------------------------------------------------------
\1486\ See Healthy Market Letter at 23, BlackRock Letter at 11,
BMO Letter at 3, and Themis Letter at 7-8.
\1487\ For stocks which do not trade with a one tick wide spread
the cost of accessing liquidity for any one instance may be higher
or lower with a reduced fee cap, however on average the cost of
accessing liquidity is not expected to change for those stocks. See
supra section VII.B.3.b for additional discussion.
---------------------------------------------------------------------------
Some commenters stated concerns that less liquid stocks may be more
susceptible to any negative effects on liquidity from a reduction in
rebates.\1488\
[[Page 81740]]
Other commenters suggested that a higher fee cap should be adopted for
illiquid stocks.\1489\ However, adopting a separate fee cap for
illiquid stocks would introduce more complexity into the market. The
Commission's response is the same as the broader concern regarding
posted liquidity: spreads may widen but the cost of accessing and
providing liquidity will on average not change.
---------------------------------------------------------------------------
\1488\ See Nasdaq Letter II at 5-7: ``This peril is particularly
acute for thinly-traded securities.'' and Virtu at 10: ``The reduced
incentives for liquidity in thinly traded securities is especially
concerning given how much liquidity improvements actually reduce an
issuer's cost of capital and impact their ability to attract
investors.'' See also Virtu Letter II at 10, Tastytrade Letter at 2.
\1489\ See Citigroup Letter at 6, TRP Letter at 4-5.
---------------------------------------------------------------------------
One commenter stated a reduction in rebates would lead to the exit
of liquidity providers, harming competition,\1490\ a lower access fee
cap is expected to reduce the fees which are used to fund rebates and
consequently rebates are expected to also see a reduction. The
Commission acknowledges that profits of liquidity providers may fall
because for stocks that remain tick-constrained, the access fee
represents a transfer from liquidity demanders to liquidity providers.
However, a net decrease in competition could serve to widen the spread
beyond a single tick and increase the proceeds of liquidity provision
thus incentivizing liquidity provision. While the Commission
anticipates and is sensitive to costs to some affected parties, the
Commission expects investors, more broadly to benefit.\1491\
---------------------------------------------------------------------------
\1490\ See Virtu Letter II at 19, Cboe Letter III at 3-4.
\1491\ Some Commenters agree. See, e.g., Verret Letter I at 9,
Retirement Coalition Letter at 2, Themis Letter at 7, ASA Letter at
4.
---------------------------------------------------------------------------
Another commenter stated that lowering the access fee cap would
result in a ``liquidity gap''; that ``it is unlikely that the liquidity
gap would be met by other market participants;'' and that ``as spreads
further widen, any cost savings non-retail investor participants
realize from a reduction in the access fee cap are likely to be more
than consumed by the rising frictional costs.'' \1492\ The Commission
disagrees that lowering the access fee cap would result in a liquidity
gap that market participants would not be able to fill, because if
quoted spreads widen beyond any reduction in maker rebates, liquidity
providers would stand to earn higher proceeds by supplying at the wider
spread. The Commission believes that competition among liquidity
providers will keep the cost of accessing liquidity from rising on
average in securities where the minimum quoting increment is not a
meaningful constraint.\1493\ In those stocks that trade with a quoted
spread equal to the tick size, the Commission expects that the
reduction in access fees will reduce the net cost of accessing
liquidity.\1494\ Therefore, non-retail investors would likely see a
reduction in overall frictional costs. Finally, one commenter stated
that reducing the access fee cap for stocks priced less than $1.00
would impact an exchange's ability to differentiate itself, and the
estimated decrease in transactions revenue from these stocks would
limit its investments in innovation and technologies.\1495\ The impact
on an exchange's ability to offer different fees and rebates for stocks
priced less than $1.00 is not likely to be large as there is not a
substantial degree of differentiation across exchanges currently.\1496\
Most exchanges charge fees near or at the fee cap to liquidity takers,
and only two exchanges offer rebates to liquidity takers. This is
unlikely to change following a decrease in the fee cap. The Commission
acknowledges a loss in revenue due to the reduction in rebates for
stocks priced below $1.00. It is possible that this could impact
exchange investment in new technologies. However, as discussed above in
this section, the amendments to Rule 612 are anticipated to lead to
more volume on exchange, and hence more trading revenue to exchanges,
offsetting this effect. Moreover, as discussed earlier in this
subsection, it is necessary to conform rebates for stocks priced below
$1.00 with those for stocks priced above $1.00.
---------------------------------------------------------------------------
\1492\ See Cboe Letter IV at 3, 5; see also Cboe Letter III, at
6.
\1493\ See section VII.B.3.
\1494\ Id.
\1495\ See Cboe Letter II at 9; see also Cboe Letter IV at 1.
\1496\ See supra table 4 and surrounding discussion.
---------------------------------------------------------------------------
One commenter suggested implementing the amendments to reduce the
access fee caps before the minimum pricing increments to isolate the
impact of the access fee cap on its own.\1497\ The Commission has
separately considered the impact of the amendments to Rule 612 with the
new access fee in place; specifically, the change in the access fee
will cause quoted spreads to widen, which may cause some stocks that
currently would qualify for a reduction in their tick size under the
adopted amendments to Rule 612 to no longer qualify for such a
reduction.\1498\ However, many stocks will continue to qualify. While
the Commission acknowledges that postponing amendments to Rule 612
would allow for time to study the access fee cap in isolation, there is
no mechanism by which a reduction in the access fee cap, on its own,
would yield the full benefits of the proposed amendments to Rule 612.
---------------------------------------------------------------------------
\1497\ See State Street Letter at 5, stating: ``We recommend . .
. [i]mplementing any changes to access fee caps before changing
quoting increments, to isolate and evaluate the effects. This
includes examining whether reducing the access fee cap may affect a
security's designation as `tick-constrained.' ''
\1498\ See supra table 7, note a.
---------------------------------------------------------------------------
The Commission acknowledges that changing the access fee cap at the
same time that the changes to Rule 612 are implemented will cause some
stocks assigned to a narrower tick size to immediately trade with a
quoted spread too wide to qualify for continued assignment to the
$0.005 tick size bucket. However, a delay in implementing changes to
Rule 612 would delay the accrual of the other benefits the Commission
has identified for these changes.
d. Other Effects of the Access Fee Cap Reduction
The Commission anticipates additional benefits inherent to adopting
a lower access fee cap for all securities.
Access fees that fund rebates contribute to complexity in markets
because they separate both the true cost of demanding liquidity and the
proceeds from supplying liquidity, as represented by the quoted half-
spread. Commenters stated that lowering the access fee cap to 10 mils
would reduce complexity; one commenter stated in the context of
supporting a significant reduction in exchange access fees, that
current pricing models ``contribute to market complexity by encouraging
rebate arbitrage strategies and the proliferation of new order types
and trading venues designed to exploit different transaction pricing
models.'' \1499\ Similarly, a second commenter supported a 10 mil fee
cap and stated that maker-taker models, ``introduce unnecessary market
complexity through proliferation of new exchange order types (and new
exchanges) designed solely to take advantage of pricing models.''
\1500\ The same commenter stated that maker-taker pricing may drive
orders off exchanges to avoid access fees, and ``benefit sophisticated
market participants, like market makers and proprietary traders, at the
expense of other market participants.'' \1501\
---------------------------------------------------------------------------
\1499\ See Vanguard Letter at 6.
\1500\ See BMO Letter at 3.
\1501\ See id.
---------------------------------------------------------------------------
One manifestation of this complexity is the potential conflict of
interest between broker-dealers and their
[[Page 81741]]
customers.\1502\ Multiple commenters stated that a benefit of a lower
access fee cap is that it would mitigate such potential conflicts of
interests.\1503\ Other commenters disagreed.\1504\ Some commenters
state that due to the complexity, opacity, and potential conflicts
inherent in the rebate structure, the Commission should go further than
in the current adopted amendments and ban rebates altogether.\1505\ The
Commission agrees that lowering the access fee cap reduces complexity
and may help alleviate potential conflicts of interest.
---------------------------------------------------------------------------
\1502\ See supra note 1518 and surrounding text discussing the
potential conflicts of interests that exchange fees and rebates may
introduce.
\1503\ See infra note 1514.
\1504\ See, e.g., Nasdaq Letter I at 2.
\1505\ See Harris Letter at 4-5, and We The Investors Letter I
at 7, arguing that the Commission should go further and ban the use
of rebates.
---------------------------------------------------------------------------
Finally, the reduction in the access fee cap will improve market
quality for stocks that remain tick-constrained. While the amendments
to Rule 612 create a smaller tick size for stocks with narrow spreads,
spreads naturally vary over time and this variation introduces the
possibility that some stocks could be misassigned to a tick size
because trading in the operative period differs from the evaluation
period due to factors exogenous to the tick change.\1506\ Panel A of
table 10 shows that approximately 13.7% of aggregate share volume will
be a false negative under the amendments \1507\--that is, this volume
will be assigned a penny tick, but will trade at a TWAQS below $0.015
and therefore trade with a tick-constrained spread a majority of the
time. Moreover, some stocks may remain tick-constrained, even at the
new half-penny tick. Reducing the access fee cap would lower the cost
of accessing liquidity because fee-funded rebates serve as a pure tax
on liquidity demanders whenever the spread is tick-constrained
(creating a wealth transfer from liquidity demanders to liquidity
providers); the reduction in the access fee cap will also reduce the
excess supply of liquidity at this price floor.\1508\
---------------------------------------------------------------------------
\1506\ See section VII.D.1.d for discussion and analysis of the
tradeoffs inherent in tick assignment.
\1507\ This number corresponds to an evaluation period of three
months and an operative period of six months, which are the
parameters of the rule text. See section VII.D.1.d.
\1508\ As discussed above (e.g., supra section VII.B.2), a
binding price floor on liquidity results in more liquidity supply
than demand. Maker-taker pricing exacerbates this problem by taxing
demand and subsidizing supply at the price floor. By reducing the
access fee, the excess supply is lessened. The lower cost of
accessing liquidity can also extend to stocks which trade with a
TWAQS greater than $0.015 to the extent to which these stocks may
occasionally trade with a spread equal to the tick size.
---------------------------------------------------------------------------
3. Exchange Fees and Rebates Determinable at the Time of Execution
In the current environment, as discussed in section VII.C.2, market
participants often have to make trading decisions without the ability
to determine the exchange fees and rebates they incur at the time of
execution, and a market participant's total cost of trading can vary by
a significant amount for orders with the same quoted execution price
once exchange fees and rebates are accounted for. Current exchange fees
and rebates are often based on the participant's relative contribution
to the exchange's monthly trading volume during the contemporaneous
month, and market participants need to grapple with the uncertainty of
forecasting future market outcomes should they wish to know what their
trading costs are at the time that they execute a trade.\1509\
Requiring fees and rebates to be determinable at the time of execution
will result in the benefits of increased transparency as to what fees
and rebates broker-dealer members are committed to pay when they trade,
reducing potential conflicts of interest, and potentially improving
broker-dealer routing decisions. These amendments will also result in
costs to exchanges associated with revising existing fee schedules to
bring them into compliance with the adopted amendments.
---------------------------------------------------------------------------
\1509\ See also Proposing Release, supra note 11, at 80292 for a
discussion of the complexity of fee schedules and the difficulty in
forecasting fees for a contemporaneous period.
---------------------------------------------------------------------------
The Commission received comments from a broad range of commenters
who expressed support for Proposed Rule 610(d) because it would provide
enhanced transparency surrounding transaction fees and rebates and
alleviate concerns related to potential conflicts of interest.\1510\
For example, one commenter stated that it agreed with the Commission's
analysis of the benefits of making fees and rebates determinable at
time of execution.\1511\ Another commenter stated that it agreed with
the Commission's assessment ``. . . of how existing exchange pricing
tier models can negatively impact market participants behavior.''
\1512\ A third commenter stated that the Rule will ``help to make
overall trading costs more transparent.'' \1513\
---------------------------------------------------------------------------
\1510\ See section IV.E for a discussion of the comment file.
\1511\ See Council of Institutional Investors Letter at 4.
\1512\ See BMO Letter at 4.
\1513\ See Retirement Coalition Letter at 2.
---------------------------------------------------------------------------
Multiple commenters pointed out the potential for exchanges'
pricing models to create a conflict of interest for broker-dealers who
route multiple customers' orders to the exchanges.\1514\ One commenter
stated that the benefits of pricing determinable at time of execution
include ``help[ing] broker-dealers make better order routing
decisions,'' and reducing order routing incentives based ``on achieving
a threshold to gain a specific fee or rebate.'' \1515\ Another
commenter also stated that exchanges have little incentive to make fees
and rebates determinable at the time of trade because the fee and
rebate structure creates ``captive customers'' that direct order flow
to a given exchange in hopes of receiving a given fee or rebate tier in
a given month.\1516\ One commenter agreed that fees determinable at
time of execution has ``the potential'' to facilitate pass-through of
fees and rebates to broker-dealers' customers, and thereby alleviate
concerns about ``perceived'' conflicts-of-interest, but characterized
such concerns about conflicts-of-interest as ``misplaced.'' \1517\
---------------------------------------------------------------------------
\1514\ See Vanguard Letter at 6 (``These pricing models can
create conflicts of interest with a broker's obligation to obtain
best execution for a customer . . .''); STA Letter at 7 (``Today,
the primary concerns on access fees are how they contribute to the
maker/taker or taker/maker pricing models offered by exchanges and
the offshoots of conflicts of interests in the routing of customer
order flow by broker dealers.''); Retirement Coalition Letter at 2
``the use of rebates creates conflicts of interest, because when an
institutional order is sent as a displayed order, the potential for
a rebate may influence where a broker sends the order, even when the
investor could receive a better execution on another market.''); CII
Letter at 3 (``The existing system disadvantages institutional
investors because we believe rebates create the kinds of conflicts
of interest identified in our policy.'').
\1515\ BMO Letter at 4.
\1516\ See BMO Letter at 4.
\1517\ See Nasdaq Letter I at 32.
---------------------------------------------------------------------------
As discussed in the Proposing Release, access fees create potential
conflicts of interest between brokers and end customers to the extent
that brokers can route orders to exchanges with worse execution quality
for end customers but more advantageous fees (i.e., a low fee or a high
rebate) for the brokers, which the brokers do not pass on to end
customers.\1518\ For example, a broker may route a customer's limit
order to an exchange with a high rebate for liquidity provision, but a
relatively low fill rate. The end result would be a high rebate payment
for the broker but potentially poor execution quality for the customer.
---------------------------------------------------------------------------
\1518\ See Proposing Release, supra note 11, at 80330.
---------------------------------------------------------------------------
One commenter stated that the supposition that rebates present
[[Page 81742]]
conflicts of interest is not supported with evidence.\1519\ There are,
however, significant reasons to believe that rebates present a conflict
of interest to agency brokers, even if there is uncertainty regarding
to what degree those potential conflicts of interest are being acted
upon. Namely, as described above, the quality of the execution accrues
to the customer while the rebate accrues to the broker, which leads to
a clear divergence of interests whenever the best rebate and the
highest quality execution opportunities differ.
---------------------------------------------------------------------------
\1519\ See Nasdaq Letter I at 2.
---------------------------------------------------------------------------
Having fees and rebates determinable at the time of execution will
mitigate these potential conflicts of interest by increasing broker-
dealer accountability to their customers.\1520\ This is because the
broker-dealer will be able to identify which fees and rebates are
associated with which customer order, which at present is not possible
at the time of execution.\1521\ Having information about the fees and
rebates paid as the order is filled will also improve a customer's
ability to negotiate routing behavior and monitor the effects that fees
and rebates have on its broker's order routing decisions and execution
quality.\1522\
---------------------------------------------------------------------------
\1520\ See Proposing Release, supra note 11, at 80330.
\1521\ See supra note 1092 and surrounding discussion on
information that customers can request from broker-dealers on net
transaction fees and rebates through Rule 606(b)(3).
\1522\ One commenter agreed with the Commission's statement in
the Proposing Release that fees being determinable only at the end
of the month, as they are currently, impedes investors' ability to
evaluate best execution and order routing. Council of Institutional
Investors Letter at 4. The Commission believes that making fees
determinable at time of execution will help investors make these
evaluations, which can contribute to these discussions of fees with
their broker-dealers.
---------------------------------------------------------------------------
In addition, fees and rebates being determinable at the time of
execution can make it easier for broker-dealers to pass the actual fees
and rebates on to the end customer.\1523\ Currently, it can be
difficult for a broker-dealer to pass on fees and rebates to individual
customers because the exchange fee and rebate pricing tier into which a
broker-dealer falls, which ultimately determines fees and rebates on an
individual trade, is typically based on the broker-dealer's relative
activity across the concurrent month and not an individual trade.\1524\
With fees and rebates known at the time of execution, it could be
possible for a broker-dealer to more quickly and easily determine the
amount to be passed back to the customer. To the extent the amendments
increase the proportion of exchange fees and rebates that broker-
dealers pass through to end customers,\1525\ this will benefit
investors and also will reduce the potential benefits broker-dealers
may receive from routing customer orders to exchanges with lower fees
or higher rebates and thereby reduce distortions in customer order
execution quality that this may cause.
---------------------------------------------------------------------------
\1523\ The Proposing Release discussed how the inability to know
the fee or rebate at the time of a trade could render it difficult
for a broker-dealer to pass on fees and rebates to customers to help
avoid a potential conflict of interest. See Proposing Release, supra
note 11, at 80329. See also Council of Institutional Investors
Letter at 4-5.
\1524\ See supra note 1084 and surrounding discussion on the
current practice of volume-based fee tiers. While the Commission has
described the tiered structure of many exchange fee schedules, the
benefits of fees and rebates being determinable at time of execution
do not depend on, or result from, the fee schedules using volume
tiers. The amendments do not ban volume tiers; exchanges can
continue to offer volume tiers as long as the tier is based on
past--rather than future--volume. Likewise, the benefits of
determinability apply even if volume tiers did not exist. For
example, some exchanges offer incentives to market makers for
frequent quoting at the NBBO--the amendments require that such
incentives be based on past quoting at the NBBO so that market
participants could determine with certainty their fee at the time of
execution.
\1525\ See supra note 1087 and surrounding discussion on the
current practice of broker-dealers passing fees and rebates through
to customers.
---------------------------------------------------------------------------
A commenter agreed with the Commission's assessment in the
Proposing Release that the ability of institutional investors and other
market participants to evaluate order execution and routing is
significantly impeded by a lack of determinability.\1526\ A lack of
determinability reduces the amount of information that a market
participant can use when evaluating order execution and routing
decisions. Without the fees and rebates being determinable, broker-
dealers may have difficulty transmitting information about fees and
rebates to customers--the broker-dealer could not commit to a fee or
rebate at execution, but would rather need to explain that uncertainty
regarding fees and rebates could not be resolved until the end of the
month--which may impede competition among broker-dealers.\1527\ In
contrast, under the adopted amendments, it will be possible for broker-
dealers to relay such information about fees and rebates incurred by
the broker-dealer to the customer at the time of execution.\1528\ This
will make the information more usable for customers such as
institutional investors, increasing their incentives to ask for such
information, as well as increasing the ability of the broker-dealer to
transmit the information in a timely manner.
---------------------------------------------------------------------------
\1526\ See Council of Institutional Investors Letter at 4-5.
\1527\ See Proposing Release, supra note 11, at 80336.
Currently, customers' lack of timely information and certainty about
the fees and rebates they incur on the execution of a trade can
impact their choice of a broker-dealer for that trade.
\1528\ For example, under the adopted amendments, it might be
possible to include such information in a report to the investor
following the execution of their order.
---------------------------------------------------------------------------
One commenter stated that, ``[f]ew brokers route directly to the
exchanges . . . Rather, most brokers pay to route their orders through
larger `[direct market access] DMA' brokers to gain the benefit of the
large brokers' exchange fee tiers. While the Proposal would simplify
life for those few large DMA brokers and proprietary trading firms who
closely track where they fall on exchange fee schedules, it wouldn't
directly help the referenced Market Participants.'' \1529\ The same
commenter stated that most Market Participants, ``judging by common
practice today,'' would be unable to account for fees, and the
requirements would not improve transparency for off-exchange trading
where venues ``are not required to charge standard fees, and where fees
are often held as competitively sensitive secrets.'' \1530\ The
Commission agrees that under common practice today it can be difficult
for most Market Participants to account for fees that are not
determinable or known at the time of execution. That said, having
exchange fees and rebates determinable at the time of execution will
make it more likely that larger DMA brokers pass exchange fees and
rebates on to their customers, including when these customers are small
brokers routing their orders through them. That is because customers
can better discuss fees and rebates with large DMA brokers, and the
information will be more useful to customers because it is more timely.
Also, while the requirements only apply to exchange fees and rebates,
exchange and off-exchange trading venues compete, and transparency in
exchange fees and rebates could prompt demand for greater transparency
in off-exchange trading fees.\1531\
---------------------------------------------------------------------------
\1529\ See Pragma Letter at 8.
\1530\ See id. at 8 (``Even if fees are determinable, it will
provide little practical transparency for most Market
Participants.'').
\1531\ See infra section VII.E.2 for additional discussion of
the competitive effects of these amendments.
---------------------------------------------------------------------------
One commenter stated that rebate tiers increase aggregate
liquidity, and that fee and rebate determinability will, ``disrupt
existing economic incentives,'' and, ``negatively impact exchange
liquidity provision and drive even more liquidity to off-exchange
venues.'' \1532\ Another commenter stated that this rule ``disrupts
existing economic incentives without justification,'' \1533\ and added
[[Page 81743]]
that they ``believe there is more aggregate liquidity in the
marketplace because of the incentives provided by exchange rebate
tiers.'' On the other hand, another commenter stated that when pricing
is not determined until the end of the month, a ``captive customer'' is
created, who ``. . . must maintain levels of qualified trading activity
or suffer an adverse economic consequence for up to an entire month's
trading activity.'' \1534\
---------------------------------------------------------------------------
\1532\ See Cboe Letter II at 9-10.
\1533\ See Cboe Letter I at 9.
\1534\ BMO Letter at 4.
---------------------------------------------------------------------------
The Commission disagrees that fee and rebate determinability are
likely to alter the economic effects related to fee and rebate tiers.
Many of the incentives created by current exchange pricing schedules
can be implemented by creating tier-based pricing schedules that are
conditioned on historic (as opposed to future) activity, and such
pricing would continue to be permissible under the amended rules.\1535\
For example, a fee schedule might base the current fee or rebate tier
on the share that the exchange member had of the exchange's total
volume (or total consolidated volume) in the previous month.\1536\ The
incentives for meeting a volume tier would remain but the benefits of
achieving the tier would be realized in the following month.
Alternatively, a fee schedule might be based on the current month's
absolute volume on the exchange (as opposed to share of volume), up
until the moment of execution, which the exchange member would
presumably know.\1537\ Again, the incentive would be approximately the
same. In general, an incentive that is based on some future quantity
that cannot be known with certainty today could likely be replicated by
offering the certainty equivalent,\1538\ which can be calculated using
a current or past quantity that can be known with certainty. This means
that the uncertain portion of current fees is not strictly necessary to
provide incentives. In this respect, any costs and benefits associated
with volume-based fee tiering are not expected to change as a
consequence of requiring fees and rebates to be determinable at the
time of execution. Therefore, the benefits of the requirements that
exchange fees and rebates be determinable at the time of execution will
likely not include alleviating such ``captive'' customers.
---------------------------------------------------------------------------
\1535\ Such pricing can have not just benefits, but also costs.
Tier-based volume pricing, for example, is used to incentivize the
concentration of order flow--i.e., a member is incentivized to route
orders to a particular exchange in order to qualify for a better
pricing tier. This in turn creates a potential conflict of interest
because the exchange member is incentivized to route customer order
flow to the exchange for the purposes of tier qualification rather
than maximizing other aspects of execution quality.
\1536\ Volume discounts like this, which are based on previous
volume and then provide discounts on future purchases, have
parallels in other industries (e.g., loyalty reward programs).
Relative to the baseline, such a schedule would incentivize a
customer to stay with an exchange for an additional month. However,
there are ways exchanges might alleviate these concerns, such as a
fee schedule that would induce a switch from one exchange to
another.
\1537\ Exchanges could also create alternative incentive
programs for new members provided these programs are comply with the
requirements of the Exchange Act.
\1538\ A ``certainty equivalent'' is a term of art in economics,
referring to the amount of a certain (that is, nonrandom) payment
that must be given to an economic agent so that the agent would be
indifferent between this payment and some random payoff.
---------------------------------------------------------------------------
One commenter stated that requiring fees to be knowable at the time
of execution would make ``participation in exchanges' growth programs
more expensive in the initial month of participation.'' \1539\ The
Commission acknowledges that a new broker-dealer will not have a
history of trading and therefore a tier schedule based on historical
trading could not be implemented until the new broker-dealer has
established a history; however, this history need not be long, and
exchanges can cater to new and small broker-dealers by constructing
tier schedules based on a short history.
---------------------------------------------------------------------------
\1539\ See Nasdaq Letter I at 33.
---------------------------------------------------------------------------
One commenter stated that: ``[R]equiring market participants to
calculate their activity from the prior period in order to determine
the volume fee and adjust their financial plans accordingly adds an
unnecessary layer of complexity. The amount of effort required to
understand the volume fee system, forecast volume fees for an upcoming
period, and confirm that fees are indeed being calculated appropriately
will especially disadvantage smaller brokers, who typically have less
resources at their disposal for needless work such as this.'' \1540\
---------------------------------------------------------------------------
\1540\ See Virtu Letter II at 11-12.
---------------------------------------------------------------------------
The Commission disagrees that the rule will add complexity. The
rule removes the need for exchange members to perform forecasts in
order to determine what fee they might be required to pay in a given
moment. This is because, in order for a fee to be known at the time of
execution as the amendments require, the fee cannot be based on
activity that will happen after the execution. The Commission
acknowledges, however, that fee schedules may remain complex. The rule
however does not require more from small brokers than is required
currently, namely it does not require them to understand the fee volume
system, forecast volume fees (indeed it eliminates the need for
forecasting), or to confirm that fees are being calculated
appropriately. Thus the Commission does not expect this rule to
disadvantage smaller brokers, and, to the extent that forecasting
future market outcomes is more difficult for small brokers, this rule
may make it easier for small brokers to compete.\1541\
---------------------------------------------------------------------------
\1541\ See section VII.E.2.c for a discussion of the effect that
fee determinability may have on competition.
---------------------------------------------------------------------------
Another commenter stated that determinable fees ``would also limit
exchanges' ability to incent market makers and other participants to
quote at the NBBO and to do so in a large number of securities,
including thinly-traded securities.'' \1542\ The Commission disagrees
because exchanges can continue to offer incentives based on past
quoting at the NBBO in a large number of securities, including thinly-
traded securities. Meaningful thresholds for pricing based on NBBO
quoting activity can still be set based on historic activity, so that
incentives to quote at the NBBO are expected to persist.\1543\
---------------------------------------------------------------------------
\1542\ See Nasdaq Letter I at 33.
\1543\ For example, one exchange offers additional rebates to
qualified market makers if, among other things, they quote at the
NBBO at least 50% of the time during the month in an average of at
least 2,700 symbols per day. See Nasdaq Stock Mkt. LLC, Equity 7,
Sec. 114, available at https://listingcenter.nasdaq.com/rulebook/nasdaq/rules/nasdaq-equity-7-section_114_market_quality_incentive_programs. Similar terms can be
offered based on historic (rather than future) quoting.
---------------------------------------------------------------------------
Another commenter stated that, ``in order to achieve th[e] stated
regulatory objective'' of certainty as to an order's net fee and rebate
price, and helping broker-dealers make better order routing decisions,
``fees and rebates would have to be known prior to the time of
execution (instead of at the point of execution, where the fee could
vary based on the type of order being accessed).'' \1544\ The
Commission acknowledges that fees can vary based on order type--for
example, removing hidden liquidity may incur a different fee than
removing displayed liquidity, and the broker-dealer may not know
whether the order will execute against hidden liquidity prior to the
time of execution. In this case, there are two sources of potential
uncertainty if fees are not determinable at execution: the broker-
dealer's ultimate position on the exchange's fee schedule, and the type
of liquidity that is accessed. Fee determinability allows broker-
dealers to know, prior to execution, what their fee will be for each
type of liquidity that might be accessed; the type of liquidity that is
accessed may not be known until
[[Page 81744]]
the time of execution. Resolving one source of uncertainty prior to
execution (the broker-dealer's position on the exchange's fee schedule)
can help broker-dealers make better routing decisions even if there
remains uncertainty along other dimensions (e.g., the presence of
hidden orders).
---------------------------------------------------------------------------
\1544\ See Citadel Letter I at 25.
---------------------------------------------------------------------------
4. Acceleration and Implementation of the MDI Rules and Addition of
Information About Best Odd-Lot Orders
The MDI Rules were designed to increase transparency into, among
other things, the best priced quotations available in the market.\1545\
The MDI Rules expanded NMS data and established a decentralized
consolidation model, pursuant to which competing consolidators will
eventually replace the exclusive SIPs for the collection,
consolidation, and dissemination of NMS data.\1546\ As discussed in
section V.A, the Commission adopted a phased transition plan for the
MDI Rules,\1547\ which has been delayed.\1548\ Because the MDI Rules
are not yet implemented, information about odd-lot orders in NMS stocks
is only available on individual exchange proprietary data feeds, and
market participants interested in quotation information for individual
odd-lot orders must purchase these proprietary feeds.\1549\ Due to the
delays in the MDI Rules' implementation, as discussed in the Proposing
Release,\1550\ the Commission is adopting an accelerated implementation
schedule, with some modifications from the proposal, so that market
participants, including investors, will be provided with the enhanced
transparency benefits earlier than anticipated in the MDI Rules.
---------------------------------------------------------------------------
\1545\ MDI Adopting Release, supra note 10, at 18601-02, 18617;
see also 17 CFR 242.600(b)(82).
\1546\ See MDI Adopting Release, supra note 10.
\1547\ See Proposing Release, supra note 11, at 80295
(describing the phased transition plan for the MDI Rules).
\1548\ See supra notes 74-78 and accompanying text.
\1549\ See MDI Adopting Release, supra note 10, at 18599. SIP
data includes odd-lot transaction information but does not include
odd-lot quotation information, except to the extent that odd-lot
orders are aggregated into round lots pursuant to exchange rules;
see also MDI Proposing Release, supra note 744, at 16739; MDI
Adopting Release, supra note 10, at 18727. SIP data includes odd-lot
transaction information but does not include odd-lot quotation
information, except to the extent that odd-lot orders are aggregated
into round lots pursuant to exchange rules; see also MDI Proposing
Release, supra note 744, at 16739; MDI Adopting Release, supra note
10, at 18727.
\1550\ See Proposing Release, supra note 11, at 80295.
---------------------------------------------------------------------------
The amendments will result in four changes to NMS data. Two of the
changes will accelerate the implementation of specific aspects of MDI,
namely the round lot definition and the inclusion of odd-lot quotations
priced better than the NBBO in NMS data. This acceleration will result
in realizing the economic effects of these MDI Rules sooner. The
Commission acknowledges that the economic effects of the acceleration
will be temporary, only lasting until the accelerated aspects of the
MDI Rules would otherwise have been implemented.\1551\ The amendments
will also impose a new requirement on the exclusive SIPs to disseminate
the accelerated odd-lot information until the exclusive SIPs are
retired, the effect of which is to result in the odd-lot information
being disseminated sooner.\1552\ The amendments, however, present the
possibility that the new requirements on the SIPs can reduce competing
consolidator competition if the additional requirements dissuade some
market participants from choosing to become competing consolidators,
which could reduce the expected benefits of the MDI Rules.\1553\ The
amendments will also require the dissemination of a standardized best
odd-lot order or BOLO. The primary economic effect of this requirement
will be to provide an additional standard benchmark that market
participants could use to gauge execution quality--particularly for
smaller or odd-lot orders.\1554\
---------------------------------------------------------------------------
\1551\ See supra section V.E for further discussion on the MDI
Rules Implementation.
\1552\ See infra section VII.E.2.c for additional discussion of
this effect. While the Rule requires the exclusive SIPs to
distribute odd-lot data, the MDI Rules do not require the competing
consolidators to disseminate odd-lot data. However, the MDI Adopting
Release anticipated that at least one competing consolidator will do
so because there would be demand for the data. See supra section
VII.C.3.
\1553\ See infra section VII.E.2.c for additional discussion of
MDI acceleration and the potential effect on competition between
competing consolidators. Requiring the SIPs to disseminate odd-lot
information may make the SIPs more likely to become competing
consolidators and give them a first-mover advantage over other
competing consolidators. However, this advantage is likely to be
limited because competing consolidators can offer a lower latency
than the SIPs currently provide, and can offer depth-of-book data in
addition to odd-lot information.
\1554\ See infra section VII.D.6.a.ii for additional discussion
on how the BOLO will make it easier for market centers and broker-
dealers to compute statistics on price improvement relative to the
best available displayed price, now required by amended Rule 605.
---------------------------------------------------------------------------
Commenters questioned the sufficiency of the proposed 90-day
implementation timeline for the acceleration of MDI Rules and the
addition of the BOLO to NMS data. One commenter discussed the need for
downstream programming changes for a variety of market participants:
SIPs, recipients of market data, and broker-dealers.\1555\ This
commenter stated that implementation is likely to take over a year.
Several other commenters stated that implementation will take longer
than 90 days.\1556\ One commenter discussed changes required for the
SIPs: ``The changes required for SIPs are relatively straightforward
from a conceptual perspective but will require significant undertakings
before they can be implemented.'' This commenter then discussed the
necessary steps, including: notice and lead time that data providers
must give their clients; product decisions vendors need to make;
development and testing for exchanges, vendors, and subscribers;
traffic and capacity decisions given the change in message traffic; and
communication and education for end users.\1557\ Another commenter
mentioned costs from converting round lots to actual share size in
processing and display systems.\1558\ The Operating Committee of the
CTA-UTP Plans commented on the need for system design, equipment
procurement, and industry testing.\1559\ This commenter referenced a
2022 Odd Lots Proposal by the SIPs, which estimated a 10-12 month time
frame for providing only top-of-book odd-lot quotations by the SIPs;
given the additional requirements in the Proposal, the commenter
estimated that implementation would take more than 12 months.
---------------------------------------------------------------------------
\1555\ See NYSE Letter I at 7.
\1556\ See FISD Letter at 3, Cboe Letter II at 11, and BlackRock
Letter at 12.
\1557\ See FISD Letter at 2-3.
\1558\ See Cboe Letter II at 11. The commenter did not provide
an estimate of these costs, only stating that, ``round lot
conversion to actual share size will likely require considerable
work by industry participants to their processing and display
systems.''
\1559\ See CTA-UTP Letter at 1.
---------------------------------------------------------------------------
In light of these comments, the Commission is modifying the
proposed compliance date for the round lot and odd-lot information
definitions to extend the time for compliance. The Commission
acknowledges that, all else equal, a shorter implementation timeline
may result in greater total costs for the acceleration of the MDI
Rules. Consistent with what commenters suggested was necessary for
systems changes and testing among a variety of market participants, the
adopted amendments extend the proposed compliance date for round lot
definitions to approximately 12 months after the effective date of the
amendments; for odd-lot information, the compliance date is extended to
[[Page 81745]]
approximately 18 months after the effective date.\1560\
---------------------------------------------------------------------------
\1560\ See section VI.C.
---------------------------------------------------------------------------
The 12-month implementation timeline for the round lot definition
aligns with the compliance date for the amendments to Rule 612,
allowing for the amended minimum pricing increments and new round lots
to begin concurrently. This concurrence is expected to reduce the
potential for operational risks and investor confusion from changing
systems separately for both the amendments to Rule 612 and the
acceleration of the MDI Rules, and reduces operational risks from a
compressed timeline. Further, to the extent that commenters
specifically discussed the timeline needed for the round lot definition
implementation (as opposed to the timeline needed for both the round
lot and odd-lot information definitions), no commenter stated that the
round lot definition would require more than 12 months.
The longer 18-month implementation timeline for the odd-lot
information definition is commensurate with the added complexity of
disseminating new data fields for odd-lot quotations at multiple levels
inside the NBBO. Additionally, the 18 month timeline is broadly
consistent with the 10-12 month timeline that the SIPs estimated to be
necessary for the 2022 Odd Lots Proposal; the longer timeline for the
odd-lot information in these amendments allows for the fact that odd-
lot information in these amendments includes quotations at each price
level inside the NBBO, whereas the 2022 Odd Lots Proposal only included
top-of-book odd-lot quotations. Further, the SIPs have been discussing
the addition of odd-lot quotation information to the SIP data for
several years and should be able to make the necessary adjustments to
their processors in the adopted timeline. Finally, no commenter stated
that the odd-lot information definition would require more than 18
months to implement. The adopted timeline allows market participants
more time for each of the steps required for implementation than was
initially proposed, which will address the concerns raised by
commenters and help market participants to complete the tasks in a
cost-effective manner. The compliance costs discussed in section
VII.D.5.c reflect costs associated with the adopted timeline.
a. Round Lot Definition
As discussed in the Proposing Release,\1561\ the round lot
definition in the MDI Rules will result in numerous economic effects,
and the amendments will result in realizing these effects sooner. The
primary effects stem from the MDI Rules round lot definition shrinking
the NBBO for stocks priced greater than $250.\1562\ Other effects of
changing the round lot definition include increased transparency and
better order execution,\1563\ as well as any effects from potentially
having more orders routed to exchanges instead of ATSs.\1564\ The costs
of changing the round lot definition derive from upgrading systems to
account for additional message traffic and modifying and reprogramming
systems.\1565\ The Commission also expects that changing the round lot
definition will impact the mechanics of other rules and
regulations.\1566\ These economic effects will be realized earlier than
is currently estimated under the existing MDI timeline because this
portion of the MDI Rules is currently not set to be implemented until
the end of the implementation timeline for the MDI Rules. Further,
because the first steps of the timeline for the MDI Rules have not been
accomplished,\1567\ and the Commission is uncertain when exactly the
round lot definition otherwise will be implemented, the degree of the
effect of the acceleration is unknown.\1568\
---------------------------------------------------------------------------
\1561\ See Proposing Release, supra note 11, at 80330.
\1562\ See MDI Adopting Release, supra note 10, section
V.C.1(b)(i), for the full discussion of the effect of changing the
round lot size on the NBBO.
\1563\ See MDI Adopting Release, supra note 10, at 18744-47 for
the full discussion of the effect of changing the round lot size on
transparency and execution quality. See FIA PTG Letter II at 4 for
agreement from a commenter.
\1564\ See MDI Adopting Release, supra note 10, at 18747 for the
full discussion of the effect of changing the round lot size on
exchange competition and order routing.
\1565\ See MDI Adopting Release, supra note 10, at 18748 for the
full discussion of the expected costs of changing the round lot
size. See also infra section VII.D.5 for an estimation and
discussion of these compliance costs as they pertain to the proposed
acceleration.
\1566\ See MDI Adopting Release, supra note 10, at 18749 for the
full discussion of the effect of changing the round lot size on
other rules and regulations. The round lot definition will
mechanically tighten the NBBO, which is used as a reference price
for numerous rules. The reference prices used for the Short Sale
Circuit Breaker and LULD Plan will be affected, though these Rules
will continue to function consistent with their stated purposes. The
NBBO is also used as a benchmark for SRO rules such as RLPs,
exchange market maker obligations, and for some order types;
exchanges could propose rule changes to maintain the current
operation of these rules. Finally, the round lot definition could
increase the benefits of 606(b)(3) reports because it could result
in an increase in the number of indications of interest in higher
priced stocks that will be required to be included in 606(b)(3)
reports.
\1567\ See supra notes 74-78 and accompanying text section for a
discussion of the delays.
\1568\ See supra section V.B.1 for a discussion of the factors
that affect when MDI will be implemented and a discussion of an
estimate of the proposed acceleration of at least two years after
the Commission's approval of the plan amendment(s) required by rule
614(e).
---------------------------------------------------------------------------
The Commission recognizes that the earlier implementation of the
round lot definition could affect the tiered tick structure by sooner
increasing the number of stocks subject to a minimum pricing increment
of less than $0.01, but the Commission does not expect this effect to
be substantial. Specifically, a mechanically tighter NBBO will reduce
the time weighted average quoted spread used to determine the
appropriate tick increment for stocks priced greater than $250.
However, higher-priced stocks also tend to have higher spreads that are
unlikely to narrow enough for the amendments to result in a smaller
minimum pricing increment.\1569\
---------------------------------------------------------------------------
\1569\ In the MDI Rules the Commission estimated an average
reduction in quoted spreads, conditional on the round lot definition
resulting in a reduction of roughly 15% for stocks priced $250-
$1,000 and 28% for stocks priced $1,000-$10,000. Given the average
quoted spread of $0.35 for stocks priced $250-1,000 and $2.90 for
stocks priced $1,000-$10,000 the expected mechanical reductions are
likely not sufficient to reduce the spreads of many of these stocks
to the point where they would qualify for a lower tick size in this
proposal. See MDI Adopting Release, supra note 10, at 18743.
---------------------------------------------------------------------------
As discussed in the Proposing Release,\1570\ the Commission also
recognizes that both the reduction in tick size and accelerating the
definition of round lot will reduce the depth of liquidity at the NBBO.
These effects might amplify each other in a small set of stocks. A
reduction in tick size will spread liquidity across more price levels,
while the implementation of the round lot definition will result in
displaying smaller quotes at the NBBO. The amendments could result in
this effect being amplified for stocks that trade above $250 with
spreads narrower than $0.015 as these stocks will receive both smaller
tick and smaller round lot sizes. The number of such affected stocks is
likely very small.\1571\ The reduction in depth at the NBBO will
temporarily reduce the information about liquidity available in the
market for market participants who rely on public data feeds. However,
the
[[Page 81746]]
eventual inclusion of the depth of book information in consolidated
market data under the MDI Rules, once implemented, will render this
effect temporary. At that point in time, consolidated market data is
expected to contain depth information at more price points, which will
largely counteract the effects of a reduction in displayed depth from
the implementation of the round lot definition and even from a
reduction in tick size.
---------------------------------------------------------------------------
\1570\ See generally, Proposing Release, supra note 11.
\1571\ See section V.B.3.b for analysis identifying such stocks.
In particular, see supra note 801 identifying only two stocks--both
highly liquid--that would have qualified for both the tick reduction
and a reduction in the round lot as of Nov. 30, 2023.
---------------------------------------------------------------------------
Multiple commenters further discussed the potential interaction of
the reduction in tick size and the MDI round lot definition. One
commenter stated that the resulting reduction in depth at the NBBO
would make the NBBO less relevant and subject to more instability:
``With a lower notional value earning protected status at the NBBO,
even accessing 100 shares of liquidity would likely move a stock in one
of the new tiers' multiple price levels. Furthermore, these smaller
notional amounts reduce the risk taken to queue jump displayed orders
by placing slightly more aggressively priced orders ahead of them.''
\1572\ A separate commenter stated that the reduced liquidity at the
NBBO will require investors executing large orders to, ``sweep across
multiple market centers, exposing them to greater execution risk.
Together, these changes would reduce the depth at the NBBO, leaving it
subject to greater volatility and, in turn, reducing reliability and
execution quality for retail investors.'' \1573\
---------------------------------------------------------------------------
\1572\ See RBC Letter at 5. Queue jumping in this context is
synonymous with pennying. See supra note 994.
\1573\ See Virtu Letter II at 8.
---------------------------------------------------------------------------
The Commission acknowledges that these are possibilities, but the
interaction of the reduction in tick size and the MDI round lot
definition is not expected to have a material impact on the NBBO of
affected stocks. In order for a stock to be impacted by both the new
round lot categories and the smaller tick size, it would need to have a
price over $250 with a spread below $0.015; this implies that the
percentage spread must be below 0.006%.\1574\ To put this in
perspective, consider the sample of all stock-days in 2023.\1575\ For
each stock-day, divide its TWAQS by its price to measure its percentage
spread. Only 1% of this sample has a percentage spread below 0.015%--
i.e., the first percentile of the sample's percentage spread is 2.5
times higher than the percentage spread of the stocks affected by both
the new round lot categories and the smaller tick size.\1576\ This
implies that the affected stocks are exceptionally liquid--they are
well within the first (i.e., most liquid) percentile when liquidity is
measured using percentage spread. The exceptional liquidity of the
affected stocks will likely protect their NBBO from material
deterioration.
---------------------------------------------------------------------------
\1574\ The percentage spread measures the cost of liquidity,
measured here as the spread, as a fraction of the execution price. A
lower percentage spread indicates that transaction costs for
liquidity demanders are a smaller fraction of the execution price.
Here, $0.015/$250 = 0.00006 = 0.006%.
\1575\ A symbol-day is the unique pair of a stock symbol and a
date. For example, one observation is AMZN on December 7, 2023; a
second observation is AMZN on December 8, 2023; a third is AAPL on
December 7, 2023, etc.
\1576\ This statistic is computed using all symbol-days in WRDS
intra-day indicators for the year 2023. The sample has 2.3 million
observations.
---------------------------------------------------------------------------
The Commission does acknowledge that the amendments will increase
the likelihood that a 100-share order will walk the book for stocks
affected by both a reduction in the tick and the round lot definition;
\1577\ however, the high price (over $250 per share) of the affected
stocks implies that the notional amount of such an order will be
substantially larger than the notional amount of a 100-share order for
a typical stock unaffected by the MDI Rules round lots. Therefore, a
round lot under the new definition will continue to reflect a
meaningful notional amount even with a reduced number of shares,\1578\
thereby ensuring that regulatory protections for round lots--such as
those governing the display, dissemination, and protection of orders
under Rules 602, 604, and 611--continue to focus on orders of
significant size. Likewise, the amendments will increase the likelihood
that particularly large orders may need to sweep across multiple market
centers; however, fixing the notional amount of an order, a high-priced
stock will require fewer shares, which reduces the need to sweep across
multiple market centers. Therefore, the amendments will continue to
protect a meaningful notional amount at the NBBO after the round lot
size is reduced for these high-priced stocks. Similarly, the amendments
will make it incrementally easier for traders to queue jump--i.e.,
penny--protected orders in affected stocks, but pennying will remain
difficult due to the relatively high price of the affected stocks. That
is, a trader would still need to commit a relatively high notional
amount to jump the queue with a protected order. Finally, the concern
with pennying is that a market participant can jump the queue by
posting economically trivial price improvement; for the stocks affected
by both the tick reduction and the round lot definition, however,
posting a protected order that improves on the NBBO is likely to
provide meaningful price improvement. To jump the queue with a
protected order, a trader would need to post an order that is: (1)
priced better than existing orders by $0.005 (which is large relative
the stock's typical spread of under $0.015), and (2) with a $10,000
notional value \1579\ (which is larger than the notional value required
to queue jump for stocks priced under $100). Such an order would
thereby require the trader to offer price improvement that is
economically large relative to the costs of trading the stock, and
relative to the notional amount required to jump the queue in most
other stocks. Therefore, the Commission continues to expect that the
acceleration of the round lot definition will protect a meaningful
amount of liquidity at the NBBO for stocks receiving the tick
reduction.\1580\
---------------------------------------------------------------------------
\1577\ A marketable order ``walks the book'' if the size of the
order is larger than the amount of liquidity available at the best
price at a market center; the order must therefore execute against
liquidity at multiple price points within the limit order book.
\1578\ For example, the notional amount reflected by a round lot
of 40 shares will generally be greater than $10,000 (this is because
a stock's price must generally be greater than $250 to be assigned a
round lot of 40 shares, and 40*$250 = $10,000). If a stock has a
round lot of 100 shares, a round lot will only reflect $10,000 of
notional if the stock price is at least $100 (so that 100*$100 =
$10,000); many stocks with a round lot of 100 shares do not have a
price greater than $100. Therefore, the lower round lot for high-
priced stocks will continue to reflect a notional amount that is at
least as high as the notional reflected in round lots for stocks
with more common prices.
\1579\ The new round lot definitions are structured so that they
protect $10,000 of notional value. See table 1 of MDI Adopting
Release, supra note 10.
\1580\ For example, suppose a trader wants to incrementally
improve the NBBO by jumping ahead of a resting protected order. If
the stock has a tick of $0.005 and a round lot of 40 shares, then
the trader must improve the price by $0.005 and post an order with a
notional value of at least $10,000, see id. The cost of queue-
jumping is therefore at least $50 (0.005*$10,000). Now consider a
stock with a tick of $0.01, a round lot of 100 shares, and a price
of $30 per share. The cost of queue-jumping for this stock is only
$30 (0.01*100*$30). Therefore, stocks that receive both a tick
reduction and a reduced round lot under these amendments are not
expected to experience more queue jumping--and consequent
deterioration of the NBBO--compared to stocks that retain the penny
tick and the 100-share round lot.
---------------------------------------------------------------------------
One commenter encouraged the Commission to ``comprehensively review
its proposed changes to tick sizes, access fees and round lots to
better evaluate how these changes together would impact liquidity.''
\1581\ As discussed in the preceding paragraphs, the Commission expects
that a very small number of stocks (two as of Nov. 30, 2023 \1582\)
would be
[[Page 81747]]
subject to both a change in round lot size and tick size because very
few stocks have both a price above $250 (to qualify for a reduced round
lot) and a TWAQS below $0.015 (to qualify for the tick reduction). The
exceptional liquidity of the stocks with both these characteristics is
unlikely to be materially affected by the interaction of the tick
reduction and the reduction in the round lot. With respect to the
reduction in the access fee cap, the Commission expects effects of that
change to be independent of the effects of the round lot definition for
two reasons. First, the reduction in the access fee cap is unlikely to
affect a stock's round lot size. This is because the reduction in the
access fee cap--to the extent that it affects quoted prices as
discussed in sections VII.B.3 and VII2--is not expected to move quoted
prices by more than one tick. Round lots, on the other hand, are
determined by whether a stock is priced above $250, $1,000, or $10,000;
the probability that the reduction in the access fee cap affects a
stock's round lot assignment is therefore miniscule. Second, the
reduction in the access fee cap and the reduction in the round lot are
expected to have separate but unrelated effects on the NBBO. Stocks
that receive a round lot less than 100 shares are expected to have a
narrower NBBO because the new round lot definition will include quotes
at better prices in core data that were previously excluded from being
reported because they consisted of too few shares.\1583\ Access fees do
not affect the existence of these better priced quotes with fewer
shares. The reduction in the access fee cap is expected to put upward
pressure on quoted spreads and therefore widen the NBBO; \1584\ this
effect operates at a per-share level because fees and rebates are
assessed per-share, making the number of shares in a round lot
irrelevant. Therefore, the Commission does not expect the round lot
definition to interact with the reduction in the access fee cap.
---------------------------------------------------------------------------
\1581\ See UBS Letter at 10.
\1582\ See supra note 801 identifying only two stocks--both
highly liquid--that would have qualified for both the tick reduction
and a reduction in the round lot as of Nov. 30, 2023.
\1583\ See MDI Adopting Release, supra note 10, at 18742.
\1584\ See section VII.B.3 for a discussion of the effect that
fees and rebates have on quoted spreads.
---------------------------------------------------------------------------
For institutions that do not purchase proprietary feeds, the MDI
Rules once implemented will result in the display of five levels of
depth-of-book in NMS market data. To the extent that the amendments
result in liquidity spread out across more price levels due to the
round lot reduction,\1585\ then these changes would reduce the value of
these NMS market data. However, the high price of stocks affected by
the round lot reduction implies that the amount of visible notional
liquidity will remain high relative the notional liquidity visible for
a typical stock unaffected by the MDI round lots.\1586\
---------------------------------------------------------------------------
\1585\ See supra note 1348 and surrounding discussion on the
effect that the tick reduction is expected to have on the value of
MDI NMS market data.
\1586\ See supra note 1578 for an example indicating that the
amount of notional liquidity reflected in a round lot will generally
be higher for stocks receiving a reduction in the round lot size
than stocks that retain a round lot size of 100 shares (due to the
higher price of stocks receiving a round lot reduction).
---------------------------------------------------------------------------
One commenter stated that, ``the Commission failed to note how much
actual volume takes place in any of the three proposed [round lot]
tiers and what challenge, if any, changes to round lot definitions
would address.'' \1587\ In the MDI Rules, the Commission estimated that
approximately 1% of stocks, 3% of share volume, and 30% of dollar
volume will be affected by the new round lot tiers.\1588\ The
Commission also discussed in the MDI Rules the effect of changing the
round lot size on transparency and execution quality.\1589\
---------------------------------------------------------------------------
\1587\ See Tastytrade Letter at 22.
\1588\ See MDI Adopting Release, supra note 10, at 18743 Table
4.
\1589\ See supra note 1563.
---------------------------------------------------------------------------
The commenter further suggested that the implementation of the
round lot definition could cause confusion among retail investors:
``Currently, retail customers, especially those trading options,
understand one option contract represents one hundred shares.
Frequently, it is simply referred to as a `round lot.' Changes in round
lot sizes will most certainly create confusion in this area for retail
investors.'' \1590\
---------------------------------------------------------------------------
\1590\ See Tastytrade Letter at 22.
---------------------------------------------------------------------------
The Commission acknowledges that there may be a learning curve
associated with the new round lot definition. However, as discussed in
the MDI Adopting Release, investor confusion will be temporary for four
reasons. First, market participants already regularly trade in
increments other than 100 shares.\1591\ Second, most NMS stocks will
continue to have a round lot of 100 shares. Third, core data will be
distributed with the size of the NBBO and best quotes in shares rather
than in the number of round lots. Fourth, broker-dealers and other
market participants will modify or develop their systems to
automatically keep track of the round-lot changes.\1592\ Further, any
confusion from the accelerated round lot definition would have occurred
eventually under the original MDI timeline, so the incremental effect
of MDI acceleration on investor confusion is minimal.
---------------------------------------------------------------------------
\1591\ See supra note 791 and surrounding text for a discussion
on the interaction of the new round lot definition and options
trading. It is unlikely that the new round lot definition will
confuse retail investors trading in options, partially because
options markets already have standard contracts on stocks with a
round lot size less than 100 shares.
\1592\ See MDI Adopting Release, supra note 10, at 18745.
---------------------------------------------------------------------------
Other commenters expressed concerns about monthly updates to
stocks' round lots. Each update requires market participants to
``reconfigure their investment platforms and trading systems to make
any modifications effective.'' \1593\ The same commenter pointed out
that, under the proposal, round lots would be assigned at discrepant
intervals from tick assignments. Commenters also stated that these
system updates may increase complexity and operational risk, and
further contribute to investor confusion.\1594\
---------------------------------------------------------------------------
\1593\ See BlackRock Letter at 9-10.
\1594\ See SIFMA Letter II at 34.
---------------------------------------------------------------------------
While any periodic system update can pose a risk of glitches, the
amendments assign round lots and tick sizes on the same schedule--every
six months in May and November. Syncing the updates like this will
reduce costs relative to the monthly round lot updates in the baseline
by reducing the number of times that firms are required to ``open the
hood'' of trading systems. To further reduce these costs and provide
opportunity for industry testing, the adopted amendments incorporate a
one-month gap between evaluation periods and the implementation of
updated round lots and tick sizes. It is possible that the amendments
to the round lot definition--i.e., the less frequent evaluation periods
and the lag between evaluation and implementation--may cause a stock's
round lot to be less reflective of its price than would have otherwise
been the case under the original MDI Rules round lot definition (e.g.,
if a stock's price falls after the evaluation period, it may be
assigned to a round lot that is too low for the next six months). This
imprecision in round lot assignment, however, is unlikely to
significantly reduce the benefits of the MDI Rules for two reasons.
First, to the extent that a stock is assigned a round lot based on
stale information, this assignment will be corrected within six months
at the next evaluation date. Second, given the significant distance
between the round lot thresholds (i.e., $250, $1,000, and $10,000), any
deviation in a stock's round lot as a result of these amendments is
likely to be due to stocks
[[Page 81748]]
that are near a threshold; for these stocks, the cost of being
including in the next smallest or largest tier is likely to be small.
Finally, one commenter suggested alternative price thresholds for
the round lot definition; these thresholds would result in five round
lot tiers.\1595\ The Commission continues to believe, as stated in the
MDI Adopting Release, that a five-tiered approach is unnecessarily
complex, and that the adopted tiers promote a smoother transition to a
price-based round lot structure.\1596\
---------------------------------------------------------------------------
\1595\ See Pragma Letter at 9.
\1596\ See MDI Adopting Release, supra note 10, at 18618. The
price-based round lot structure ensures that there is $10,000 of
notional value protected under the new round lot definitions. See
supra note 1579.
---------------------------------------------------------------------------
b. Including Odd-Lots in NMS Data
As discussed in the Proposing Release,\1597\ the acceleration of
the implementation of the MDI Rules that expand the NMS data to include
odd-lot information inside the NBBO will result in sooner realizing
some, but not all, economic effects of this aspect of the MDI
Rules.\1598\ The odd-lot information could be useful to consumers of
SIP data that could use it to make better inferences about market
conditions, thereby leading to investment decisions that more fully
reflect market conditions and increased market efficiency. This odd-lot
information could also lessen the effect of a reduction in displayed
depth at the NBBO resulting from either a smaller tick size or a
smaller round lot. Specifically, expediting inclusion of odd-lot data
will allow individual investors whose broker-dealers subscribe to the
data to visually monitor the market sooner than they would
otherwise.\1599\
---------------------------------------------------------------------------
\1597\ See Proposing Release, supra note 11, section V.D.5.
\1598\ See MDI Adopting Release, supra note 10, section
V.C.1(c)(i), for the full discussion of the effects of including
odd-lot information inside the NBBO in its definition of core data.
Also, the MDI Rules do not require that the competing consolidators
to disseminate odd-lot information, but the Commission anticipated
in the MDI Adopting Release that at least one would do so. The
requirement that the exclusive SIPs disseminate odd-lot information
helps ensure that the economic effects of the acceleration of the
MDI Rules occur. See infra section VII.D.5.c for a discussion of the
costs to the exclusive SIPs.
\1599\ See MDI Adopting Release, supra note 10, section
V.C.1(c)(i).
---------------------------------------------------------------------------
Multiple commenters remarked on the growing importance of odd-lot
activity for the overall equities market. One commenter stated that
odd-lots ``provide a meaningful source of liquidity across all trading
sessions and stocks, representing 54.8% of all trades in the U.S.
financial markets, up from 43% at the beginning of 2020.'' \1600\
Similarly, another commenter stated that including odd-lot information
in consolidated market data will ``improve transparency and increase
the usefulness of the consolidated tape given the growing prevalence of
market activity in sub 100 share quantities . . . . Allowing for access
to this information, as proposed, would therefore likely result in
increased pre-trade transparency for both retail and institutional
investors and bolster execution quality.'' \1601\
---------------------------------------------------------------------------
\1600\ See Cboe Letter II at 10.
\1601\ See BlackRock Letter at 11.
---------------------------------------------------------------------------
In addition, the amendments will change the timing and magnitude of
compliance costs and other costs.\1602\ One commenter estimated that
quotation traffic will increase at least 35% as a result of adding odd-
lot data to the SIP feeds; this estimate was based on a previous
proposal by the CTA and UTP Operating Committees, which proposed a more
limited inclusion of odd-lot data to the SIP feeds.\1603\ The
associated costs will include: the cost for exclusive SIPs to upgrade
existing infrastructure and software to handle the dissemination of
additional message traffic,\1604\ the cost to SROs to implement system
changes required in order to make the data needed to generate odd-lot
information available to exclusive SIPs, and the cost of technological
investments market participants might have to make in order to receive
the SIP data.\1605\
---------------------------------------------------------------------------
\1602\ See MDI Adopting Release, supra note 10, at 18759 for the
full discussion of the costs associated with expanding core data to
include odd-lot information inside the NBBO. See also infra section
VII.D.5.c for further discussion of compliance costs.
\1603\ See FISD Letter at 3. The Commission agrees that the
addition of information on odd-lot quotes that are priced at or more
aggressively than the NBBO may substantially increase message
traffic. See MDI Adopting Release, supra note 10, at n.2019.
\1604\ Multiple commenters agreed with the amendments' effect on
message traffic. See, e.g., Citadel Letter I at 26 and FIA PTG
Letter II at 4.
\1605\ See supra note 1602.
---------------------------------------------------------------------------
While these aforementioned economic effects of including odd-lots
in NMS data will be realized sooner, the Commission does not expect
that the amendments will accelerate all the effects described in the
MDI Rules related to adding to NMS data odd-lot information inside the
NBBO. The amendments will not accelerate the decentralized
consolidation model and will therefore not accelerate the benefits from
allowing some market participants to reduce data expenses required for
trading by providing a reasonable alternative to some market
participants to proprietary data.\1606\ As such, the amendments will
also not accelerate the cost to users of proprietary data whose
information advantage will dissipate somewhat. In particular, the
Commission does not believe that adding the specified odd-lot
information to the exclusive SIPs will result in low-latency traders
substituting the exclusive SIPs for their current proprietary data
usage. This is because a key component of the MDI Rules for this
functionality is an expected reduction in latency of NMS data
anticipated from the competing consolidator model of NMS data
distribution.\1607\ The exclusive SIPs are not expected to be fast
enough to replace proprietary data because existing SIP latency will
not be reduced or affected by this Rule. Thus, the amendments will not
accelerate the benefits anticipated in the MDI Rules that pertain to
using low-latency odd-lot information. Instead, the Commission expects
these effects to be realized after the implementation of all MDI Rules.
---------------------------------------------------------------------------
\1606\ Id.
\1607\ See MDI Adopting Release, supra note 10, at 18752 n.1939.
---------------------------------------------------------------------------
Market participants who decide to receive and use odd-lot quotation
information from the exclusive SIPs under these amendments will also
incur costs if the acceleration results in additional systems changes
when competing consolidators begin offering odd-lot information.
Specifically, market participants that decide to receive odd-lot
quotation information from exclusive SIPs will need to make systems
changes upon implementation of the acceleration of the MDI Rules in
order to receive the odd-lot quotation information. Because the data
specifications of the competing consolidators are unknown and could
differ from the data specification of the exclusive SIPs, market
participants receiving odd-lot information from the exclusive SIPs
could also need to make systems changes again to receive the odd-lot
information from a competing consolidator upon full implementation of
the MDI Rules.\1608\ If there are significant fixed costs associated
with system changes that are incurred on each change, then multiple
system changes will be inefficient and could increase costs. Because
market participants who receive odd-lot quotation information from the
exclusive SIPs may need to make an extra systems change stemming from
this Rule--one change to receive the data from the exclusive SIPs, and
[[Page 81749]]
potentially another change to receive the data from a competing
consolidator--some market participants may decide not to implement
systems changes to make use of the accelerated implementation of the
odd-lot information and, instead, wait until the MDI Rules are fully
implemented. This would dampen some of the benefits of accelerating the
inclusion of odd-lot quotation information.
---------------------------------------------------------------------------
\1608\ One commenter pointed out that this duplication of effort
becomes more likely if SIPs do not choose to register as competing
consolidators. See BlackRock Letter at 12. The Commission agrees
with this assessment.
---------------------------------------------------------------------------
To the extent that some market participants store SIP data for
various purposes (such as transaction cost analysis), the acceleration
of the MDI Rules could hasten an increase in storage costs because the
amount of SIP data increases with the inclusion of odd-lot data. Many
factors affect these costs in total, such as the number of market
participants storing SIP data, the data structures they use to store
SIP data, whether these market participants will choose to store all or
just some of the SIP data provided by the amendments, and the period
over which the amendments will affect these storage costs. Because the
Commission does not have information on how many market participants
will store MDI odd-lot information and the methods they will use to do
so, the Commission is unable to estimate these costs.
One commenter stated that displaying odd-lot quotes ``could lead
investors to expect prices that are not available.'' \1609\ The
Commission acknowledges that there may be a learning curve associated
with the dissemination of odd-lots on the SIP. However, any confusion
from the accelerated dissemination of odd-lot quotes would have
occurred eventually under the original MDI timeline, so the incremental
effect of MDI acceleration on investor confusion is minimal. Further,
as discussed in the MDI Adoption Release, the Commission acknowledges
that many retail investors may not directly view the entire content of
expanded core data--retail brokers may decide not to offer their
customers direct access to all of the odd-lot information but may
rather customize products derived from odd-lot information.\1610\ The
provider of these customized products is expected to supply the
information in a way that does not confuse the provider's customers. To
the extent that retail brokers allow some customers to directly utilize
all of the odd-lot information, the customers who choose to do so will
likely be sophisticated--as evidenced by their seeking out the
information--and will likely not be confused.\1611\
---------------------------------------------------------------------------
\1609\ See Schwab Letter II at 36.
\1610\ See MDI Adopting Release, supra note 10, at 18753.
\1611\ See MDI Adopting Release, supra note 10, at 18754.
---------------------------------------------------------------------------
c. Dissemination of Odd-Lots in SIP Data
The Amendments require the exclusive SIPs to disseminate odd-lot
data.\1612\ As discussed in the Proposing Release,\1613\ this
requirement will help realize the benefits of accelerating the
implementation of including odd-lot information in NMS data while
imposing costs on exclusive SIPs and potentially on market
participants.\1614\ The MDI Rules do not require the competing
consolidators to disseminate odd-lot data. However, the Commission
estimated in the MDI Rules that at least one competing consolidator
will do so because there will be demand for the data.\1615\ These
amendments, though, do not accelerate the competing consolidator model.
Unlike competing consolidators, each exclusive SIP is the only
distributor of the entirety of its data and may lack the incentive to
disseminate the data. As a result, the Commission cannot rely on the
exclusive SIPs to disseminate the odd-lot information prescribed by the
MDI Rules absent a requirement to do so; the benefits of the
acceleration could therefore be at risk without the Amendment to Rule
603's requirement for the SIPs to disseminate.\1616\
---------------------------------------------------------------------------
\1612\ See 17 CFR 242.603(b)(3) for rule text relating to this
requirement under Rule 603.
\1613\ See Proposing Release, supra note 11, at 80332.
\1614\ See infra section VII.D.5.c for additional discussion of
the costs the exclusive SIPs are expected to incur.
\1615\ See supra note 1155.
\1616\ The Commission recognizes that the exclusive SIPs have
some incentive to offer odd-lots as indicated by the exclusive SIPs
seeking comment on doing so. See, e.g., Proposal of the CTA and UTP
Operating Committees Regarding Odd Lots on the SIPs (Mar. 2022),
available at https://www.ctaplan.com/publicdocs/ctaplan/CTA_Odd_Lots_Proposal_2022.pdf.
---------------------------------------------------------------------------
While the inclusion of the odd-lot data could impose costs on those
who receive and use exclusive SIP odd-lot data,\1617\ the requirement
that exclusive SIPs disseminate the data could also impose costs on
those who receive but do not have an interest in using odd-lot
information provided in SIP data. These costs would vary based on how
the exclusive SIPs decide to implement the dissemination of the MDI
odd-lot information. For example, if the exclusive SIPs offer a
separate data feed for odd-lot quotation information, then market
participants that do not have an interest in this information may not
incur any additional costs because they would not need to subscribe to
this data feed. If the exclusive SIPs instead incorporate MDI odd-lot
quotation information into an existing data feed, then market
participants may incur costs to update their systems to filter out the
unwanted odd-lot information; the Commission is unable to estimate
these costs because they would vary across market participants and
depend upon each market participant's existing infrastructure, which is
unknown to the Commission. Further, such SIP data users could incur the
cost of any SIP data fee increases intended to offset the costs to
exchanges and exclusive SIPs.\1618\ However, SIP data fees did not
increase when the exclusive SIPs started to include odd-lot trades.
---------------------------------------------------------------------------
\1617\ These costs include systems changes and data storage. See
section VII.D.4.b for a discussion of these costs.
\1618\ Any changes in fees for SIP data would need to be filed
by the Equity Data Plans and approved by the Commission. See supra
note 887 and accompanying text for further discussion.
---------------------------------------------------------------------------
d. Best Odd-Lot Order Definition
The amendments go beyond the MDI Rules by requiring that NMS data
also include information on the best priced odd-lot orders across all
markets. Including the best odd-lot order in a standardized form will
offer market participants a standard benchmark, like the NBBO, to use
to measure execution quality. As discussed in the Rule 605 Amendments
\1619\ a market center may be able to internalize an order and claim
price improvement relative to the NBBO even if better priced odd-lots
are available at another market center. A standardized best odd-lot
benchmark may give market participants that receive it valuable
information for evaluating broker-dealers and market centers. Including
this benchmark in NMS data allows the information to be readily
available to a broad set of market participants, including those
investors to whom broker-dealers choose to make this information
available.\1620\
---------------------------------------------------------------------------
\1619\ See Rule 605 Amendments, supra note 10, section
IX.D.b.2.c.vii for further discussion of the benefits of disclosing
execution quality benchmarked to the best available displayed price.
\1620\ Because the amount of information disseminated as a
result of the amended rule BOLO dissemination requirement would
likely result in significantly less message traffic compared to the
amount of information that would be disseminated as a result of the
amended rule MDI odd-lot information dissemination requirement, we
expect market participants will be able to receive the BOLO
information required here without having to make significant system
upgrades, unlike the MDI odd-lot information. Therefore, a broader
set of market participants may choose to receive the BOLO from the
exclusive SIPs (and later on, competing consolidators) than all of
the MDI odd-lot information. Additionally, because of the lower
message traffic, more broker-dealers may make information on the
BOLO available to their customers than MDI odd-lot information.
---------------------------------------------------------------------------
[[Page 81750]]
Currently, this information is only available to market
participants who have proprietary data feeds, and even then there could
be differences across market participants with these data in terms of
how exactly market participants calculate the best odd-lot order (or
how many proprietary feeds they include). The best odd-lot information
in the NMS data will provide a standardized benchmark that reflects the
best odd-lot price consolidated across all national securities
exchanges and national securities associations. This benchmark may
allow more market participants to better monitor the execution quality
of their broker-dealers and send more trading volume to broker-dealers
with better performance.\1621\ One commenter highlighted the value of
this new benchmark: ``. . .transparency requires that the units used to
represent the range of prices available in the market match the units
in which participants typically quote and trade.'' \1622\ For market
participants who receive the BOLO and typically execute small trades,
the best odd-lot order will provide them more relevant information on
available orders. Thus, including the best odd-lot information could
enhance competition among broker-dealers leading to better trade
execution and perhaps a lower cost to customers for execution services.
---------------------------------------------------------------------------
\1621\ While the Commission does not expect most retail traders
would engage in this sort of benchmarking due to a lack of technical
capacity to do so among most retail traders, institutional traders
likely have such capacity and so would engage in this type of
monitoring. Institutional traders have strong incentives to monitor
all aspects of transaction costs as these costs can significantly
affect portfolio performance. See Amber Anand, et al., Performance
of Institutional Trading Desks: An Analysis of Persistence in
Trading Costs, 25 Rev. Fin. Stud. 557 (2012).
\1622\ See IEX Letter I at 31. Another commenter agreed more
generally that the best odd-lot order will enhance the usefulness of
odd-lot information and enhance liquidity--see Cboe Letter II at 10.
---------------------------------------------------------------------------
One commenter expressed concern that the best odd-lot order would
result in ``displaying locked/crossed markets.'' \1623\ It is possible
for the best odd-lot bid to be at a price equal to or higher than the
best odd-lot ask; in these cases, the best odd-lot order would show a
locked or crossed market.\1624\ Academic research shows that the NBBO
does get locked and crossed from time to time.\1625\ These tend to be
fleeting events.\1626\ Because the BOLO will provide prices inside the
NBBO, the BOLO will likely be crossed or locked more frequently than
the NBBO. However, it is unclear what practical effect a locked or
crossed BOLO would have on financial markets or those that use the
BOLO. Market participants are already well versed in using the NBBO,
which can be locked and crossed from time to time, so it is likely that
they would use similar techniques for dealing with locked and crossed
markets when, for example, benchmarking relative to the BOLO.\1627\
Further, market participants who subscribe to proprietary data feeds
already have access to information on when the best odd-lot orders may
lock or cross each other; the rule amendment merely extends this
information to market participants who do not subscribe to proprietary
data feeds. As more market participants see the information contained
in the BOLO, there may be fewer instances of locked and crossed odd-lot
quotes--e.g., more market participants will have the information needed
to arbitrage crossed odd-lot markets. Finally, a locked or crossed BOLO
will be less disruptive than a locked or crossed NBBO because Rule 610
of Regulation NMS requires SROs to adopt rules requiring their members
reasonably to avoid displaying quotations that lock or cross protected
quotations.\1628\ However, the BOLO does not establish a protected
quote, and so a locked or crossed BOLO would not trigger the same
reaction by SROs and their members as a locked or crossed NBBO.
---------------------------------------------------------------------------
\1623\ See SIFMA Letter II at 34 requesting an analysis of the
effect of the BOLO on the display of locked and crossed markets.
\1624\ See supra note 1051 for the definition of locked and
crossed markets. A locked or crossed market occurs when there is a
passive buy order on one venue at a price greater or equal to the
price of an existing passive sell order at another venue; the fact
that these orders have not executed against each other indicates
that there is a friction between the trading venues.
\1625\ See Craig W. Holden & Stacey Jacobsen, Liquidity
Measurement Problems in Fast, Competitive Markets: Expensive and
Cheap Solutions, 69 J. Fin. 1747 (2014). This paper estimates that
1.7% of trades occur when the NBBO is locked, and 0.5% of trades
occur when the NBBO is crossed--see table 1, Panel A, column 4
therein. The authors conjecture that some of these instances arise
due to a data issue where quotes have been canceled, but the
cancellation was not recorded by the time of the trade.
\1626\ Id.
\1627\ See Id. for an example of one methodology used when
employing market data in the presence of locked or crossed markets.
\1628\ 17 CFR 242.610(d)(1).
---------------------------------------------------------------------------
Other commenters discussed the effect BOLO may have on investor
confusion. Two commenters stated that: ``Calculating and publishing an
odd-lot NBBO risks creating significant investor confusion due to the
appearance that a new benchmark is being established even though odd-
lots are treated differently than round-lots under Commission
regulations. Rather than taking steps to prevent unnecessary investor
confusion, the Commission encourages it by suggesting that the odd-lot
NBBO is a `standard benchmark' that could be used by investors `to
measure the amount of price improvement they receive for the execution
of their orders.' '' \1629\ The commenters continued: ``The odd-lot
NBBO is not a standard benchmark, since the size associated with these
quotes will vary greatly as opposed to the actual NBBO, which always
represents a round-lot.'' \1630\ Similarly, some commenters stated that
the BOLO will not provide a useful benchmark and may instead distort
price improvement statistics.\1631\
---------------------------------------------------------------------------
\1629\ See Citadel Letter I at 26-27, and SIFMA Letter II at 43-
44.
\1630\ Id.
\1631\ See JPMorgan Letter at 7, FIA PTG Letter II at 5, and ASA
Letter at 6.
---------------------------------------------------------------------------
The Commission acknowledges that the BOLO is not, at present, a
widely used and standard benchmark. This is likely because the
requisite data is not broadly distributed and is only available to
market participants who have proprietary data feeds. In contrast, the
NBBO is broadly distributed in NMS data and is also more widely used as
a benchmark. The Commission believes that including the BOLO in NMS
data will similarly allow market participants to more easily use the
BOLO as a benchmark if they choose to.\1632\
---------------------------------------------------------------------------
\1632\ See infra section VII.D.6.a.ii discussing how the BOLO
will make it easier for market centers and broker-dealers to compute
statistics on price improvement relative to the best available
displayed price, now required by amended Rule 605.
---------------------------------------------------------------------------
The Commission also recognizes that an odd-lot price that is better
than the NBBO may not reflect sufficient quantity to execute certain
orders, particularly larger-sized orders, and, as a result, price
improvement relative to the BOLO will be more relevant in some cases
than for others. However, market participants are already well versed
in interpreting nuanced benchmarks--for example, holding the round lot
size constant, the NBBO may reflect a different amount of dollar
liquidity based on the price of the stock. Furthermore, the size
available at the NBBO of any particular stock might be a great deal
more than a single round lot. This means that market participants
already deal with the distinction between the price of a benchmark and
the amount of shares available at that benchmark price. Indeed, while
one commenter points out the challenge of comparing a 500-share order
to a 10-share odd-lot,\1633\ a similar challenge already exists in
comparing a 1000-
[[Page 81751]]
share order to a 100-share round lot. This challenge is understood and
handled already; market participants already know that quantity must be
taken into account when making comparisons. The Commission expects that
market participants who will benchmark their trades with the BOLO will
generally have comparable levels of sophistication as investors who
currently use the NBBO benchmark; given that users of the NBBO
benchmark are already adept at accounting for order size, these users
should not be confused by the BOLO. It is important that market
participants have access to a variety of benchmarks to meet their
various purposes, and the BOLO will provide a useful data point for
market participants to consider in addition to the NBBO.
---------------------------------------------------------------------------
\1633\ See Citadel Letter I at 27.
---------------------------------------------------------------------------
5. Compliance Costs
Various market participants will incur one-time implementation
costs as well as ongoing compliance costs to comply with the Rule.
These costs and their computations are discussed in greater detail
below, but are summarized in table 15. Some of the costs are associated
with the acceleration of aspects of the MDI Rules and will only
represent new costs (which are not already anticipated under the MDI
rules) if the exclusive SIPs do not become competing consolidators once
the MDI rules are fully implemented.
Table 15--Compliance Cost Estimates
----------------------------------------------------------------------------------------------------------------
Total
Rule Affected One-time Ongoing Number of Total one- ongoing
entities costs costs entities time costs costs
----------------------------------------------------------------------------------------------------------------
612.......................... All trading $156,000 ......... 277 $43,212,000 .........
venues \a\.
612.......................... Listing 33,000 9,000 5 165,000 $45,000
exchanges \b\.
612.......................... SIPs \c\........ 13,000 9,000 2 26,000 18,000
612.......................... Broker-dealers 33,000 ......... 1,161 38,313,000 .........
with order
entry systems
\d\.
612.......................... Broker-dealers 11,000 ......... 270 2,970,000 .........
with smart
order routers
\e\.
610.......................... Exchanges \f\... 57,000 ......... 15 855,000 .........
600, 603..................... Exchanges \g\... 3,500 6,500 16 56,000 104,000
600, 603, 612................ SIPs \h\........ 613,000 174,000 2 1,226,000 348,000
----------------------------------------------------------------
Total.................... ................ ......... ......... .............. 86,823,000 515,000
----------------------------------------------------------------------------------------------------------------
Sources: Across estimates below, salaries are derived from SIFMA's Management & Professional Earnings in the
Securities Industry 2013, modified to account for an 1,800-hour work-year and inflation, and multiplied by
5.35 to account for bonuses, firm size, employee benefits and overhead. The burden hours estimates are based
on Commission's experiences with burden estimates.
\a\ See Proposing Release, supra note 11, at 80333. The Proposing Release's estimate of $140,000 is adjusted to
$156,000 to account for an 11.4% increase in the Producer Price Index for Data Processing, Hosting and Related
Services from December 2014, when the $140,000 estimate was first made. See U.S. Bureau of Labor Statistics,
Producer Price Index by Industry: Data Processing, Hosting and Related Services: Hosting, Active Server Pages
(ASP), and Other Information Technology (IT) Infrastructure Provisioning Services [PCU5182105182105],
retrieved from FRED, Federal Reserve Bank of St. Louis; available at https://fred.stlouisfed.org/series/PCU5182105182105 PCU5182105182105 (Mar. 11, 2024).
\b\ The $33,000 estimate per listing exchange is based on the following calculations: $19,950 (hourly rate for
Sr. Programmer at $399 for 50 hours) + $6,860 (hourly rate for Sr. Systems Analyst at $343 for 20 hours) +
$3,730 (hourly rate for Compliance Manager at $373 for 10 hours) + $2,940 (hourly rate for Director of
Compliance at $588 for 5 hour) $33,000, for a total annual monetized burden of $165,000 (i.e., $165,000 =
$33,000 x 5 listing exchanges). The $9,000 estimate per listing exchange is based on the following
calculations: ($2,640 (hourly rate for Compliance Attorney at $440 for 6 hours) + $746 (hourly rate for
Compliance Manager at $373 for 2 hours)) x 4 tick size revisions per year) $9,000, for a total annual
monetized burden of $45,000 (i.e., $45,000 = $9,000 x 5 listing exchanges).
\c\ The $13,000 estimate per listing exchange is based on the following calculations: $4,788 (hourly rate for
Sr. Programmer at $399 for 12 hours) + $1,715 (hourly rate for Sr. Systems Analyst at $343 for 5 hours) +
$3,730 (hourly rate for Compliance Manager at $373 for 10 hours) + $2,940 (hourly rate for Director of
Compliance at $588 for 5 hour) $13,000. The $9,000 estimate per listing exchange is based on the following
calculations: ($2,640 (hourly rate for Compliance Attorney at $440 for 6 hours) + $746 (hourly rate for
Compliance Manager at $373 for 2 hours)) x 4 tick size revisions per year) $9,000.
\d\ The $11,000 estimate per system change is based on the following calculations: ($2,005 (hourly rate for
Attorney at $401 for 5 hours) + $2,980 (hourly rate for Compliance Manager at $298 for 10 hours) + $4,640
(hourly rate for Programmer Analyst at $232 for 20 hours) + $1,325 (hourly rate for Senior Business Analyst at
$265 for 5 hours) [ap] $11,000. The Commission expects that broker-dealers are likely to have to undertake 3
system changes, for a total one-time expense of $33,000. See also Transaction Fee Pilot Adopting Release,
infra note 1644, at 5271 n.770.
\e\ The $11,000 estimate per broker-dealers with smart order routers is based on the following Manager at $298
for 10 hours) + $4,640 (hourly rate for Programmer Analyst at $232 for 20 hours) + $1,325 (hourly rate for
Senior Business Analyst at $265 for 5 hours) [ap] $11,000. See also Transaction Fee Pilot Adopting Release,
infra note 1644, at 5274 n.796 where the cost to broker-dealers to update systems for the TSP was estimated to
be $9,000. Here, we are allowing for an additional 10 hours of Programmer Analyst time.
\f\ See Proposing Release, supra note 11, at 80333.
\g\ The additional $3,500 in one-time costs and $6,500 in ongoing costs represent a 5% addition over the costs
reported in the MDI release. See supra note 10, section V.C.2(d)(ii) to account for the new requirement to
send the necessary data to generate odd-lot information to the exclusive SIPs.
\h\ The $613,000 estimate in one-time costs is based on the following calculations: $33,000 (costs under the
amendments to Rule 612 to update data specifications and internally and externally test the updates) +
$167,670 ($83,790 (hourly rate for Sr. Programmer at $399 for 210 hours) + $61,740 (hourly rate for Sr.
Systems Analyst at $343 for 180 hours) + $7,460 (hourly rate for Compliance Manager at $373 for 20 hours) +
$5,880 (hourly rate for Director of Compliance at $588 for 10 hours) + $8,800 (hourly rate for Compliance
Attorney at $440 for 20 hours) + $412,500 (costs for external services). The $174,000 estimate in ongoing
costs is based on the following calculations: $50,301 ($25,137 (hourly rate for Sr. Programmer at $399 for 63
hours) + $18,522 (hourly rate for Sr. Systems Analyst at $343 for 54 hours) + $2,238 (hourly rate for
Compliance Manager at $373 for 6 hours) + $1,764 (hourly rate for Director of Compliance at $588 for 3 hours)
+ $2,640 (hourly rate for Compliance Attorney at $440 for 6 hours)) + $123,725 (costs for external services).
See infra notes 1745, 1747, 1749, and 1750 and accompanying text for relevant details on these cost estimates.
a. Estimates for Rule 612
Each trading venue will have to update systems to comply with the
change in tick size for some NMS stocks under the Rule 612 amendments.
Due to similarities with the changes that were required by the TSP, the
Commission estimated, in the Proposing Release, that the amendments to
Rule 612 would impose the same costs to trading venues as those
estimated for the TSP.\1634\
[[Page 81752]]
These costs were estimated at $140,000 in 2014 at the time of the
TSP.\1635\
---------------------------------------------------------------------------
\1634\ See Proposing Release, supra note 11, at 80333.
\1635\ An exchange commenting on the Tick Size Pilot estimated
$140,000 as its expected expense to comply with the Tick Size
Pilot's requirement to change the tick size for some stocks. See
James G. Ongena, Chicago Stock Exchange (CHX), Comment Letter Re:
File No. 4-657; Notice of Filing of the Proposed National Market
System Plan to Implement a Tick Size Pilot Program On a One-Year
Pilot Basis (Dec. 2014), available at https://www.sec.gov/comments/4-657/4657-67.pdf.
---------------------------------------------------------------------------
Some commenters stated that the $140,000 estimated cost to trading
venues in the proposal was too low because exchanges would have to
acquire additional hardware and update various systems.\1636\ These
commenters did not provide alternative estimates for the implementation
costs. As discussed in the Proposing Release, this $140,000 estimate is
derived from exchange feedback on the costs associated with the
TSP.\1637\ This estimate acknowledges that the market participants may
have hardware and system costs associated with the amendments. Given
that those hardware and system changes are similar in nature to those
associated with the TSP, and the commenters did not provide analysis to
the contrary, the Commission continues to believe that this estimate is
reasonable. One commenter suggested that the costs of processing and
disseminating trading information may increase linearly with any
increases in message traffic.\1638\ However, estimating the costs in
this manner is not possible as the Commission does not know the current
costs incurred by exchanges in processing and disseminating trading
information, is unaware of data sources that could provide reliable
estimates, and commenters did not provide such information. As
discussed in section VII.D.1.c, the Commission acknowledges that
message traffic may increase due to the amendments to Rule 612, and so
the costs of processing and disseminating message traffic may similarly
increase. There is, however, uncertainty as to whether exchanges will
need to incur additional hardware investments, and the degree of such
investments, if they are needed, will likely differ from exchange to
exchange, as it would depend on the capacity of their existing
infrastructure to handle increased data.
---------------------------------------------------------------------------
\1636\ See FIF Letter at 9-10 and FISD Letter at 3.
\1637\ See Proposing Release, supra note 11, at 80333 n.618 and
surrounding text.
\1638\ See FIF Letter at 9 (``Some FIF members would estimate
that the increased server, bandwidth and other costs would be
roughly proportional to the increase in message traffic'').
---------------------------------------------------------------------------
To account for likely increases in the costs of computer hardware
since 2014, the Commission is revising the estimated costs from
$140,000 to $156,000 per trading venue.\1639\ As shown in table 15, the
Commission estimates that the compliance costs associated with the
amendments of Rule 612 across all trading venues are $43 million. This
estimate is computed by multiplying an estimated $156,000 in one-time
costs incurred by each trading venue to update systems to comply with
the amendments to Rule 612, by the estimated number of trading venues,
which is 277 trading venues.\1640\
---------------------------------------------------------------------------
\1639\ The Commission has updated the expected costs to $156,000
from $140,000 to reflect the roughly 11.4% increase in the Producer
Price Index for Data Processing, Hosting and Related Services from
December 2014, when the $140,000 estimate was first made. See U.S.
Bureau of Labor Statistics, Producer Price Index by Industry: Data
Processing, Hosting and Related Services: Hosting, Active Server
Pages (ASP), and Other Information Technology (IT) Infrastructure
Provisioning Services [PCU5182105182105] (Mar 11, 2024, retrieved
from FRED, Federal Reserve Bank of St. Louis, available at https://fred.stlouisfed.org/series/PCU5182105182105.
\1640\ The technical aspect of a broker-dealer that internalizes
customer orders updating its system to reflect the tiered tick
regime is likely similar to that of an exchange or an ATS. Thus, the
Commission is applying the same cost estimate for wholesalers and
other broker-dealers that execute customer orders to update their
systems as that applied to exchanges and ATSs. In Q1 2023 there were
16 registered exchanges, 33 ATSs, and 228 other FINRA members
(including wholesalers) that executed orders off-exchange. In the
first quarter of 2023, there were 277 total entities affected. See
Rule 605 Amendments, supra note 10, at 26542.
---------------------------------------------------------------------------
Under the amendments to Rule 612, listing exchanges will have to
calculate NMS stocks' time weighted average quoted spreads and transmit
their associated tick size to the exclusive SIPs. The Commission does
not believe the reduction in the number of tick sizes relative to the
proposal will significantly affect compliance costs since the TWAQS
still needs to be calculated for each stock and assigning a stock to a
tick size is not computationally or conceptually difficult once the
TWAQS has been computed. Thus, the Commission is keeping the estimate
for listing exchanges the same at $33,000 per listing exchange.\1641\
---------------------------------------------------------------------------
\1641\ The $33,000 estimate per listing exchange is based on the
following calculations: $19,950 (hourly rate for Sr. Programmer at
$399 for 50 hours) + $6,860 (hourly rate for Sr. Systems Analyst at
$343 for 20 hours) + $3,730 (hourly rate for Compliance Manager at
$373 for 10 hours) + $2,940 (hourly rate for Director of Compliance
at $588 for 5 hour) [ap] $33,000. Salaries for estimates in this
section are derived from SIFMA's Management & Professional Earnings
in the Securities Industry 2013, modified to account for an 1,800-
hour work-year and inflation, and multiplied by 5.35 to account for
bonuses, firm size, employee benefits and overhead.
---------------------------------------------------------------------------
Commenters stated that modifications to the data specifications
with regards to transferring the tick size information from the listing
exchanges to the SIPs would require both internal and external testing
by the primary listing exchange and the SIPs.\1642\ Commenters stated
that modifications to data specifications require software changes and
require testing.\1643\ The Commission anticipates that the SIPs will
not have to acquire additional hardware or develop new systems in order
to incorporate the minimum pricing increment indicator; they will
rather need to update existing specifications. The Commission expects
that the amendments to Rule 612 will require a one-time cost for
updating existing systems and will not increase the cost of becoming a
competing consolidator once the MDI Rules are implemented, because the
amendments are not expected to increase the cost of establishing new
systems. Accordingly, the Commission estimates a one-time cost of
$13,000 \1644\ and ongoing costs of $9,000 per year \1645\ for the two
SIPs.
---------------------------------------------------------------------------
\1642\ See generally CTA-UTP Operating Committee Letter and FIF
Letter.
\1643\ Id.
\1644\ The $13,000 estimate per listing exchange is based on the
following calculations: $4,788 (hourly rate for Sr. Programmer at
$399 for 12 hours) + $1,715 (hourly rate for Sr. Systems Analyst at
$343 for 5 hours) + $3,730 (hourly rate for Compliance Manager at
$373 for 10 hours) + $2,940 (hourly rate for Director of Compliance
at $588 for 5 hour) $13,000. Salaries for estimates in this section
are derived from SIFMA's Management & Professional Earnings in the
Securities Industry 2013, modified to account for an 1,800-hour
work-year and inflation, and multiplied by 5.35 to account for
bonuses, firm size, employee benefits and overhead.
\1645\ The $9,000 estimate per listing exchange is based on the
following calculations: ($2,640 (hourly rate for Compliance Attorney
at $440 for 6 hours) + $746 (hourly rate for Compliance Manager at
$373 for 2 hours)) x 4 tick size revisions per year) $9,000.
---------------------------------------------------------------------------
These estimates are based on the Commission's understanding that
the listing exchanges currently have access to the data needed to
calculate the time weighted average quoted spreads because such data,
specifically the NBBO, are needed for the exchanges to compile Rule 605
reports.\1646\ Thus, the Commission does not expect that the exchanges
will incur additional costs associated with gathering data.
Additionally, the listing exchanges have experience computing a share-
weighted measure of average quoted spreads for their Rule 605
reports.\1647\ The listing
[[Page 81753]]
exchanges also already have connections to the exclusive SIPs, and once
competing consolidators replace the exclusive SIPs it is the competing
consolidators that will have the responsibility to connect to the
exchanges in order to receive data. Thus, under the MDI Rules the
exchanges will not incur additional costs to connect to the competing
consolidators.\1648\ Additionally, the SIPs have experience
distributing regulatory data and so the costs represent those of adding
the tick size to existing data. Consequently, the Commission expects
that having the listing exchange compute time weighted average quoted
spreads and transmit the associated tick to the exclusive SIPs
currently, or to the competing consolidators once the exclusive SIPs
are discontinued, will require listing exchanges to modify existing
systems, rather than build or acquire new systems or hardware.
---------------------------------------------------------------------------
\1646\ See, e.g., 17 CFR 242.605(a)(ii)(E), requiring the
reporting of the number of shares executed with price improvement,
and 17 CFR 600(b)(36) defining ``executed with price improvement''
to mean, for buy orders, execution at a price lower than the
national best offer at the time of order receipt and, for sell
orders, execution at a price higher than the national best bid at
the time of order receipt.
\1647\ See, e.g., 17 CFR 242.605(a)(ii)(A), requiring the
reporting of the average quoted spread for executions of covered
orders, and 17 CFR 600(b)(12), defining the average quoted spread as
the share-weighted average of the difference between the national
best offer and the national best bid at the time of order receipt
or, for order executions of midpoint-or-better limit orders, the
difference between the national best offer and the national best bid
at the time such orders first become executable. Additionally, some
listing exchanges have issued white papers that include statistics
based on time weighted average quoted spreads. See, e.g., Nasdaq
Intelligent Tick, supra note 150, Chart 3 and Cboe Proposal, supra
note 150, Exhibit 1.
\1648\ See MDI Adopting Release, supra note 10, at 18612 n.1133
and surrounding text. The costs for the competing consolidators to
connect to the exchanges is accounted for in the MDI Rules and thus
would not represent costs associated with this proposal.
---------------------------------------------------------------------------
The Commission expects that broker-dealers with order entry systems
will need to modify their existing systems to comply with the tick size
changes and will not need to acquire new hardware or develop new
systems for this specific aspect of the adopted rule. In the Proposing
Release, the Commission had estimated the cost of a broker-dealer
system change at $11,000.\1649\ One commenter stated that for most
firms a significant technological build will not be needed,\1650\ other
commenters stated that broker-dealers would have to undertake
significant systems work, or acquire additional hardware, as a result
of the amendments to Rule 612, specifically if there is a significant
increase in message traffic.\1651\ The message traffic implications and
costs are discussed in section VII.D.1.c. This section deals
specifically with modifications to broker-dealer order entry systems.
---------------------------------------------------------------------------
\1649\ See Proposing Release, supra note 11, at 80332-34.
\1650\ See Apex Letter at 15.
\1651\ See FIF Letter at 9, TradeStation Letter at 6, Citigroup
Letter at 2, and FISD Letter at 3.
---------------------------------------------------------------------------
One commenter stated that the $11,000 estimate was too low because
the proposed amendments to Rule 612 would necessitate additional
expenses in order to update ``their order management, execution
management, customer trading, middle-office trade processing,
reporting, settlement, surveillance and compliance systems.'' \1652\ In
the Proposing Release, when the Commission estimated the cost of a
broker-dealer system change at $11,000, it assumed that a single system
change would be needed in response to the Rule 612 amendments.\1653\ In
light of the various functions described by the commenter which would
need to be updated, the Commission is revising the estimated one-time
cost to broker-dealers to $33,000 per broker-dealer with an order-entry
system as reported in table 15.\1654\ The revised estimate stems from
the expectation that broker-dealers are more likely to have to
undertake three system changes, rather than one,\1655\ although the
Commission recognizes that these costs could be greater if additional
hardware investment is needed.
---------------------------------------------------------------------------
\1652\ See FIF Letter at 9.
\1653\ See Proposing Release, supra note 11, at 80332.
\1654\ See table 15 note d.
\1655\ The Commission believes that the order management,
execution management, and customer trading functions highlighted by
the commenter are sufficiently similar to be covered under a single
system change. The middle-office trade processing, reporting, and
settlement functions constitute another system change. Surveillance
and compliance systems constitute a third system change.
---------------------------------------------------------------------------
The Commission estimates that there are 1,161 broker-dealers with
order entry systems.\1656\ Thus, the Commission estimates that the
amendments will lead to a one-time aggregate cost of around $38.5
million (i.e., $38.5 million [ap] $33,000 x 1,161) across broker-
dealers with order entry systems to update their systems to account for
the new tick sizes.
---------------------------------------------------------------------------
\1656\ Using CAT data from December 2023, the Commission
calculated the total number of unique Central Registration
Depository Numeric Identifiers ``CRDs'' that originated an order to
estimate the number of entities with an order entry system.
---------------------------------------------------------------------------
The Commission expects that broker-dealers with smart order routers
will also need to modify their existing systems to comply with the tick
size changes and will not need to acquire new hardware or develop new
systems. The Commission estimates a one-time cost of $11,000 to broker-
dealers operating smart order routers.\1657\ These broker-dealers
already have systems that can adjust for tick sizes that change around
the $1.00 threshold. Thus, the Commission expects that they will modify
existing systems rather than build new systems. Any broker-dealer that
will need to build new systems will likely incur costs greater than
$11,000 to do so. One commenter stated that significant system changes
and hardware costs would also be required by broker-dealers operating
smart order routers due to an increase in message traffic.\1658\ The
Commission acknowledges that the one-time costs to broker-dealers
operating smart order routers may be greater than estimated if
additional hardware investment will be required.\1659\
---------------------------------------------------------------------------
\1657\ The $11,000 estimate per broker-dealers with smart order
routers is based on the following Manager at $298 for 10 hours) +
$4,640 (hourly rate for Programmer Analyst at $232 for 20 hours) +
$1,325 (hourly rate for Senior Business Analyst at $265 for 5 hours)
[ap] $11,000. See also Securities Exchange Act Release No. 84875
(Dec 19, 2018), 84 FR 5202 (Feb 20, 2019) (``Transaction Fee Pilot
Adopting Release'') at 5274 n.796 where the cost to broker-dealers
to update systems for the TSP was estimated to be $9,000. Here, we
are allowing for an additional 10 hours of Programmer Analyst time.
\1658\ See FIF Letter at 9.
\1659\ See supra section VII.D.1.c for additional discussion
about costs to market participants stemming from increases in
message traffic.
---------------------------------------------------------------------------
The Commission estimates an upper bound of 270 broker-dealers
operating smart order routers.\1660\ This number provides an upper
bound as it assumes that all entities with direct connections to
exchanges or ATSs use a smart order router, which is likely an over-
estimate. Aside from potential additional costs due to increased
message traffic, the Commission thus estimates a one-time cost of $3.0
million (i.e., $3.0 million [ap] $11,000 x 270) for market participants
to update smart order routers.\1661\ If fewer than 270 broker-dealers
operate their own smart order routers, then the $3.0 million estimate
is likely higher than the aggregate cost for these broker-dealers to
adjust their order routing systems to comply with these amendments.
---------------------------------------------------------------------------
\1660\ This number is estimated by counting the number of unique
CRDs that submitted an order directly to an exchange or ATS in the
month of December 2023.
\1661\ The Commission also expects there may be other costs
associated with updating systems to account for an increase in
message traffic resulting from the new tick sizes. See supra section
VII.D.1.c for additional discussion.
---------------------------------------------------------------------------
Further, the Commission believes that broker-dealers operating
smart order routers already subscribe to SIP data and will subscribe to
consolidated market data products once the competing consolidators
become operative. Thus, they will not incur a separate data expense to
receive the regulatory messages necessary to comply with Rule 612
amendments. The Commission also assumes that system updates will impose
a similar cost on larger and smaller entities given that, once code is
written, scaling it up is relatively inexpensive.
[[Page 81754]]
Lastly, the Commission recognizes that Rule 612 amendments could
increase the overall implementation costs of the MDI Rules. In
particular, stocks that will become less tick-constrained as a result
of the smaller tick size following these amendments could have more
odd-lot quotes inside the NBBO than anticipated when the Commission
adopted the MDI Rules.\1662\ As a result, the costs to SROs and
competing consolidators of collecting, transmitting, consolidating, and
disseminating odd-lot information will be greater than those described
in the MDI Rules. The Commission is unable to provide an estimate of
this cost because it would require predicting a complex interaction
between behavior changes from multiple types of market participants and
the resulting effect on the number of ticks inside the NBBO and the
volume of odd-lots submitted inside the NBBO. However, the cost
increase may not be significant, because the Commission generally
estimates that the infrastructure cost increases associated with an
increase in message traffic from the amendments to be approximately
1%.\1663\
---------------------------------------------------------------------------
\1662\ This is a result of a smaller tick size allowing
liquidity to spread over more levels, reducing the depth at each
level and could increase the number of odd-lot quotes at each level.
See supra section VII.D.1.b for additional discussion.
\1663\ See supra note 1334 and accompanying text.
---------------------------------------------------------------------------
Multiple commenters stated that many broker-dealers, particularly
those with retail customers, would have to incur additional costs for
investor education and customer assistance in order to handle any
investor confusion arising from the amendments to Rule 612.\1664\ One
commenter specifically mentioned that the amendments to Rule 612 would
complicate ``good-til-cancelled'' orders (``GTC orders'') as the orders
could be placed under one tick size and could still be active after a
tick size change.\1665\ This scenario is unlikely given that there will
be a period of one-month between end of the evaluation period and the
tick size implementation during which market participants will be able
to know which stocks will be changing their tick size. For an issue to
arise, the GTC order would have to be in place over the course of that
month and the trader would have to be unaware of the upcoming tick size
change. Retail facing broker-dealers will likely implement some method
of notifying their customers that the tick size for some stocks will
change following the end of the evaluation period. To the extent to
which customer confusion causes these costs to materialize, the
Commission would expect that these costs would likely be greater in the
time immediately following the implementation of the amendments, and
they would decrease over time as investors become accustomed to the new
tick size regime.
---------------------------------------------------------------------------
\1664\ See, e.g., FISD Letter at 2, TastyTrade Letter at 19, and
TradeStation Letter at 6.
\1665\ See TradeStation Letter at 6.
---------------------------------------------------------------------------
b. Estimates for Rule 610
As in the Proposing Release, the Commission estimates a $57,000
one-time cost to exchanges to comply with changes to Rule 610.\1666\
This estimate assumes that exchanges will combine in the same Rule 19b-
4 filing their proposals to amend their fees and rebates and make fees
and rebates determinable at the time of execution, and that this
process will not increase the cost of those filings. The Commission
recognizes that if these filings are not efficiently combined, then the
costs to exchanges could be higher than $57,000. The Commission
estimates also assume that LTSE will not file a proposed rule change
with the Commission because it does not currently charge access fees or
offer rebates, but that the other 15 exchanges will file proposed rule
changes. If so, these amendments will lead to an estimated one-time
total cost of $855,000 across exchanges to comply with Rule 610.\1667\
---------------------------------------------------------------------------
\1666\ See Proposing Release, supra note 11, at 80333.
\1667\ The Commission does not expect other market participants
to incur significant incremental costs associated with the change in
the access fees and rebates. As shown in table 4, market
participants deal with over 100 fee changes per year across all
exchanges and thus it reasonable to expect that one fee change by
the exchanges to bring their fees into compliance with these
amendments would represent an economically trivial incremental cost
to these market participants.
---------------------------------------------------------------------------
c. Estimates for Rules 600 and 603
The exclusive SIPs and exchanges will incur compliance costs
associated with accelerating the inclusion of odd-lot data inside the
NBBO in SIP data, adding the BOLO to SIP data, and accelerating the
implementation of the round lot definitions as amended in this release.
The round lot definitions (but for amendments to them in this release)
and the inclusion of odd-lot data inside the NBBO were both parts of
the MDI Rules. Thus, the amendments will accelerate the compliance
costs associated with these aspects of the MDI Rules. One difference is
that the MDI Rules anticipated that these changes to NMS data would
occur after the competing consolidator model was up and running. Thus,
the MDI Rules did not anticipate that the exclusive SIPs would incur
such costs unless they chose to become competing consolidators. The
addition of the best odd-lot order to the SIP data was not part of the
MDI Rules and will thus be a new cost under the amendments.
Accordingly, the discussion below distinguishes costs to the exclusive
SIPs in terms of those included in the MDI Rules and new costs from
these amendments.
The Commission estimates a one-time cost of $3,500 and ongoing
costs of $6,500 per year for at least two years for exchanges to comply
with the amendments to Rules 603 and 600.\1668\ This estimate accounts
for the acceleration of the necessary data to generate the odd-lot
information, including the best odd-lot order, and transmit this
information to the exclusive SIPs. The costs reported here account for
an increase in the costs associated with the MDI Rules that will
require the exchanges to transmit to competing consolidators all of the
data necessary for generating consolidated market data.
---------------------------------------------------------------------------
\1668\ In the MDI Adopting Release, supra note 10, at 18764, the
Commission estimated costs to the exchanges of collecting and
transmitting the necessary information to the competing
consolidators to be approximately $70,000 in one-time costs and
approximately $130,000 in ongoing costs. The additional $3,500 in
one-time costs and $6,500 in ongoing costs represent a 5% addition
over the costs in the MDI release to account for the new requirement
to send the necessary data to generate odd-lot information to the
exclusive SIPs (i.e., $3,500 [ap] $70,000 x 0.05 and $6,500 [ap]
$130,000 x 0.05). See infra note 1841 and accompanying text.
---------------------------------------------------------------------------
Consequently, for the exchanges, the costs associated with
providing the exclusive SIPs with odd-lot information will represent an
acceleration of costs anticipated in the MDI Rules rather than new
costs, with a few differences. First, the odd-lot information will be
transmitted to the exclusive SIPs as opposed to the competing
consolidators. Second, the ongoing costs of these amendments will be
incurred only until the exclusive SIPs are retired, which the
Commission estimates will be at least two years after the Commission's
approval of the plan amendment(s) required by Rule 614(e).
Compliance with the amendments to Rules 603 and 600 will require
the exclusive SIPs to develop, operate, and maintain systems to collect
and disseminate the odd-lot information inside the NBBO as well as the
best odd-lot order. The Commission expects that these costs will
primarily consist of costs that an exclusive SIP would incur if it were
to convert to a competing consolidator. Thus, for exclusive SIPs that
would have become competing consolidators in the absence of these
amendments, initial compliance costs
[[Page 81755]]
represent an acceleration of costs under the MDI Rules, rather than new
additional costs. Further, the ongoing costs for exclusive SIPs to
comply with Rules 600 and 603 will be incurred only until the exclusive
SIPs are retired, after which time these costs will consist of ongoing
costs that were previously accounted for in the MDI Rules. The
Commission estimates that the exclusive SIPs will incur a one-time cost
of approximatively $613,000 and ongoing costs of approximatively
$174,000 per year.\1669\
---------------------------------------------------------------------------
\1669\ The $613,000 estimate in one-time costs is based on the
following calculations: $33,000 (costs under the amendments to Rule
612 to update data specifications and internally and externally test
the updates; see supra note 1651, see also section VII.D.5.a) +
$167,670 ($83,790 (hourly rate for Sr. Programmer at $399 for 210
hours) + $61,740 (hourly rate for Sr. Systems Analyst at $343 for
180 hours) + $7,460 (hourly rate for Compliance Manager at $373 for
20 hours) + $5,880 (hourly rate for Director of Compliance at $588
for 10 hours) + $8,800 (hourly rate for Compliance Attorney at $440
for 20 hours) + $412,500 (costs for external services). The $174,000
estimate in ongoing costs is based on the following calculations:
$50,301 ($25,137 (hourly rate for Sr. Programmer at $399 for 63
hours) + $18,522 (hourly rate for Sr. Systems Analyst at $343 for 54
hours) + $2,238 (hourly rate for Compliance Manager at $373 for 6
hours) + $1,764 (hourly rate for Director of Compliance at $588 for
3 hours) + $2,640 (hourly rate for Compliance Attorney at $440 for 6
hours)) + $123,725 (costs for external services). See infra notes
1830, 1832, 1834 and 1835 and accompanying text for relevant details
on these cost estimates.
---------------------------------------------------------------------------
The Commission recognizes some uncertainty in the assumption that
exclusive SIPs will become competing consolidators. If one or both
exclusive SIPs are not planning to become competing consolidators under
the MDI Rules, and the amendments do not change their plans, then the
estimated initial and ongoing costs in table 15 represent new costs
associated with the amendments. If the amendments were to prompt one or
both exclusive SIPs to become competing consolidators, when they
otherwise would not have done so, then the costs in table 15
underestimate the total costs of these SIPs becoming competing
consolidators. In the MDI Rules, however, the Commission anticipated
that both exchanges operating exclusive SIPs would have strong
incentives to enter the competing consolidator market.\1670\ The
amendments require that the exclusive SIPs build out the capacity to
disseminate aspects of the data required by the MDI Rules. This could
increase the likelihood that the exclusive SIPs will choose to become
competing consolidators because they will already have implemented some
of the technology needed to comply with the requirements of a competing
consolidator, thereby lowering their subsequent cost of becoming a
competing consolidator. In this context, the Commission continues to
expect that the exclusive SIPs will become competing consolidators, and
the estimated costs in table 15 largely represent costs that the
exclusive SIPs would have borne in the process of becoming competing
consolidators.
---------------------------------------------------------------------------
\1670\ See MDI Adopting Release, supra note 10, at 18761.
---------------------------------------------------------------------------
The Commission recognizes that the amendment to Rule 600 could
increase the initial costs of becoming a competing consolidator as well
as the ongoing costs of competing consolidators, but the Commission
believes that such costs are already accounted for in the MDI
Rules.\1671\ In particular, competing consolidators could incur
additional compliance costs to estimate and disseminate the best odd-
lot order. To the extent such costs are not accounted for in the MDI
Rules, they will likely be a small fraction of the compliance costs of
including odd-lot information in SIP data stated above. Indeed, the
competing consolidators will already have the information necessary to
calculate the BOLO, so most of the cost incurred under the amendment to
Rule 600 will be the initial cost of coding the information and the
cost of processing and monitoring that code in real time.
---------------------------------------------------------------------------
\1671\ See supra section VII.D.5 for further discussion of how
or whether this requirement would alter the compliance costs of
competing consolidators.
---------------------------------------------------------------------------
6. Interactions With Recently Adopted Rules
The Commission acknowledges that the effects of any final rule may
be impacted by recently adopted rules that precede it. Accordingly,
each economic analysis in each adopting release considers an updated
economic baseline that incorporates any new regulatory requirements,
including compliance costs, at the time of each adoption, and considers
the incremental new benefits and incremental new costs over those
already resulting from the preceding rules. We discuss below economic
effects stemming from interactions between the final rule and other
recently adopted rules.\1672\
---------------------------------------------------------------------------
\1672\ As explained above, the Commission considers recently
adopted rules, but not recent proposals, as part of its baseline
against which it measures the economic effects of its rules. See
supra section VII.C and notes 1034 and 1047.
---------------------------------------------------------------------------
a. Amendments to Rule 605
Commenters have specifically questioned the Commission's analysis
of interactions between the four EMS Proposals,\1673\ of which only the
Rule 605 Proposal has been adopted.\1674\ Because the amendments to
Rule 605 were not yet adopted at the time of the Proposing Release and
were thus not a part of the baseline in the Proposing Release,\1675\
the economic effects described in the Proposing Release may differ from
those described here to the extent those effects change due to the
amendments to Rule 605. Below, we discuss specific impacts the
amendments to Rule 605, which is now part of the baseline, may have on
the expected economic effects of the final rules compared to
description of those effects in the Proposing Release, as well as the
impact the final rules may have on the effects of amended Rule 605. In
response to comments, we also consider whether the amended Rule 605
data is needed to assess the impact of the final rules. Overall, the
inclusion of the amendments to Rule 605 in the baseline does not
significantly change the costs and benefits of the final rules, and the
final rules adopted herein have significant benefits even taking into
account the adopted amendments to Rule 605 as part of the baseline.
---------------------------------------------------------------------------
\1673\ See supra section II for further discussion.
\1674\ See Rule 605 Amendments, supra note 10.
\1675\ See supra note 1672.
---------------------------------------------------------------------------
i. Impact of Amended Rule 605 on the Final Rules
First, the Commission considers whether the amendments to Rule 605,
which are now part of the baseline, may have affected certain benefits
of the final rules compared to how those benefits were described in the
Proposing Release. Some commenters stated that the four EMS Proposals,
including the amendments to Rule 605 and these final rules, have
similar objectives, such that the benefits of each rule may be
overlapping, and that therefore each successive rule would have fewer
benefits than were described in each proposing release.\1676\ However,
the
[[Page 81756]]
final rules address different and significant issues in the national
market system distinct from those addressed in Rule 605.\1677\ The
final rules have benefits, such as certain improvements in market
quality for stocks that receive a smaller tick size, lower trading
costs for liquidity demanders in certain stocks that experience a
reduction in their access fees, and increased transparency and reduced
complexity of exchange access fees and rebates, that are distinct from
the benefits resulting from the amendments to Rule 605 and could not
conceivably have been achieved through the amendments to Rule
605.\1678\ While, as one commenter stated,\1679\ both rules may improve
competition, the issues being addressed in these final rules and in the
amendments to Rule 605, and the mechanisms used to address them, differ
significantly, making these benefits additive rather than overlapping.
For example, the amendments to Rule 605 will increase competition among
trading venues through greater transparency,\1680\ while these final
rules will increase competition between orders on trading venues in
some stocks by removing barriers to sub-penny quoting. Both of these
competitive effects are expected to improve execution quality, but
through different mechanisms and independently of one another.
---------------------------------------------------------------------------
\1676\ See, e.g., Virtu Letter II at 55-56 (stating ``that the
proposals are designed to accomplish the same overarching goals,''
that ``each rule ignores the possibility that the other three rules
may already address the Commission's concerns,'' that ``the expected
benefit the Commission believes its rules will achieve is
overlapping,'' and that an important policymaking question that
arises from these overlapping objectives is whether ``the estimated
benefits [are] purely additive'') see also SIFMA Letter II at 100
(stating that ``the Proposed Rules may, each individually, largely
affect the same aspects of equity markets, including the economics
of liquidity provision, spreads (particularly for retail investors),
and costs (particularly for wholesalers)''); Virtu Letter II at 20
(stating that, if any of the other rules (including Rule 605) are
successful at achieving their stated purpose, ``competition would be
enhanced without the Proposed [Tick Size] Rule (and its significant
risks and costs) and the claimed benefits of the Proposed Rule are
overstated''); Virtu Letter III at 2 (stating that the amendments in
the Rule 605 Amendments may ``otherwise address any concerns that
formed the impetus for the [EMS] Proposals,'' including the
Proposing Release).
\1677\ See supra section II for further discussion.
\1678\ See supra sections VII.D.1.b, VII.D.2, and VII.D.3 for
further discussion of the benefits of the final rules and supra
section VII.C.5 for a discussion of the benefits resulting from the
amendments to Rule 605.
\1679\ See supra note 1676.
\1680\ See Rule 605 Amendments, supra note 10, at 26543-75.
---------------------------------------------------------------------------
In addition, the adoption of the amendments to Rule 605 may, to
some extent, enhance certain benefits of the final rules compared to
the benefits described in the Proposing Release. Specifically, the
final rules are expected to increase the extent to which broker-dealers
make decisions based on execution quality.\1681\ At the same time, the
amendments to Rule 605 improve broker-dealers' access to information
about the execution quality of market centers.\1682\ Thus, to the
extent that broker-dealers base their order routing decisions more on
execution quality as a result of the final rules, the improved access
to market center execution quality information under amended Rule 605
will help facilitate those decisions.\1683\
---------------------------------------------------------------------------
\1681\ Specifically, the increase in transparency in exchange
access fees and rebates in the final rules is expected to decrease
the extent to which broker-dealers' routing decisions are based on
access fees and rebates and increase the extent to which these
decisions are based on other factors, including the execution
quality of market centers See supra section VII.D.3.
\1682\ For example, the amendments to Rule 605 increase the
granularity of time-to-execution buckets, which will improve broker-
dealers' ability to compare execution speeds across trading venues
and route their orders accordingly. See Rule 605 Amendments, supra
note 10, at 26561.
\1683\ See Rule 605 Amendments, supra note 10, at 26544-26547
(discussing the impact of the amendments to Rule 605 on competition
between broker-dealers).
---------------------------------------------------------------------------
The adopted amendments to Rule 605 may also, in certain
circumstances, cause the benefits of the final rules stemming from
transparency to be somewhat lower than those described in the Proposing
Release; however, the Commission expects these impacts to be minor.
Specifically, in the Rule 605 Amendments, the Commission anticipated
that the increase in transparency and competition on the basis of
execution quality as a result of the amendments to Rule 605 might make
broker-dealers less likely to route customer orders based on exchange
fees and rebates.\1684\ At the same time, the amendments to Rule 610
that would make fees and rebates determinable at the time of execution
are expected to reduce broker-dealer conflicts of interest related to
fees and rebates.\1685\ Likewise, the lower access fee cap is also
expected to reduce broker-dealer conflicts of interest.\1686\ If the
amendments to Rule 605 result in exchange fees and rebates becoming
less important for broker-dealer customers' order routing decisions,
the benefits resulting from a reduction in conflicts of interest under
the final rules--caused by reducing the access fee cap and increasing
the transparency of exchange fees and rebates--may be reduced compared
to how they were described in the Proposing Release. However, this
reduction in benefits, compared to the Proposing Release, is likely to
be minor, for several reasons, and the Commission still expects the
benefits described above, in comparison to the baseline, to be
realized. First, not all orders are subject to and directly benefit
from increased transparency under amended Rule 605.\1687\ Second, the
amended Rule 605 reporting requirements only require reporting by
larger broker-dealers; while these broker-dealers handle the vast
majority of customer accounts, they only handle around 60% of customer
order flow in terms of number of orders.\1688\ Therefore, the
amendments to Rule 605 are not expected to directly impact customer
order routing decisions for a significant subset of order flow,\1689\
such that the final rules lowering the access fee cap and increasing
the transparency of exchange fees and rebates are still expected to
have additional benefits above and beyond those of the amendments to
Rule 605.
---------------------------------------------------------------------------
\1684\ See Rule 605 Amendments, supra note 10, at 26586.
\1685\ See supra section VII.D.3 for a discussion of the
economic effects of requiring exchange fees and rebates to be
determinable at the time of execution.
\1686\ See supra section VII.D.2 for a discussion of the
economics effects of reducing the access fee cap on conflicts of
interest.
\1687\ For example, in the Rule 605 Amendments, an analysis of
Tick Size Pilot data found that, between April 2016 and March 2019,
approximately 25% of orders were flagged as having special handling
requests, which would exclude them from the scope of Rule 605
reporting requirements. See Rule 605 Amendments, supra note 10, at
26514.
\1688\ The Rule 605 Amendments estimated that only 85 out of
1,245 broker-dealers with at least one customer account would
qualify as a larger broker-dealer and therefore be required to
prepare Rule 605 reports; however, these 85 broker-dealers are
responsible for more than 98% of customer accounts and more than 60%
of customer orders. See Rule 605 Amendments, supra note 10, at 26428
(table 13).
\1689\ The Rule 605 Amendments acknowledge that, if smaller
broker-dealers are also incentivized to produce execution quality
information for their customers as a result of the expanded scope of
Rule 605 to include larger broker-dealers, the benefits of increased
competition could indirectly extend to smaller broker-dealers as
well. See Rule 605 Amendments, section IX.C.1.(D)(1), supra note 10,
at 26428.
---------------------------------------------------------------------------
Third, the Commission anticipated that the amendments to Rule 605,
by expanding the scope of covered orders to include odd-lots, would
encourage broker-dealers to compete for these orders on the basis of
execution quality.\1690\ The final rule's requirement that the
exclusive SIPs disseminate odd-lot data \1691\ is expected to
accelerate the benefits of accelerating the implementation of including
odd-lot information in NMS data,\1692\ which may include facilitating
better execution quality for these orders by broker-dealers who newly
have access to information about odd-lots.\1693\ To the extent that the
increase in competition for odd-lot execution quality under amended
Rule 605 has already incentivized broker-dealers to increase their
usage of existing sources of odd-lot data (such as proprietary data
feeds) in routing decisions, this would reduce the number of broker-
dealers without pre-existing access to odd-lot information and thus may
reduce the benefits from
[[Page 81757]]
disseminating odd-lot information in the SIP as described in the
Proposing Release. However, if broker-dealers that rely on odd-lot
information from proprietary data feeds are able to reduce their costs
by switching to using odd-lot information from the SIP,\1694\ this
would result in benefits even to those broker-dealers with pre-existing
access to odd-lot information.\1695\
---------------------------------------------------------------------------
\1690\ See Rule 605 Amendments, supra note 10, at 26552.
\1691\ See 17 CFR 242.603(b)(3) for rule text relating to this
requirement under Rule 603.
\1692\ See supra section VII.D.4.c.
\1693\ See supra note 1151 and corresponding text. See MDI
Adopting Release, supra note 10, at 18753, stating that ``if a
broker-dealer previously did not have access to odd-lot information,
then a broker-dealer receiving the additional information may help
facilitate best execution of its clients' orders.''
\1694\ See MDI Adopting Release, supra note 10, at 18753 and
18793-95 (discussing market participants substituting MDI odd-lot
information for exchange proprietary data feeds).
\1695\ As another example, the Rule 605 Amendments stated that
one indirect effect of the amendments to Rule 605 might be an
increase in incentives for reporting entities to compete in areas
other than improved execution quality, including lowering their
access fees. See Rule 605 Amendments, supra note 10, at 26575. This
could also reduce incentives to route based on fees and rebates,
which would reduce the benefits of increased transparency under the
final rules. However, this would only be the case in limited
circumstances, i.e., when exchanges are not able to differentiate
themselves based on execution quality. Furthermore, the Rule 605
Amendments also acknowledged that Rule 605 reporting entities may
pass some of the costs of amended Rule 605 on to their customers.
See Rule 605 Amendments, supra note 10, at 26586. If exchanges pass
on their compliance costs by raising their access fees, then the
benefits of the final rules may be heightened by the adopted
amendments to Rule 605, rather than lessened.
---------------------------------------------------------------------------
ii. Impact of the Final Rules on Amended Rule 605
In addition to the impact amended Rule 605 may have on the effects
of the final rules compared to those described in the Proposing
Release, the Commission also considered the reverse, i.e., whether the
final rules may impact the effects of amended Rule 605 going forward.
Specifically, the final rules will enhance certain benefits and reduce
certain costs of amended Rule 605. As discussed above, the amendments
accelerating MDI Rules related to including information about odd-lots
into SIP data will accelerate the realization of the benefits of this
information.\1696\ In turn, the MDI Rules increase the usefulness of
price improvement statistics included in amended Rule 605 using the
best available displayed price as the benchmark by providing market
participants with price improvement information relative to a benchmark
price that more accurately reflects liquidity available in the
market.\1697\ Increasing the usefulness of price improvement statistics
promotes incentives for reporting entities to seek out or offer price
improvement relative to the best displayed price, taking into account
all available displayed liquidity (including odd-lots).
---------------------------------------------------------------------------
\1696\ See supra section VII.D.4.c.
\1697\ See 17 CFR 242.600(b)(14) (defining the ``best available
displayed price'') and 17 CFR 242.605(a)(1)(ii)(M) through (Q); see
also Rule 605 Amendments, supra note 10, section III.B.4(g) for
further discussion of these amendments.
---------------------------------------------------------------------------
In addition, the value of reporting price improvement relative to
the best displayed price relative to the NBBO, now required by amended
Rule 605,\1698\ will increase for those stocks for which the reductions
in the tick size in the final rules result in an increase in the number
of price levels within the spread.\1699\ If there are more price
increments within the spread, it is more likely that the best displayed
price will be different from the NBBO. Similarly, the availability of a
greater number of price increments within the spread increases the
value of the separate reporting of execution quality information for
midpoint-or-better NMLOs \1700\ because the prevalence of these orders
is likely to increase.\1701\ Furthermore, if a reduction in the tick
size results in a reduction of depth at the NBBO, this increases the
usefulness of the recently adopted measures of size improvement
included in amended Rule 605 reports.\1702\ Finally, the final rule
requiring NMS data to also include information on the best priced odd-
lot orders across all markets \1703\ will reduce ongoing compliance
costs related to compiling information about price improvement relative
to the best displayed price under amended Rule 605.\1704\ This is
because reporting entities will be able to access standardized
information about the best odd-lot order, rather than needing to use
odd-lot trade and quote data to calculate the best odd-lot order
themselves.\1705\
---------------------------------------------------------------------------
\1698\ See supra note 1176 and corresponding text.
\1699\ See supra section VII.D.1.b for further discussion.
\1700\ See supra note 1179 and corresponding text.
\1701\ An analysis of CAT data in the Rule 605 Amendments found
that, in the quartile of stocks with the lowest quoted spreads (an
average quoted spread of around $0.026), midpoint-or-better orders
still compromise a non-negligible percent of order flow,
representing 5.15% of submitted orders (4.32% of submitted shares).
This is compared to the quartile with the highest quoted spreads (an
average quoted spread of $20.26), where midpoint-or-better orders
are 9.66% of submitted orders (8.62% of submitted shares). See Rule
605 Amendments, supra note 10, at 26428 n.1448.
\1702\ See supra note 1182 and corresponding text.
\1703\ See supra section VII.D.4.d.
\1704\ See supra note 1178.
\1705\ The amendments to include in Rule 605 information about
price improvement relative to the best displayed price, size
improvement, and beyond-the-midpoint NMLOs (which are a subset of
midpoint-or-better NMLOs) were also considered in the Rule 605
Proposal; see Rule 605 Proposal, supra note 117, at 3817, 3819, and
3810. The Commission acknowledges that, to the extent that it
occurs, an increase in the cost of processing and storing
consolidated market data may be higher for larger broker-dealers,
who will be required to prepare Rule 605 reports for the first time
under the adopted amendments to Rule 605. As a result, the
additional cost of preparing Rule 605 reports may be higher for
these broker-dealers as a result of the final rules. See supra
section VII.D.1.c for a discussion of how the final rules may
increase the cost of processing and storing consolidated market
data.
---------------------------------------------------------------------------
One commenter stated that because of the proposed amendments to the
definition of ``categorized by order size'' in Rule 605,\1706\
subsequent changes to the definition of tick sizes and round lots would
``create customer confusion'' regarding their Rule 605 reporting
requirements.\1707\ The Commission does not believe that market centers
and brokers-dealers will be confused about their reporting obligations
under amended Rule 605 as a result of the new round lot definition and
the new minimum tick size under the final rules. The rule texts for
both amended rules are clearly stated. Further, the use of notional
value in the order size categories under the adopted amendments to Rule
605 will help end users of these reports understand the effect of a
change in round lot size for a security because a notional value range
will remain constant even if the size of a round lot changes.\1708\
---------------------------------------------------------------------------
\1706\ The final amendments to Rule 605 likewise included an
amended definition of ``categorized by order size'' that requires
orders to be categorized according to whether they are round lots.
See Rule 605 Proposal, supra note 117, at 3807; proposed Rule
600(b)(19). As amended, rather than requiring the reporting of order
sizes in terms of whether an order was less than one share, an odd-
lot, or in one of five categories based on numbers of round lots,
final Rule 605 requires the reporting of order sizes in terms of
notional values, with each order size category further separated
into whether an order is a round lot, odd-lot, or fractional order,
for a total of 24 reporting categories. See Rule 605 Amendments,
supra note 10, at section III.B.1; adopted Rule 600(b)(18). Prior
Rule 600(b)(13) required reporting of order sizes in one of four
categories based on numbers of round lots, with no reporting of
fractional orders or odd-lots.
\1707\ See Tastytrade Letter at 5.
\1708\ See Rule 605 Amendments, supra note 10, at 26428 n.375.
It may be the case that, within a given notional order size bucket
in Rule 605 reports, the distribution of orders across round-lot and
odd-lot categories may change for some stocks following the
implementation of the new round lot definition.
---------------------------------------------------------------------------
iii. Delaying the Final Rules Until Amended Rule 605 Data are Available
Third, in response to comments, the Commission considers whether
adoption of the final rules should be delayed until amended Rule 605
data are available.\1709\ Several commenters suggested that the
Commission wait to adopt the final rules until after the amended Rule
605 data are available so that amended Rule 605 data could be
[[Page 81758]]
used to assess whether the final rules are necessary.\1710\
---------------------------------------------------------------------------
\1709\ See supra section VII.C.5 for discussion of the
implementation timeline for the adopted amendments to Rule 605. See
supra section VI for further discussion of the compliance dates for
the final rules.
\1710\ See, e.g., Virtu Letter II at 24 (recommending that ``the
Commission amend Rule 605 to provide more comprehensive execution
quality statistics on retail activity based on input from investors
and market participants, and then pause to study and assess market
quality based on the newly collected data before determining whether
to move forward with the Proposed Rule''); Citadel Letter II at 1-2
(stating that ``the data provided pursuant to an updated Rule 605
should be the primary input in determining whether the other
proposals are necessary)'' See also Letter from Ellen Greene,
Managing Director, Equities & Options Market Structure, and Joseph
Corcoran, Managing Director, Associate General Counsel, SIFMA, dated
Aug. 14, 2024 at 3; SIFMA Letter II at 3; Virtu Letter II at 1-2;
and comments discussed in supra note 122 and accompanying text.
---------------------------------------------------------------------------
Although the information disclosed under Rule 605 is a significant
source of information about execution quality, the Commission did not
rely on, and does not believe that it is necessary to rely on, Rule 605
data (either adopting or pre-existing) in its analyses in the Proposing
Release or in the adoption of the final rules.\1711\ Instead, the
Commission utilized other data sources for conducting the relevant
analyses, including with respect to execution quality, which it
believes has sufficiently informed the Commission and the public on the
issues being addressed in the final rule.\1712\ Other commenters
suggested waiting until after the amended Rule 605 data is available so
that amended Rule 605 data could be used to evaluate the impact of the
implementation of the final rules.\1713\ The Commission acknowledges
that Rule 605 data is an important source of public information about
order execution quality. However, as stated by another commenter, there
are other data products that provide relevant information on execution
quality that can be used to evaluate the impact of the final
rules.\1714\ Waiting until amended Rule 605 data are available to adopt
the final rules would delay the significant benefits of the final rules
to be realized, and the Rule 605 Amendments cannot and do not solve the
main concerns that the final rules address by reducing the tick size,
lowering the access fee cap, and accelerating the round lot definition
are designed to solve.
---------------------------------------------------------------------------
\1711\ This was supported by a commenter, who stated that
``better data from Rule 605 reports, among other sources, could be
useful in making additional decisions about tick size in the future.
But it is certainly not needed to decide whether to make changes to
the tick size now.'' See IEX Letter I at 4.
\1712\ Data used in the Proposing Release, supra note 11,
included CAT data, see, e.g., Proposing Release at 80340 n.673;
MIDAS data, see, e.g., id. at 80297 Tables 1, 2; TAQ data, see,
e.g., id. at 80313 Table 6; WRDS intraday indicators, see, e.g., id.
at 80316 Table 8; Rule 606(a)(1) reports, see, e.g., id. at 80306
n.467; and Tick Size Pilot data and Rule 606(a)(1) reports, see,
e.g., id. at 80320 Table 9. While the Proposing Release included an
estimate of the number of trading venues who report Rule 605
statistics, see id. at 80333, Rule 605 data itself was not used. One
commenter stated that the Proposing Release, ``relies, in part, on
data from Rule 605 reports--which use metrics that the Commission
has acknowledged are deficient and in need of modification.'' The
commenter proceeded to cite the Proposing Release at 80321 n.557,
which discusses the horizon over which realized spreads are
calculated in both the TSP analysis and in Rule 605 reports. See
Virtu Letter II at 5 and n.8. The realized spread is a common and
useful metric that will continue to be reported under the amendments
to Rule 605. While both the TSP analysis and Rule 605 reports
calculate realized spreads using a five-minute horizon, the
commenter is incorrect in stating that the Proposing Release relied
on data from Rule 605 reports. In the TSP analysis, the Commission
calculates realized spreads using trade and quote data from TAQ;
when completing Rule 605 reports, trading centers calculate realized
spreads using similar trade and quote data. Rule 605 reports are
therefore not necessary to obtain the realized spreads used in the
TSP analysis.
\1713\ See, e.g., SIFMA Letter II at 3 (stating that a
quantitative analysis ``can only be done effectively after the
implementation and operation of the proposed amendments to Rule 605
to allow the Commission and the public to measure the impact of
modified tick sizes and/or access fee caps''); see also Citadel
Letter I at 29 (stating that ``if both proposals were to be
finalized, it appears that market participants and regulators would
be unable to accurately assess the true impact of the market
structure changes contained in this Proposal, precluding an `apples-
to-apples' before-and-after comparison'').
\1714\ See IEX Letter III at 3 (``There are myriad sources of
information that both regulators and market participants draw on to
consider how orders are handled and how markets compete with and
compare to each other.''). See also supra section II for additional
discussion.
---------------------------------------------------------------------------
b. Implementation Costs From Overlapping Compliance Periods
Several commenters stated that the Commission should consider the
cumulative costs of implementing the proposed amendments and other
recent Commission rules and proposed rules.\1715\ Specifically, one
commenter requested that the Commission ``publish a thorough analysis
of the cumulative effects of the Interconnected Rules that accounts for
interconnections and dependencies among them and any other rules the
Commission has proposed or intends to propose in the near term,'' and
``tak[e] into account not just the expected effects on investors and
our capital markets but also practical realities such as implementation
timelines as well as operational and compliance requirements.'' \1716\
We consider here recently adopted rules, including the Settlement Cycle
Adopting Release, February 2024 Form PF Adopting Release, May 2023 SEC
Form PF Adopting Release, Dealer Adopting Release, Rule 605 Amendments,
Beneficial Ownership Adopting Release, Rule 10c-1a Adopting Release,
Short Position Reporting Adopting Release, Treasury Clearing Adopting
Release, and Customer Notification Adopting Release.\1717\
---------------------------------------------------------------------------
\1715\ See supra note 1035.
\1716\ ICI Letter II.
\1717\ See supra section VII.C.
---------------------------------------------------------------------------
Consistent with its long-standing practice, the Commission's
economic analysis in each adopting release considers the incremental
benefits and costs for the specific rule--that is the benefits and
costs stemming from that rule compared to the baseline. The Commission
acknowledges the possibility that complying with more than one rule may
entail costs that could exceed the costs if the rules were to be
complied with separately. Two of the rules identified by commenters
have compliance dates that occur before the effective date of the final
amendments.\1718\ The compliance periods for other rules overlap in
part, but the compliance dates adopted by the Commission in recent
rules are generally spread out over an approximately two-year period
extending to June 2026,\1719\ which could limit the number of
implementation activities occurring simultaneously. Where overlap in
compliance periods exists, the Commission acknowledges that there may
be additional costs on those entities subject to one or more other
rules as well as implications of those costs, such as impacts on
entities' ability to invest in other aspects of their businesses.\1720\
---------------------------------------------------------------------------
\1718\ The compliance date for the May 2023 SEC Form PF Adopting
Release occurred on December 11, 2023 and June 11, 2024, and the
compliance date for the Settlement Cycle Adopting Release occurred
on May 28, 2024.
\1719\ See supra section VII.C.
\1720\ See, e.g., MFA Comment Letter II (asserting that the
adoption of multiple proposals would impose ``unprecedented
operational and other practical challenges'').
---------------------------------------------------------------------------
Affected entities subject to the amendments may be subject to one
or more of the other recently adopted rules depending on whether those
entities' activities fall within the scope of the other rules.
Specifically, the Rule 605 Amendments, which require disclosures for
order executions in NMS stocks, affects market centers and certain
larger broker-dealers that were not required to publish Rule 605
reports prior to the Rule 605 amendments. The Beneficial Ownership,
Rule 10c-1a, Short Position Reporting, Dealer, and Customer
Notification Adopting Releases also apply to certain brokers and
dealers \1721\--although due to differing
[[Page 81759]]
requirements, these rules may not all apply to any given broker or
dealer. The Treasury Clearing Adopting Release applies to certain
participants of the covered clearing agencies which could include
broker-dealers.\1722\ We acknowledge that entities subject to multiple
rules may still experience increased costs associated with implementing
multiple rules at once as well as implications of those costs, such as
impacts on entities' ability to invest in other aspects of their
businesses.
---------------------------------------------------------------------------
\1721\ See Beneficial Ownership Adopting Release, supra note
1029 at 76897, 76945; Rule 10c-1a Adopting Release, supra, note 1030
at 75647, 75717-18; Short Position Reporting Adopting Release, supra
note 1031 at 75150; Dealer Adopting Release, supra note 1028 at
14938, 14967-71; Customer Notification Adopting Release, supra note
1034 at 47689, 47725.
\1722\ See Treasury Clearing Adopting Release, supra note 1045,
at 2717, 2791.
---------------------------------------------------------------------------
In addition, while the Commission received comments on the
interaction of the MDI Rules and these amendments,\1723\ commenters did
not specifically address costs associated with overlapping compliance
periods. When the Commission adopted the MDI Rules, it outlined a
phased transition plan for implementation.\1724\ Based on the times
provided in the transition plan for implementation of the MDI Rules,
the Commission estimated that the full implementation of the MDI Rules
will be at least two years after the Commission's approval of the plan
amendment(s) required by Rule 614(e).\1725\ Therefore, the length of
time affected market participants will have to come into compliance
with both the MDI Rules and these amendments, and the likelihood of
limited overlap in compliance periods, will mitigate compliance
costs.\1726\
---------------------------------------------------------------------------
\1723\ See, e.g., Citadel Letter I at 26; SIFMA Letter II at,
e.g., 41-42; Robinhood Letter at 44.
\1724\ See MDI Adopting Release, supra note 10, at 18699-18701.
\1725\ See supra note 1139 and accompanying text.
\1726\ See also section VII.D.4 (discussing the acceleration and
implementation of the MDI Rules).
---------------------------------------------------------------------------
One commenter stated that the complete implementation of the MDI
Rules will undermine the Commission's economic analysis of amendments
to Rules 610 and 612. The commenter stated: ``Implementation of the MDI
Rules would . . . likely mute any potential benefits of or weaken the
case for the additional costs associated with the Tick Size Proposal .
. . At minimum, the Commission is obligated to consider the fully
implemented MDI Rules as part of the `baseline' against which the
asserted need for this new rule, and its impact, are assessed.'' \1727\
---------------------------------------------------------------------------
\1727\ See Robinhood Letter at 44.
---------------------------------------------------------------------------
The MDI Rules form part of the baseline for the amendments to Rules
610 and 612. The MDI Rules and these amendments will increase
transparency, achieve better order execution, lower costs, and lead to
better investment decisions and increased market efficiency. However,
the MDI Rules and these amendments--while sharing broad goals--achieve
their benefits through distinct channels; therefore, implementation of
the MDI Rules is unlikely to mute benefits arising from these
amendments. The channel through which the MDI Rules achieves their
benefits is data--the MDI Rules will increase the granularity of data
that is included in NMS data and further introduce a decentralized
competing consolidator model to lower the cost of purchasing this data.
For market participants who already purchase proprietary data from
exchanges, the MDI Rules may have a limited direct impact because these
participants will not experience a change in their information set.
Nonetheless, these participants will still see many benefits from the
reduction in tick size and access fee cap. For example, they will be
able to quote at more precise prices, which more accurately reflect the
competitive cost of liquidity, and experience fewer instances of tick
constraints. Also, any stocks that remain tick-constrained would have
fewer distortions due to excess liquidity as a result of the reduction
in the access fee cap. Therefore, the benefits of the MDI Rules do not
lessen the benefits of the amendments to Rules 610 and 612 for these
market participants.
For market participants who do not purchase proprietary data from
exchanges, the MDI Rules, once fully implemented, as well as the market
data amendments associated with this release, may complement the
benefits of amendments to Rules 610 and 612. This is because the MDI
Rules, including the amendments in this release, will result in more
information being made available to non-consumers of proprietary data,
while amendments to Rules 610 and 612 remove constraints on trading by
alleviating tick constraints. The amendments in this release to Rules
610 and 612, the amendments to the MDI Rules, and the MDI Rules in
total (once fully implemented), will create an environment where there
is more, as well as better, information available to market
participants that do not subscribe to proprietary data products. These
market participants will also have an improved ability to trade on that
information due to the lower expected cost of transacting for some
stocks due to the lower tick size and access fee. The eventual
inclusion of depth-of-book information in core data under the MDI Rules
will also mitigate costs from the tick reduction causing liquidity to
spread across multiple price points \1728\--market participants will
more readily be able to see the liquidity available at these price
points.
---------------------------------------------------------------------------
\1728\ See section VII.D.1 for a discussion of this effect.
---------------------------------------------------------------------------
E. Effect on Efficiency, Competition, and Capital Formation
1. Efficiency
The amendments will improve price efficiency, namely the degree to
which the price of a stock reflects its fundamental value. The
improvement in price efficiency is expected largely to come through the
reduction in the tick size and the reduction of the access fee cap. The
acceleration of portions of the MDI Rules could also increase price
efficiency, but those effects are largely to accelerate the economic
impact already anticipated in the MDI Rules.
Lowering the tick size for some NMS stocks with prices equal to or
greater than $1.00, as well as lowering the access fee cap for all
stocks to 10 mils for stocks with prices equal to or greater than
$1.00, or to 0.10% for stocks with prices lower than $1.00, will
increase price efficiency.\1729\ The reduction in the tick size for
some stocks along with the reduction of the access fee cap for all
stocks will lower transaction costs.\1730\ When trading becomes less
costly, market participants have an increased incentive to gather more
information because doing so is more profitable.\1731\ Gathering more
information and trading on that information means that prices are more
reflective of the fundamental value of the firm. Consequently, for
stocks that receive an improvement in market quality due to the lower
tick size or the reduction in the access fee, the Commission expects an
improvement in price efficiency.\1732\
---------------------------------------------------------------------------
\1729\ See NASAA Letter at 9, and Vanguard Letter at 6.
\1730\ As discussed in section VII.B, the reduction in the fee
cap will primarily lower trading costs and distortions for stocks
that are tick-constrained; for stocks that are not tick-constrained,
the reduction in the fee cap may lead to wider quoted spreads which
will offset the lower fee. See supra note 1508 for a discussion of
the fee cap's contribution to liquidity distortions for tick-
constrained stocks; see also supra section VII.D.2.c on the effect
of the fee cap reduction for stocks that are not tick-constrained.
\1731\ See, e.g., Dixon, supra note 1277 for a discussion of
this concept in the context of short selling.
\1732\ Id.
---------------------------------------------------------------------------
The Commission further expects that quoted spreads will better
reflect the cost of liquidity as a result of these amendments. As
discussed in section
[[Page 81760]]
VII.B.3,\1733\ high access fees and rebates can distort liquidity
supply and demand.\1734\ The tick acts as a price floor that prevents
the spread from reflecting the true cost of liquidity; likewise, access
fees--and the rebates they fund--further distort quoted spreads by
taxing liquidity demand and subsidizing liquidity supply at this price
floor. By reducing both the tick and the access fee cap, quoted spreads
will better reflect the cost of liquidity and allow for more efficient
liquidity provision.
---------------------------------------------------------------------------
\1733\ See also Proposing Release, note 11, at 80328: ``If tick
sizes were infinitely small, and absent other distortions, then fees
and rebates would not affect the cost of trading because markets
would simply adjust quotes by the amount of the rebate such that the
spread with rebates included is the same. However, current U.S.
equity markets differ from this frictionless construct because there
is a finite tick. In this environment, and particularly for stocks
with narrower spreads, high access fees and rebates can distort
liquidity supply and demand by artificially increasing the cost of
taking liquidity and the revenue to providing liquidity. This
dynamic creates an environment with too much liquidity supply
relative to liquidity demand.''
\1734\ One commenter stated that the Commission did not explain
these distortions, see Virtu Letter II at 16. The subsequent
discussion (as well as the discussion surrounding note 1508, supra)
explains the distortions.
---------------------------------------------------------------------------
Some commenters expressed the concern of diminished intra-tick
pricing from reduced access fees.\1735\ As shown in the Proposing
Release, the NYSE, Nasdaq, and Cboe exchange families each operate both
a maker-taker venue and an inverted venue.\1736\ Variation in fee and
rebate schedules across trading venues effectively allow for intra-tick
pricing.\1737\ However, that intra-tick pricing via access fees and
rebates is an imperfect solution to the problem of a stock having a
tick that is too large.\1738\ Reducing the access fee cap can reduce
the degree to which maker-taker trading centers can offer effective
intra-tick pricing which could potentially lead to pricing
distortions--i.e., less efficient prices.\1739\ However, the ability
for markets to establish efficient prices will increase overall due to
the reduced tick size for stocks with TWAQS less than or equal to
$0.015. The reduced tick size reduces the need for intra-tick pricing
by providing a finer pricing grid on which market participants can
submit orders. Specifically, a tick size of $0.005 with an access fee
and rebate of 10 mils provides an effective pricing grid with price
pints at every $0.005 and plus or minus $0.001, which is finer than the
current pricing grid with a tick size of $0.01 and an access fee/rebate
of $0.003.\1740\ For stocks not subject to the reduced tick size, the
Commission acknowledges some reduction in the ability to price intra-
tick. However, the efficiency loss is limited due to the fact that, by
definition, these stocks have sufficiently wide average spreads, and
thus sufficient ticks within the spread, to avoid qualifying for the
lower tick size. For these stocks, there is less of a need to price
intra-tick.
---------------------------------------------------------------------------
\1735\ See also Citadel Letter I at 16, CCMR Letter at 26.
\1736\ See 87 FR 80313.
\1737\ See also supra note 1128.
\1738\ See Quarter Penny Tick, supra note 1127.
\1739\ See supra note 1128 for an explanation of how variation
in fee and rebate schedules across trading venues can increase price
fidelity by allowing for more effective intra-tick pricing. Reducing
the access fee cap reduces this effective intra-tick pricing by
limiting the degree to which fee and rebate schedules can differ
from one another.
\1740\ Under the preexisting Rule 610 fees, maker-taker venues
can offer liquidity demanders net prices that are 30 mils worse than
quoted prices; inverted venues, in contrast, tend to offer net
prices that are 30 mils better than quoted prices, creating a grid
of price points that are 40 to 60 mils separated from each other.
E.g., the net prices available at a maker-taker venue may be
$10.003, $10.013, etc., while the net prices available at an
inverted venue may be $9.997, $10.007, etc.; the price points are 40
to 60 mils apart from each other. Under the reduced fee cap, maker-
taker venues will only be able to offer liquidity demanders net
prices that are 10 mils worse than quoted prices; to maintain a grid
of price points that are 50 mils apart, an inverted venue could
offer a net price that is 40 mils better than the quoted price by
instituting a 40 mil rebate for liquidity takers. E.g., the net
prices available at a maker-taker venue will be $10.001, $10.013,
etc.; with a 40 mil rebate, inverted venues could offer net price
points of $9.996, $10.006, etc., so that the price points are 50
mils apart from each other. See also Quarter Penny Tick, supra note
1127, for a discussion of this concept.
---------------------------------------------------------------------------
The reduction in the tick is expected to reduce the cost of
transacting on exchanges, which will result in an increase in orders
being sent to lit markets.\1741\ The reduction in the access fee cap is
expected to widen quoted spreads, and the quote differential between
lit exchanges and other trading venues may widen. This need not,
however, lead to lower demand for lit liquidity, as the access fee,
which currently disincentivizes investors to access liquidity on
exchanges, will also be lower.\1742\ Thus, the net effect of the rules
is an increase in orders on lit markets and an improvement in price
efficiency.
---------------------------------------------------------------------------
\1741\ See infra section VII.E.2.a and VII.E.2.b for additional
discussion. Specifically see discussion surrounding infra note 1752.
\1742\ See, e.g., IEX Letter VI at 1-6.
---------------------------------------------------------------------------
Some commenters stated that a finer tick size could cause
information leakage. Large orders may need to be divided into smaller
orders due to the fragmentation of liquidity across multiple price
levels; executing a large order may therefore reveal more information,
which could increase price impact and trading costs.\1743\ The
Commission acknowledges that information leakage may, in turn, reduce
incentives to collect information ex-ante and thereby reduce price
efficiency. However, on balance, the effect of the amendments to
increase price efficiency by, on average, reducing trading costs. This
is because the cost of information leakage primarily shows up in the
form of wider spreads when a large trade is anticipated and higher
costs of trading for large orders. If information leakage was a
primary--rather than mitigating--effect of the smaller tick size, then
in the TSP analysis discussed in section VIII.D.1.b.ii spreads would
not have narrowed and round-trip trading costs would not have fallen
for stocks with narrow spreads when the tick size was reduced. But this
is not what that analysis found; thus, while increased information
leakage could be a factor mitigating the reduction in spreads due to
the smaller tick size, the net effect is expected to be lower trading
costs for stocks receiving the smaller tick size. Additionally, the
commenter's concerns are mitigated by the fact that the amendments do
not include the proposed smaller tick sizes (0.001 and 0.002).
---------------------------------------------------------------------------
\1743\ See supra notes 1219, 1220 and 1398, as well as related
discussion in sections VII.D.1.b.i and VII.D.1.e. The Commission
acknowledges that tick-constrained stocks receiving a tick reduction
may experience an increase in execution costs for sufficiently large
orders--evidence from the TSP suggests that this occurs for orders
that are quite large (approximately 50 round lots). See discussion
surrounding note 1284.
---------------------------------------------------------------------------
Lowering access fees also increases the efficiency with which the
quote conveys information regarding the cost of liquidity. As discussed
in section VII.D.2.d, access fees that fund rebates contribute to
complexity and lack of transparency in markets because they separate
both the true cost of demanding liquidity and the proceeds from
supplying liquidity, as represented by the quoted half-spread. Reducing
the wedge between the spread and the true cost also reduces conflicts
of interest between broker-dealers and their customers. Multiple
commenters stated that a lower access fee cap would help mitigate the
conflict of interest because lowering the access fee is expected to
reduce rebates available and thus the incentive to route based on
rebates instead of execution quality.\1744\
---------------------------------------------------------------------------
\1744\ See RBC Letter at 4 stating (``we believe that lowering
the access fee cap would lead to a reduction in broker conflicts of
interest . . .''); NASAA Letter at 9 (stating ``reducing access fee
caps could help reduce incentives for broker-dealers to route orders
to trading venues that benefit those broker-dealers (such as venues
in which a broker-dealer is rebated), but may provide suboptimal
execution to the detriment of the broker-dealer's customers.''). See
also Themis Letter at 7; supra section VII.D.3.
---------------------------------------------------------------------------
[[Page 81761]]
Making fees and rebates determinable at the time of execution,
along with the reduction of the access fee cap could also increase
price efficiency by helping minimize potential conflicts of
interest.\1745\ Fees and rebates create a potential conflict for a
broker in situations where incentives related to transaction fees,
which are paid by the broker, potentially conflict with incentives to
obtain execution quality, which may affect the customer.\1746\ This
conflict, if acted on, can lead to inefficient order routing and worse
transaction outcomes for customers; \1747\ it can also lead to an
inefficient incorporation of information into stock prices, harming
market efficiency. Making access fees determinable at the time of
execution will enhance efficiency by providing market participants with
certainty concerning the fees that they will be charged per
transaction. This certainty could also allow broker-dealers to examine
their own best-execution performance more efficiently. Greater
certainty about fees and rebates in advance of routing an order could
also increase the efficiency of the broker-dealers' best execution
assessments by providing them with greater certainty about the full
cost of a transaction when executing the order. Additionally, to the
extent that determinable fees make it easier for broker-dealers to
communicate fees and transmit them to end customers, doing so could
help eliminate distortions that might occur due to potential conflicts
of interest. However, to the extent that exchanges are not able to as
effectively replicate some incentives that were based on using
historical activity,\1748\ requiring these fees to be determinable at
time of execution may reduce efficiency.
---------------------------------------------------------------------------
\1745\ One commenter stated that they agreed with the
Commission's analysis of these effects, see Council of Institutional
Investors Letter at 5; see also Themis Letter at 7.
\1746\ See Vanguard Letter at 6. See also supra sectionVII.D.2
for additional discussion of access fees.
\1747\ See Retirement Coalition Letter at 2.
\1748\ See supra section VII.D.3 for further discussion on
basing fees and rebates on historical activity.
---------------------------------------------------------------------------
Accelerating the addition of odd-lot information to NMS data and
the inclusion of information relating to the best odd-lot quote will
realize some of the price efficiency benefits articulated in the MDI
Rules at an earlier date, providing improved price efficiency earlier
than anticipated in the MDI Rules. Specifically, research suggests that
adding information on the shares available at price levels inside the
NBBO may improve price efficiency.\1749\ Currently only market
participants who subscribe to proprietary data feeds can view the odd-
lot information and thus adjust trading strategies and decisions based
on that information. Expanding the SIP feeds to include odd-lot
information will provide new information to those investors who
subscribe to the SIP data but do not subscribe to proprietary data
feeds.\1750\
---------------------------------------------------------------------------
\1749\ See Robert P. Bartlett et al., The Market Inside the
Market: Odd-Lot Quotes, Rev. Fin. Stud. (Sep. 19, 2023), available
at https://doi.org/10.1093/rfs/hhad074.
\1750\ There is the possibility that competing consolidators may
not choose to distribute odd-lot information (because the MDI Rules
do not require them to do so), in which case the positive effects of
including odd-lots in NMS data on price efficiency will be lost.
This outcome is unlikely because the odd-lot information is valuable
in terms of having information relevant to stock prices, see
Bartlett et al., id, and the alternative to odd-lot information from
the competing consolidators would be to subscribe to all of the
proprietary data feeds, which is expensive. Given that there will be
significant demand for the odd-lot information, competing
consolidators will therefore offer the data.
---------------------------------------------------------------------------
2. Competition
a. Modification of Rule 612
The amendments will promote competition both on price on a given
venue and across venues. This will occur because the amendments will
allow liquidity providers to compete at more price points on exchange.
By limiting the affected stocks to those with low spreads, the
amendments ameliorate possible effects of pennying which may accompany
a finer pricing grid.\1751\
---------------------------------------------------------------------------
\1751\ See supra section VII.D.1.b for additional discussion of
the effects of pennying.
---------------------------------------------------------------------------
In addition, the amendments will improve the efficiency of on-
exchange trading, allowing exchanges to better compete with off-
exchange market makers. Empirical evidence suggests that, on average,
relaxing tick constraints leads to volume moving onto exchanges
primarily by improving market quality on the exchanges.\1752\ Research
suggests that this occurs both because of the reduction in transaction
costs resulting from a finer pricing grid, and also because a tick that
is too wide creates long queues for limit order execution and increase
the incentives to send orders off-exchange.\1753\
---------------------------------------------------------------------------
\1752\ See two industry studies attached to MEMX Letter at 43-63
and 64-70. These studies examine the relaxation of tick constraints
following reverse splits and find that the reduction in on-exchange
trading costs results in an increase in on-exchange trading volume.
See also Panel A of table 2 in Bidisha Chakrabarty, et al., Tick
Size Pilot Program and Price Discovery in US Stock Markets, 59 J.
Fin. Mkt. 100658 (2022). This academic study uses the TSP and finds
that an increase in tick constraints results in an increase in off-
exchange trading volume. See also Amy Kwan et al., Trading Rules,
Competition for Order Flow and Market Fragmentation, 115 J. Fin.
Econ. 330 (2015). This academic study examines the change in the
tick from $0.01 to $0.0001 for orders priced in the vicinity of
$1.00, and finds that more volume is executed on exchanges when the
trade price dips below $1.00 and is therefore subject to the smaller
tick.
\1753\ Id.
---------------------------------------------------------------------------
The increase in message traffic expected from the reduction in the
tick sizes for certain stocks will result in a mild increase in costs
to process such traffic for those who receive the relevant data
feeds.\1754\ Because such technological costs are largely fixed, and do
not depend on the size of the broker-dealer, this could disadvantage
smaller broker-dealers in the market to provide broker-dealer services
to investors.
---------------------------------------------------------------------------
\1754\ See supra section VII.D.1.c.
---------------------------------------------------------------------------
Some commenters stated that variable tick sizes could increase
confusion among investors trading on-exchange, thereby driving orders
off-exchange.\1755\ The potential for investor confusion is addressed
generally in section VII.D.1.d. To the point about confusion due to a
smaller tick size driving order flow off of exchanges, this is
unlikely. The ability to trade at finer price points, and the reduced
need to wait in the queue should contribute to on-exchange trading, not
off-exchange trading. Indeed, relaxing of tick constraints has been
associated empirically with volume moving on-, not off-exchange.\1756\
---------------------------------------------------------------------------
\1755\ See Themis Letter at 6 and Cboe Letter II at 7.
\1756\ See supra note 1752.
---------------------------------------------------------------------------
One commenter asked the Commission to consider the competitive
effects of Rule 612 on stocks that have similar quoted spreads but fall
just on either side of the threshold, specifically similar ETPs that
may have quoted spreads that are similar but fall on either side of the
threshold and so receive different tick sizes.\1757\ The commenter
considers two issuers, Issuer A and Issuer B, and explains that a
narrower tick sizes for Issuer A could attract more liquidity to Issuer
A's stock and less liquidity to Issuer B's stock. Once Issuer A's stock
attracts more liquidity, its spreads could potentially narrow further,
perpetuating a cycle in which Issuer B's shares are unable to catchup
to Issuer A. The Commission, however, does not expect significant
competitive effects in this situation. While the evidence suggests that
stocks with quoted spreads less than the threshold will, on average,
benefit from the lower tick, those benefits attenuate as spreads
widen.\1758\ Thus, for stocks or ETPs with
[[Page 81762]]
spreads right at the threshold, the differential effect of the smaller
tick size may be relatively small.
---------------------------------------------------------------------------
\1757\ See SIFMA Letter II at 41.
\1758\ See supra section VII.D.1.
---------------------------------------------------------------------------
b. Lower Access Fee Caps
The amendments to Rule 610(c) reducing the access fee cap will have
varying effects on competition between trading venues as well as
competition between broker-dealers. The Commission does acknowledge
that it would limit the ability of exchanges to differentiate
themselves from other exchanges on the basis of their pricing
schedules; however, the Commission expects that exchanges will continue
to set fees and rebates at or near the access fee cap. A lower access
fee cap mechanically reduces the range over which pricing tiers can
vary, potentially reducing the economic differences between pricing
tiers thereby reducing the benefits from routing order flow for the
purpose of qualifying for one tier over another. This can reduce the
competitive wedge between high and lower volume broker-dealers due to
volume discounts making it easier for lower volume broker-dealers to
compete with larger volume broker-dealers.
Commenters disagreed regarding the effects of the 10 mils cap
relative to the 15 mils/30 mils alternative in terms of the competitive
dynamics between on- and off-exchange venues, with some commenters
arguing that reducing the access fee cap would cause a shift to on-
exchange trading while others arguing it would cause a shift to off-
exchange trading.\1759\ The Commission's discussion in section
VII.D.2.c suggests that it is unlikely that significant activity will
be driven off-exchange, and it is possible that activity may come on-
exchange as a result of the lower access fee cap. Moreover, the lower
access fee cap will improve competition relative to the 15 mils/30 mils
alternative in that it will reduce information asymmetries among
investors, better aligning displayed prices with the actual costs of
transactions.\1760\
---------------------------------------------------------------------------
\1759\ We focus on the competitive effect of the access fee cap
relative to the alternative, as failure to reduce the access fee cap
in the presence of the tick constraint would lead to a loss of price
coherence. For commenters stating that reducing the access fee cap
would increase off-exchange volume, see, e.g., Nasdaq Letter II at 4
and Cboe Letter IV at 2-3. For commenters stating that reducing the
access fee cap would increase on-exchange volume, see IEX Letter VI
at 7.
\1760\ See IEX Letter VI at 4-6.
---------------------------------------------------------------------------
Some commenters expressed that the reduction in access fees will
impede exchange competition by reducing their ability to offer
differentiated pricing.\1761\ One commenter stated that this will
particularly disadvantage new exchanges with limited opportunities for
differentiation.\1762\ Another commenter stated that the inability to
differentiate based on fees and rebates will lead volume to congregate
on the listing exchange to the detriment of non-listing
exchanges.\1763\
---------------------------------------------------------------------------
\1761\ See Fidelity Letter at 14, Virtu Letter II at 10, Cboe
Letter II at 8, and Citadel Letter I at 24.
\1762\ See Virtu Letter II at 19.
\1763\ See Cboe Letter II at 7.
---------------------------------------------------------------------------
The Commission acknowledges that lowering the access fee cap will
mechanically limit the extent to which exchanges can potentially
differentiate themselves based on varying pricing schedules and
diminish their ability to compete on the basis of their pricing
schedules. However it is not clear if the amount of differentiation or
degree of competition will diminish because exchanges do not appear, in
the markets today, to be competing on the basis of offering
substantially different pricing schedules or models.\1764\ The vast
majority (85%) of on-exchange trading volume executes on
exchanges with maker-taker pricing models and with baseline access fees
near the cap and rebates slightly lower than the access fee cap.\1765\
The few exchanges which deviate from this pricing style do not execute
a large proportion of trading volume.\1766\ Additionally, smaller or
newer exchanges (which do not belong to one of the three large exchange
families) have adopted similar pricing schedules and thus do not seem
to be competing for order flow by differentiating their pricing
schedule. For example, MEMX, the exchange with the most market share
not affiliated with one of the three large exchange families, adopted a
similar maker-taker pricing schedule to what is prevalent in the
market. Additionally, IEX, a formerly flat-fee exchange, has recently
switched to a maker-taker pricing model, citing the need to incentivize
liquidity provision.\1767\ This is in line with the discussion in
section VII.C.2, which explains how the structure of markets today
incentivize the adoption of maker-taker pricing where access fees are
set at or near the access fee cap in order to fund large maker rebates
as a means of attracting competitively priced quotes, which in turn
increase the trading volume executed on the exchange. Lowering the
access fee cap does not change this dynamic, and so the Commission
expects that exchanges will continue to set fees and rebates at or near
the lowered access fee cap.
---------------------------------------------------------------------------
\1764\ As explained in section VII.C.2, exchanges can
differentiate themselves by offering different fee schedules--e.g.,
inverted, flat fee, or maker-taker with numerous price strata.
Reducing the access fee cap can reduce the variation in rebates and
fees across venues by narrowing the viable range for fees and
rebates thereby making the different exchange price schedules more
similar. However, many price schedules are already quite similar
despite the 30 mil access fee cap allowing for a greater degree of
differentiation. For instance, the data reported in table 4 does not
show that there is currently a large degree of variation in the
highest fees charged, particularly among maker-taker exchanges which
dominate the market. Additionally inverted exchange fees are all set
close to the access fee cap.
\1765\ See table 5.
\1766\ Id.
\1767\ Self-Regulatory Organizations; Investors Exchange LLC;
Notice of Filing and Immediate Effectiveness of Proposed rule Change
Pursuant to IEX Rule 15.110 to Amend IEX's Fee Schedule, Securities
Exchange Act Release No. 98063 (Aug. 4, 2023), 88 FR 54373 (Aug. 10,
2023).
---------------------------------------------------------------------------
When stating that a lower access fee cap would limit competition by
restricting differentiation, one commenter pointed out that when one
exchange switched from a flat rebate model to a tiered pricing model
that exchange quoted at the NBBO more often.\1768\ This is consistent
with a tiered pricing structure discouraging order routing to competing
venues. In this case switching to a tiered pricing schedule
incentivized that exchange's members to not route orders to competing
exchanges to collect the benefits associated with high volume tiers.
More orders sent to the exchange incentivizes more aggressive quoting
on the exchange leading the exchange to quote at the NBBO more often.
However, the effect on tiering from the amendment's reduction in the
access fee cap would be different from this example because the
reduction in the access fee cap would apply to all exchanges, meaning
that the effect on competition from tiering will also be diminished
across all exchanges. Therefore, the effect that a lower access fee cap
would have on one exchange's ability to more consistently quote at the
NBBO is likely to be weaker than that stated by the commenter.
---------------------------------------------------------------------------
\1768\ See Cboe Letter III at 8.
---------------------------------------------------------------------------
One commenter expressed the concern that the reduced access fee cap
may also result in exchanges increasing the cost to access market data
or sell preferential access to exchange data to some members but not to
others.\1769\ The adopted amendments address this concern as compared
to the proposal by eliminating the proposed requirement for an access
fee cap of 5 mils on some stocks. One commenter stated that while it is
possible that a sufficiently low cap could generate this concern, the
access fee cap of 10 mils strikes the right balance.\1770\ Furthermore,
as explained in section VII.D.2.b, the Commission does not expect
exchange transaction
[[Page 81763]]
revenues on transactions priced greater than $1.00 to substantially
change as a 10 mil access fee cap is expected to be high enough that
exchanges can continue to realize their current net capture rates for
these transactions.
---------------------------------------------------------------------------
\1769\ See RBC Letter at 4.
\1770\ See Verret Letter I at 8.
---------------------------------------------------------------------------
Reducing the access fee cap may also impact competition between
broker-dealers who are exchange members, to the extent that a lower
access fee cap diminishes the marginal benefit of qualifying for a
pricing tier with lower fees or higher rebates over another tier with
less preferential terms. Different transaction pricing tiers,
particularly volume-based pricing tiers providing more favorable fees
and/or rebates to exchange members who execute higher relative order
volume, introduce a competitive wedge between those exchange members
who qualify for the better tiers and those who do not. Another
commenter stated that high access fees disproportionately affect
smaller firms and investors, and lowering the access fee cap would
promote a more competitive and diverse market landscape.\1771\ The
commenter stated that exchanges employ pricing tiers to extract rents
from smaller exchange members which are then split between the exchange
and their high tier members, and lowering the access fee cap would
limit the extent to which this can occur.\1772\ Under the assumption
that a reduction in access fees would be accompanied by a reduction in
transaction rebates, one possible effect of reducing the access fee cap
would be to diminish the relative differences in the fees charged and
rebates offered between different pricing tiers.\1773\ As shown in
table 4 multiple exchanges have fee or rebate tiers which vary within a
range that is greater than 10 mils. Therefore, lowering the access fee
cap to 10 mils will necessitate that pricing tiers would have to be
placed within a narrower price range. The Commission expects that as
the differences between pricing tiers become less economically
meaningful, the competitive wedge introduced by pricing tiers will
diminish, which would make it easier for exchange members lacking scale
to compete with exchange members that qualify for preferential pricing
tiers (i.e., tiers with higher rebates/lower fees).
---------------------------------------------------------------------------
\1771\ See Verret Letter I at 9.
\1772\ Id. at 6, 7.
\1773\ According to table 4 the differences in pricing tiers for
many exchanges exceed 10 mils therefore if the number of pricing
tiers does not decrease, by necessity, the average difference
between the tiers would diminish.
---------------------------------------------------------------------------
Another commenter stated that although the access fee cap would be
lowered, it could still allow sufficient room for existing differences
in preferential pricing to persist.\1774\ The commenter used the
example of a large bank and small broker both paying a 30 mils rebate
with the bank receiving a 32 mils rebate and the broker receiving 24
mils. Under a 10 mils access fee cap, the exchange could offer a 12
mils rebate to the bank and 4 mils to the broker. In that case the
differential between the bank and the broker would remain the same
despite the reduced access fee cap. In the example provided by the
commenter, the exchanges could continue to offer a fully funded rebate
to some exchange members, which is 8 mils greater than offered to other
exchange members, because that 8 mil differential would be allowed
under a 10 mil access fee cap. The Commission acknowledges that
exchanges would be able to continue to offer differentiated pricing;
however, a lower access fee cap would reduce the extent to which
pricing tiers can differ and limit the aggregate fees available to the
exchange to redistribute among its members in the form of rebates. It
is more difficult for an exchange to fund high rebates, particularly
those greater than the fee cap, under a lower access fee cap.
---------------------------------------------------------------------------
\1774\ See Healthy Markets Letter I at 23-24.
---------------------------------------------------------------------------
c. Acceleration of the MDI Rules, Addition of Information About Best
Odd-Lot Orders, Fees and Rebates Determinable at Time of Execution
Accelerating the inclusion of odd-lot information in the NMS data,
along with the implementation of the MDI Rules round lot definition,
might lead to increased competition between exchanges and ATSs and OTC
market makers, including wholesalers. NMS stocks priced greater than
$250.00 are expected to benefit sooner from a tighter NBBO, thereby
increasing the competitiveness of the best displayed protected quotes.
Greater visibility of more competitively priced odd-lot orders inside
the NBBO could increase the competitive position of exchanges and ATSs
and attract greater order flow. This effect will be temporary, only
lasting until the full implementation of the MDI Rules. After the full
implementation of the MDI Rules, the effect on competition is accounted
for in the MDI Adopting Release and is not ascribed to these
amendments.
Making exchange fees and rebates determinable at the time of
execution will enable the customers of broker-dealers to better discuss
transaction fees and rebates with their broker-dealers, and potentially
request data on the exchange fees incurred by an order,\1775\ which
will increase competition between broker-dealers along this dimension,
leading to better order execution and lower costs.\1776\ In particular,
while there is currently no requirement to either pass on the fees and
rebates to customers, or account for fees and rebates when assessing
execution quality, there may be competitive pressure to do so as a
result of the amendments because a competing broker-dealer will be able
to include fees and rebates in its transaction cost analysis, or simply
pass them through to the customer. One commenter stated that the
requirements for exchange pricing under this rule change will be ``even
more anti-competitive'' than the current practice, because this would
mean that ``smaller brokers can't attract new flows based on modelling
of what such flows will do to their rates upon arrival.'' \1777\ The
Commission disagrees with the statement that this rule change will make
the market for offering executing broker services more anti-
competitive. As described in section VII.D.3, this rule change allows
brokers to determine their fees and rebates at execution and thereby
eliminates the need for forecasting future market outcomes in order to
anticipate the fee that will be incurred by an order. To the extent
such forecasting is more difficult for small brokers, the rule change
will make it easier for small brokers to compete.
---------------------------------------------------------------------------
\1775\ For example, some brokers allow customers to direct an
order to a particular exchange. By making the fees and rebates
determinable at execution, the broker may be better able to inform
the customer of the net transaction price of a prospective directed
order.
\1776\ Under the baseline it would be difficult in many cases
for a broker-dealer to allocate specific rebates received or fees
paid to one customer's trade because the fees or rebates in a given
month are based, in many instances, on that broker-dealer's total
trading volume across all customer accounts, see section VII.C.2.b.
However, if the fees and rebates are determinable at the time of
execution the broker-dealer could feasibly track a specific fee or
rebate to a specific trade, making it possible for a customer to
receive such information.
\1777\ See Danny Mulson Letter.
---------------------------------------------------------------------------
Including odd-lot information in the exclusive SIPs and providing
the best odd-lot order information will enhance competition among
broker-dealers. Making the best odd-lot order information accessible
through the exclusive SIPs will facilitate better analysis of a broker-
dealer's execution quality than is available with just NBBO data.\1778\
Thus, it could be easier for
[[Page 81764]]
some customers to monitor the performance of their broker-
dealers.\1779\
---------------------------------------------------------------------------
\1778\ See supra note 1621, for a discussion on the incentives
that institutional traders have to monitor all aspects of
transaction costs.
\1779\ It is possible that some institutional traders have
access to proprietary data feeds that provide the ability to
benchmark trades against odd-lot orders. Or, they could contract
with specialized firms that have access to the data and provide
transaction cost analysis.
---------------------------------------------------------------------------
Accelerating the inclusion of odd-lot data into the exclusive SIPs
will increase competition among data providers of odd-lot information
prior to the full implementation of the MDI Rules, though it will do so
less than envisioned in the MDI release for the period until the MDI
Rules are fully implemented. Specifically, under the implementation
schedule in the MDI Rules, adding odd-lot information to core data was
to occur during the parallel operation period. Adding odd-lot
information to the current exclusive SIPs will enable the exclusive
SIPs to compete directly with the exchanges' proprietary data products
for use in visual display settings. Without this change, the only means
to get odd-lot information is to subscribe to multiple proprietary data
feeds. This will change when odd-lots are a part of SIP data.
Unlike the data provided by the competing consolidators, the
current exclusive SIPs are not fast enough for use in certain
trading.\1780\ Thus, the competition for odd-lot data will be limited
to odd-lot information used in visual display settings. To the extent
that some market participants subscribe to proprietary data for use in
visual display settings, the introduction of odd-lot information to the
exclusive SIPs will provide competition to this segment of the market
and reduce the prices of odd-lot information provided by the
proprietary data feeds. However, the Commission does not believe that
this market is very large. Currently, for most display settings, market
participants use SIP data or one of the top-of-book data products
offered by one of the three highest volume exchange groups; it is
unclear to what extent market participants subscribe to proprietary
data with odd-lot information for use in visual display settings.
---------------------------------------------------------------------------
\1780\ See MDI Rules for a discussion of the SIPs' higher
latency relative the proprietary feeds offered by exchanges. In
particular, footnote 26 on page 18599 summarizes commenters' views
on the disadvantages of using SIP data instead of proprietary feeds.
---------------------------------------------------------------------------
With respect to competition for top-of-book (TOB) data, the
exclusive SIPs face competition from exchanges' TOB data products. As
discussed in the MDI adoption,\1781\ these proprietary products are
typically less expensive and contain less content--being derived from a
single exchange or exchange family--than the exclusive SIPs. If the
exclusive SIPs charge more for data on account of the increased costs
associated with disseminating odd-lot information, then this may
provide a competitive advantage to providers of proprietary TOB
products.
---------------------------------------------------------------------------
\1781\ See MDI Adopting Release, supra note 10 at 18603.
---------------------------------------------------------------------------
Requiring the exclusive SIPs to disseminate the accelerated odd-lot
information until the exclusive SIPs are retired will guarantee that
the odd-lot information will be disseminated.\1782\ However, this
requirement may also affect competition among competing consolidators
once the MDI Rule is fully implemented. On the one hand, these new
requirements on the SIPs could reduce competition among competing
consolidators and therefore reduce the expected benefits of the MDI
Rules. This reduction in competition could occur because the amendments
may increase the competitive advantage of exclusive SIPs relative to
non-SIP competing consolidators because the SIPs will have established
a market for odd-lot information before having to face competition.
That is, the SIPs will have time to acquire customers for odd-lot
information before other competing consolidators can enter. These
customers may then face costs should they switch to a non-SIP competing
consolidator; these switching costs may dissuade entry by non-SIP
competing consolidators and thereby lower competition.\1783\
---------------------------------------------------------------------------
\1782\ See supra section VII.D.4.c for additional discussion.
While the amendments require the exclusive SIPs to distribute odd-
lot data, the MDI Rules do not require the competing consolidators
to disseminate odd-lot data. However, the MDI Adopting Release
anticipated that at least one competing consolidator will do so
because there would be demand for the data.
\1783\ See MDI Adopting Release, supra note 10, for further
discussion of how competing consolidators have higher barriers to
entry than exclusive SIPs, such as in the form of compliance costs
associated with Regulation SCI.
---------------------------------------------------------------------------
On the other hand, the Commission is uncertain whether the SIPs
will become competing consolidators.\1784\ The amendments' requirement
for SIPs to disseminate odd-lot information reduces the incremental
costs that the SIPs would need to bear in order to become competing
consolidators. Therefore, these amendments make it more likely that the
SIPs will register as competing consolidators, which would improve
competition relative to a scenario in which they do not compete.
---------------------------------------------------------------------------
\1784\ See supra note 1782.
---------------------------------------------------------------------------
Further, non-SIP competing consolidators will still have an
opportunity to compete for significant market share. As discussed
above, SIPs face latency disadvantages relative exchanges.\1785\ If
competing consolidators can offer a lower latency product, then they
can capture a part of the market that the amendments will not affect--
those customers who will use odd-lot information in ways other than
visual display.\1786\ Likewise, competing consolidators can offer
depth-of-book data under the MDI Rules, which the SIPs are not required
to disseminate under these amendments. If these markets are
significantly bigger than the odd-lot visual display market, the
competitive advantage of the exclusive SIPs will be less likely to
dissuade entry, and non-SIP competing consolidators could have
sufficient incentive to enter the market.\1787\
---------------------------------------------------------------------------
\1785\ See supra note 1780.
\1786\ See discussion around note 1780, supra explaining that
the SIPs' latency disadvantage makes their data useful for visual
display.
\1787\ In the MDI Adopting Release, the Commission anticipated
that both exchanges operating exclusive SIPs would have strong
incentives to enter the competing consolidator market. See MDI
Adopting Release, supra note 10, at 18761.
---------------------------------------------------------------------------
3. Capital Formation
The Commission expects that the amendments will promote capital
formation. First, the combined effect of the amendments will be to
increase liquidity generally, which will increase incentives to trade
and therefore price efficiency. Price efficiency in turn promotes
capital formation. The Commission also expects that the alleviation of
tick constraints and the lower access fee cap will work together and
separately to lead to displayed prices that are more reflective of
supply and demand for the underlying securities, also promoting capital
formation.
One commenter stated that a narrower tick could increase volatility
and decrease liquidity which could discourage companies from going
public.\1788\ As discussed in section VII.D.1, stocks receiving the
$0.005 tick on average will not experience harmful liquidity effects.
On the contrary, as discussed in section VII.D.1, the expectation is
that on average liquidity will improve for stocks with narrow quoted
spreads that receive the tick size reduction. Additionally, as
discussed in section VII.D.1, the narrower tick will not result in
increased volatility.\1789\ Consequently, even if there was a link
between liquidity and volatility, and the decision to go public, those
channels
[[Page 81765]]
aren't expected to be affected in the manner mentioned by the
commenter. Further, the link between tick sizes and IPOs is not clearly
defined in existing research.\1790\
---------------------------------------------------------------------------
\1788\ See RBC Letter at 3.
\1789\ See supra notes 1209 and 1210 and surrounding text for a
discussion of tick sizes and volatility.
\1790\ Research on this topic is exceptionally difficult. As
stated in the report Assessment of the Plan to Implement a Tick Size
Program, ``There are myriad factors influencing companies' decisions
about whether to go public or remain private--and, if an IPO is
desired, in which country to list shares. These include the
availability of capital outside the public equity market, the
regulatory burdens placed on public companies, market conditions,
broader macroeconomic trends and differences in economic conditions
between countries globally. Additionally, broader historical context
may reveal certain periods of strong IPO issuance, particularly
during times of high speculative activity in markets, as anomalous
and unsustainable.'' See Securities and Exchange Commission,
Assessment of the Plan to Implement a Tick Size Pilot Program
(Jul.3, 2018), available at https://www.sec.gov/files/TICK%20PILOT%20ASSESSMENT%20FINAL%20Aug%202.pdf (last accessed Feb.
6, 2024).
---------------------------------------------------------------------------
Commenters expressed the concern that wider spreads and reduced
depth would negatively impact capital formation for growth
companies.\1791\ Commenters specifically mentioned the importance of
rebates for small and medium-sized growth companies, without which
``market makers may no longer find it profitable to make tight
markets.'' \1792\ Two considerations enter in determining the effect of
capital formation. First, for illiquid stocks, spreads are the primary
determinant of revenue for liquidity providers. The rebate makes less
of a difference on a percentage basis then for stocks that are more
liquid. Second, the Commission does not expect the cost of transacting
in illiquid securities to rise, net of fees and rebates.\1793\ While
the Commission acknowledges the crucial role of the ability of
investors to transact for capital formation, it is not quoted spreads
that matter to investors but rather the net spread available on
exchange. In sum, liquidity is expected to improve for stocks with
narrow quoted spreads that receive the tick size reduction--as
discussed in section VII.D.1--and liquidity is not expected to be
harmed for stocks that do not receive the tick size reduction--as
discussed in section VII.D.2. Therefore, the amendments are expected to
improve liquidity and thus will not impede capital formation through
this channel.
---------------------------------------------------------------------------
\1791\ See Nasdaq Letter I at 24.
\1792\ See Nasdaq Letter I at 25. See also Virtu Letter II at
10.
\1793\ One commenter, stating that the reduced access fee cap
would reduce ``incentives for liquidity in thinly traded
securities,'' cited a study showing that an improvement in liquidity
from stock splits resulted in significant reductions in the cost of
capital for firms that did a stock split (Virtu Letter II at 8,
citing Ji-Chai Lin, Ajai K. Singh &Wen Yu, Stock Splits, Trading
Continuity, and the Cost of Equity Capital, 93 J. Fin. Econ. 474,
475 (Jan. 1, 2009)). As stated above, because the reduction of the
access fee will not result in an increase in the cost of liquidity,
see supra section VII.D.2.c, discussing this point, there is no
reason to expect the cost of capital to increase as a result of
lowering the access fee cap.
---------------------------------------------------------------------------
F. Reasonable Alternatives
This section considers alternatives to the amendments. In the
Proposing Release, we considered the benefits and costs of multiple
categories of alternative, and variations within those
categories.\1794\ For brevity we do not repeat that discussion here.
Instead, this section focuses on additional alternatives suggested by
commenters, to the extent they are not incorporated into the adopted
amendments. We organize subsections around key elements of the Rule:
tick size, minimum trading increment, access fee, and MDI and
BOLO.\1795\
---------------------------------------------------------------------------
\1794\ See Proposing Release, supra note 11, at 80339.
\1795\ Some commenters discussed ``no action'' as an alternative
to the proposed rules. See, e.g., Virtu Letter II at 22-23. For
purposes of the economic analysis, the baseline describes the world
as it would exist without the rules.
---------------------------------------------------------------------------
1. Tick Size Alternatives
a. Alternative Criteria for Selecting Stocks Receiving a Smaller Tick
Size
Commenters suggested alternative methodologies for identifying
which stocks should receive a smaller tick size.\1796\ These
alternative methodologies are discussed in greater detail in section
VII.D.1.b.iii and generally center on adding additional criteria, in
addition to the TWAQS, to determine which stocks should qualify for a
lower tick size.\1797\ In that section, analysis failed to find
evidence that the additional criteria would avert harm to market
quality. One reason for this is likely that much of the information
contained in these additional thresholds suggested by commenters is
already contained in the TWAQS.\1798\
---------------------------------------------------------------------------
\1796\ See supra section VII.D.1.b.iii for additional discussion
of these methodologies.
\1797\ See supra note 1311 for discussion of specific commenter
suggestions.
\1798\ See supra note 1318 and surrounding discussion.
---------------------------------------------------------------------------
Moreover, implementing these alternatives would increase the
complexity of the amendments from the perspective of the listing
exchanges, who would be required to track and implement multiple
thresholds to identify tick-constrained securities. Increased
complexity would increase the compliance costs of the amendments for
these entities. Complexity would also increase for broker-dealers and
investors, who would be required to take these changes into account.
These alternatives would likely not affect the compliance costs of the
rules for other market participants relative to the adopted amendments.
This is because these alternatives would not change how these entities
learn which stocks are subject to the $0.005 tick and which are subject
to the $0.01 tick size in terms of assessing lists from the listing
exchanges' websites, and the need to update systems to implement the
different tick sizes.
The biggest effect of these alternatives relative to the adopted
amendments is that they would reduce the number of securities receiving
a reduced tick size. For example, one proposed alternative would limit
the number of securities receiving a smaller tick size to an estimated
58 stocks.\1799\ Commenters stated that limiting the sample via
additional thresholds and criteria would ensure that only the stocks
that are absolutely the most likely to benefit from a smaller tick size
would receive the smaller tick size.\1800\ However, the drawback to
this more limited approach is that the analysis presented in sections
VII.D.1.b.ii and VII.D.1.b.iii suggests that many stocks that would not
qualify for the lower tick size under these alternative thresholds
would likely still benefit from reducing the tick size. This conclusion
is supported by the findings in table 9 which demonstrate that across
many dimensions TSP stocks with narrow spreads that are nonetheless in
the bottom quartile based on depth, price, or trading volume, i.e.,
those that commenters suggest could perhaps be excluded from receiving
the lower tick size, still experienced market quality improvements
across many dimensions with a smaller tick. This analysis also fails to
find statistically significant evidence that such stocks would be
harmed. Consequently, adding additional criteria would add complexity
to the implementation of the Rule, increasing the compliance costs of
the rule, and would have lower benefits than the adopted amendments
because it would leave some stocks with a wider tick size than would be
optimal.
---------------------------------------------------------------------------
\1799\ See Cboe Letter II at 5.
\1800\ See supra section VII.D.1.b.iii for additional
discussion.
---------------------------------------------------------------------------
b. Alternative Threshold for Lower Tick Size
Some commenters suggested that the Commission adopt a threshold for
the lower tick size that is different from the adopted amendments. The
most common alternative suggested was a TWAQS of $0.011
threshold.\1801\
---------------------------------------------------------------------------
\1801\ See, e.g., UBS Letter at 10 and JPMorgan Letter at 4.
With a $0.011 threshold, following the methodology employed in table
7, an estimated 1,216 stocks would receive the lower tick size, with
a $0.02 threshold an estimated 2,339 stocks would receive the lower
tick size.
---------------------------------------------------------------------------
[[Page 81766]]
The Commission estimates that the costs to implement this
alternative would be similar to the adopted amendments because all
affected entities would be required to perform the same work as in the
adopted amendments. From an implementation perspective, the key
difference between this alternative and the adopted amendments would be
the considerably reduced number of stocks that would qualify for the
alternative's lower tick size.
This alternative would more specifically target trading volume that
is nearly always trading at the minimum trading increment. This
alternative would leave stocks with quoted spreads between $0.011 and
$0.015 with the $0.01 tick size, whereas the adopted amendments assign
a tick size of $0.005 to such stocks. Using the same methodology as is
used in table 7 there would be an estimated 1,216 stocks receiving a
$0.005 tick size under this alternative, a reduction of approximately
572 stocks compared to the adopted amendments. These omitted stocks
have between 1.1 and 1.5 ticks intra-spread on average. Research and
Commission analysis as well as commenters' analyses suggest that 2 to 4
ticks intra-spread is likely an optimal range for stocks.\1802\ Thus,
assigning stocks with a quoted spread between $0.011 and $0.015 to a
tick size of $0.01--resulting in 1.1-1.5 ticks intra-spread--is likely
to produce worse market quality outcomes than assigning these stocks a
$0.005 tick--which would result in 2.2-3 ticks intra-spread. Thus,
under this alternative these stocks, on average, would be expected to
have lower overall market quality relative to the adopted amendments.
Specifically, analysis in table 8 of stocks in bin 2, which had 1-2
ticks intra-spread during the TSP, experienced significant improvements
in market quality when the TSP tick size was relaxed--providing
evidence that such stocks would benefit from a lower tick size.
Consequently, this alternative, by failing to reduce the tick size for
stocks that evidence suggests would benefit from a tick size decrease,
would have lower benefits compared to the adopted amendments, while
having similar costs.
---------------------------------------------------------------------------
\1802\ See supra section VII.D.1.b.ii.
---------------------------------------------------------------------------
c. Alternative Measurement Horizons for TWAQS and Effective Periods for
Tick Sizes
Some commenters suggested alternative measurement horizons to
determine the TWAQS as well as alternative periods that the tick size
would be effective.\1803\ Much of the analysis of alternative
measurement horizons for the time weighted quoted spread as well as
alternative operative periods for tick sizes is contained in VII.D.1.d.
In sum, the analysis in that section presents tradeoffs. On the one
hand, a shorter evaluation period ties the tick size to the most recent
market experience for a given stock potentially resulting in the most
relevant tick size to be assigned. On the other hand, if that time
period is associated with transient spikes in quoted spreads, such as
during the first quarter of 2020 coincident with the onset of the
Covid-19 pandemic, then the time period used to assign tick sizes would
not be representative of current market conditions and a stock may be
assigned a sub-optimal tick size.
---------------------------------------------------------------------------
\1803\ See, e.g., FIA PTG Letter II at 2 and JPMorgan Letter at
4.
---------------------------------------------------------------------------
There is also a tradeoff associated with the length of time that
tick sizes are effective. More frequent updating means the tick size
can adjust more rapidly to changes in the trading environment for a
given stock and thus could increase the amount of trading volume
associated with optimal tick sizes. The downside is that more frequent
updates would also increase the cost and complexity of the amendments
as market participants would have to adjust to tick sizes that change
more frequently.
The Commission analyzed many iterations of evaluation period and
tick size operative period and found evidence consistent with these
tradeoffs.\1804\ Consequently, depending on the combination of period
used to determine the TWAQS and the effective period for the tick size,
the total fraction of trading volume that trades in the preferred range
may increase or decrease relative to the adopted amendments as
suggested by the analysis in table 10. Additionally, the costs and
complexity of the alternatives would similarly be affected by
alternative horizons chosen with more frequent updating associated with
higher costs and complexity and less frequent updating associated with
lower costs and complexity relative to the adopted amendments.
---------------------------------------------------------------------------
\1804\ See supra section VII.D.1.d.
---------------------------------------------------------------------------
For example, as shown in panel A of table 10, an alternative which
would have a one-month evaluation period and a one-month effective
period for a tick size would reduce the amount of estimated trading
volume that trades with a wide tick and likely would have benefited
from a smaller tick to 8.5%, compared to an estimated 13.7% in the
adopted amendments. Consequently, more trading volume would be assigned
a tick size that is expected to improve market quality for the stock.
However, relative to the adopted amendments this alternative would have
12 tick size changes per year instead of the adopted 2 changes, thus it
would increase the complexity and compliance costs associated with the
rule. Overall, relative to the adopted amendments, this alternative
results in a significant increase in complexity but only achieves a
relatively modest increase in effectiveness in terms of trading volume
with the appropriate tick size.
On the other side of the spectrum, a rule that uses a 12-month
operative period could produce the opposite effect. It would reduce
complexity somewhat by reducing the number of revisions, but with a
significant decrease in effectiveness of the tiered tick size regime.
For instance, again using data from table 10, this alternative would
increase the amount of trading volume that retains the larger tick size
but would likely benefit from a smaller tick size to 20.8%, up from an
estimated 13.7% associated with the adopted amendments.
2. Access Fee Alternatives
a. 15 mils/30 mils Access Fee
Some commenters recommended adopting a two-tier approach to
existing Rule 610(c)'s uniform maximum access fee cap. Specifically,
these commenters recommended an access fee of 15 mils for stocks with a
$0.005 cent tick and maintaining the 30 mils maximum access fee for
stocks that continue to have a $0.01 cent tick (15 mils/30 mils
alternative).\1805\ This 15 mils/30 mils
[[Page 81767]]
alternative would be applied to stocks priced $1.00 or more.\1806\
---------------------------------------------------------------------------
\1805\ See, e.g., NYSE Letter I at 7 and Nasdaq Letter IV at 2.
Nasdaq's alternative assumes that the Commission will adopt only one
additional reduced tick size bucket of $0.005 and a uniform access
fee of $0.10. Nasdaq Letter IV. See also NYSE, Schwab, and Citadel
Letter at 2 (``we recommend a reduction that is proportionate to the
proposed reduction in the minimum quoting increment for tick-
constrained symbols. This would reduce the current $.0030/share cap
to $.0015/share for the symbols with a half-penny minimum quoting
increment''); Nasdaq Letter I at 2; MMI Letter at 7 (``access fees
should be scaled based on 30% of the minimum pricing increment'');
Robinhood Letter at 5, 56-59 (``the Commission should tie access fee
caps to be consistently proportional to the applicable tick size at
the current proportion of 30%''); and MEMX Letter at 23-24 (``Lower
the access fee cap in tick constrained NMS Stocks to $0.0015 to
maintain the proportionality of access fees and tick sizes, and
include auction fees within the scope of the rule to prevent
competitive distortions that would otherwise result if listing
exchanges were permitted to use auction fees to avoid a lower fee
cap'').
\1806\ Commenters recommending this alternative did not address
the treatment of sub $1.00 stocks. For purposes of this discussion,
we assume a proportionately reduced 0.15% access fee cap for those
stocks. This access fee cap percentage relative to the adopted
amendments would mitigate the expected reduction in exchanges'
revenue resulting the lower access fee cap. Cf. Cboe Letter I (not
recommending any reduction in access fees but, if access fees are
reduced, recommending that the access fee [for stocks priced equal
to or greater than $1.00] should not be reduced below $.0015 for
tick constrained securities with a $0.005 [tick] increment. For
securities priced less than $1.00, the access fee cap must remain
unchanged to support competition, differentiation, and liquidity
provision.'').
---------------------------------------------------------------------------
According to one commenter, this approach would be beneficial
because the higher rebate cap for $0.01 tick stocks would maintain
incentives to provide on-exchange lit liquidity.\1807\ The Commission
acknowledges that the lower access fee cap on stocks with the higher
tick is likely to widen spreads. As discussed in sections VII.D.2.c and
VII.E.1, these wider spreads will not lead to a diminution of lit
liquidity on exchanges. Rather, liquidity providers will adjust their
quotes to reflect the change in fees and rebates, resulting in higher
quoted spreads without an increase in transaction costs nor a decrease
in lit liquidity.
---------------------------------------------------------------------------
\1807\ Nasdaq Letter IV at 11 and passim. See also supra notes
529 and 535 and accompanying text.
---------------------------------------------------------------------------
Relative to the adopted amendments, this alternative would raise
costs by causing stocks to oscillate between the two tick sizes,
resulting in the tick size being consistently mis-assigned for the
oscillating stocks. Specifically, the 15 mils/30 mils alternative would
result in a feedback loop scenario on tick size assignment from tying
fee caps to tick size. As discussed in sections VII.B.3 and VII.C.2.b,
access fees are tied to rebates, which in turn influence quoted
spreads. Consider, for example, a stock that trades in an Evaluation
Period, on average, with a 1.4 cent TWAQS and 30 mils access fee cap
(and rebate). Under the amendments to Rule 612, this stock would
receive a 0.5 cent tick. If this stock were to also to be subject to
the 15 mils/30 mils alternative, it would thus be subject to a 15 mils
fee cap once it receives the smaller tick size, instead of a 30 mils
one, and its average TWAQS would increase to 1.7 cents (increase by
twice the 0.15 cents reduction in rebates as result of the lower access
fee cap).\1808\ But with a TWAQS of 1.7 cents the stock in a subsequent
Evaluation Period would receive a tick of 1 cent and the access fee cap
would again be 30 mils.\1809\ At that point, the aforementioned process
would begin again: the stock would yo-yo between tick sizes of 0.5
cents and 1 cent and access fee caps of 15 mils and 30 mils, generating
investor confusion and additional costs.\1810\ In contrast, consider
what would happen to this same stock under a uniform 10 mils access fee
cap. The average quoted spread would widen from 1.4 cents to 1.8 cents,
at which point the stock would not be subject to a tick size reduction
and there would be no oscillations. The same problem would occur even
with a lower threshold for a tick size reduction (e.g., if the TWAQS
threshold for the 0.5 cent tick were to be set at 1.1 cents instead of
1.5 cents, then the 15 mils/30 mils alternative would result in spreads
oscillating between 1.05 and 1.35 cents for some stocks, causing them
to yo-yo between tick sizes). Moreover, oscillation would also occur if
one were to add other metrics to the TWAQS threshold for smaller tick
sizes because there will still be stocks near the TWAQS threshold. The
simplest, and perhaps the only, way to avoid this feedback loop is to
use a uniform access fee cap.
---------------------------------------------------------------------------
\1808\ It is not necessary to assume that the exchange rebates
the entire access fee. Rather, it is sufficient that the difference
between the rebate under the access fee cap of 0.30 mils and the
rebate under the access fee cap of 0.15 mils is 0.15. The example is
not changed if the rebate goes from 0.28 mils to 0.13 mils.
\1809\ As discussed above in section VII.D.1, the stock's having
a 0.5 cent tick would likely lead to a narrower quoted spread than
1.7 cents. However, if the smaller tick causes the quoted spread to
fall by anything less than 0.2 cents, the tick and access fee cap
would revert back to 1 cent and 30 mils after the next Evaluation
Period.
\1810\ One commenter stated that, ``the Commission should
consider the possibility of tick size oscillation for some stocks
that fall close to the threshold values for TWAS.'' The commenter
stated that such oscillation will impose excessive costs. See Mitre
Corp. Letter at 5. See also section VII.D.1.d for a quantitative
analysis on the tradeoff between appropriate tick assignment and the
number of tick changes when evaluating the length of the evaluation
and operative periods. The oscillation discussed in this alternative
does not present such a tradeoff--this oscillation results in both
more tick changes and more tick misassignment.
---------------------------------------------------------------------------
The 15 mils/30 mils alternative would also increase complexity
because the higher level of fees and rebates create a larger wedge
between the quoted half-spread and the true cost of demanding
liquidity.\1811\ Further, this alternative would reduce some benefits
related to the minimum pricing increment, most substantially for stocks
with the penny tick. As discussed in section VII.C.1.b, these stocks
may have, at times, an economic spread that is less than one penny. At
those times, the difference in outcome between a 30 mils and 10 mils
access fee would be a relatively large reduction in distortions.\1812\
For stocks assigned to the half-penny tick that remain tick-
constrained, the reductions would be more minor in an absolute sense
(15 mils versus 10 mils). The analysis for potential conflicts of
interest is similar.\1813\ While the Commission acknowledges that there
is some access fee that would be so low as to create strain on exchange
business models, 10 mils appears well above this point.\1814\ Like the
adopted amendments, this alternative would not affect an exchange's
ability to earn its baseline net capture on trading volume priced
greater than $1.00.\1815\ This alternative may also result in
complications for orders priced below $1. Specifically, for orders
priced below $1 this alternative would lead to one of two outcomes that
could have negative effects for stocks trading right at the $1.00
threshold. For these stocks, the minimum pricing increment is $0.0001
regardless of the access fee applied. If the fee cap for sub-$1 orders
were to be kept proportional at 0.30% and 0.15% for trades priced above
$1.00, then if those stocks prices drop below $1.00 this would result
in a situation where there was a group of stocks with the same tick
size (i.e., $0.0001) but two different fee caps. This could place
stocks with the higher fee cap at a competitive disadvantage relative
to the stocks with the lower access fee cap. This outcome is also more
complicated than both the baseline and adopted rule which both have at
most one fee cap per tick size. Alternatively, if the fee cap for
orders priced below $1 were to be set uniformly at 0.30%, then this
alternative would create a discontinuity in the cost of accessing
liquidity at the $1.00 price threshold for stocks assigned the 15 mils
fee cap; likewise, if the fee cap for orders priced below $1 were to be
set uniformly at 0.15%, then it would create a discontinuity at the
$1.00 price point for stocks with the 30 mil fee cap. This
discontinuity could create distortions in liquidity provision as
discussed in section VII.D.2.c.\1816\ Consequently, this alternative
results in either a situation in which there are two
[[Page 81768]]
fee caps for the $0.0001 tick, or there exists a discontinuity in the
cost of accessing liquidity at the $1 price point causing distortions
in liquidity provision. Thus, this alternative appears to create costs
without corresponding benefits.
---------------------------------------------------------------------------
\1811\ See supra section VII.D.2.d.
\1812\ See supra section VII.D.2.d. For stocks receiving the
penny tick, the reduction in the access fee cap from 30 mils to 10
mils will reduce transaction costs for stocks that experience
periods in which they are tick-constrained, and further reduce the
probability that a stock becomes tick-constrained.
\1813\ See supra section VII.D.2.d
\1814\ See supra section VII.D.2.b
\1815\ See supra section VII.D.2.b
\1816\ Specifically, see supra note 669, on the issue of
discontinuities in the cost of accessing liquidity near the $1
threshold.
---------------------------------------------------------------------------
One commenter suggested the Commission plan to further study the
question of access fee caps in combination with the change in the
tick.\1817\ For the reasons discussed above and elsewhere in this
release, the Commission adopts the 10 mil access fee cap. Commission
staff, however, will review and study the effects of the amendments as
described in section VII.D.
---------------------------------------------------------------------------
\1817\ Nasdaq Letter IV at 11. This commenter recommended this
alternative as phase 1 in a three-phase data gathering process
where, in phase 2, the Commission would collect a year of data from
phase 1's changes and then consider a further access fee cap
reduction for stocks with a $0.005 tick. In phase 3, the Commission
would collect an additional year of data to consider a lower fee cap
for stocks with a $0.01 tick. Id. at 2 and 11.
---------------------------------------------------------------------------
b. Higher or Lower Uniform Access Fee Cap
The Commission could have adopted different uniform access fee
caps. An access fee cap must stay below 50% of the minimum pricing
increment in order to preserve price coherence.\1818\ Alternatively, if
the access fee cap is set below an exchanges' net capture rate, then it
can adversely affect existing exchange pricing practices.\1819\ Some
commenters suggested retaining the current uniform 30 mils cap for
stocks with prices above $1.00.\1820\ A uniform 30 mils level would be
above 50% of the minimum pricing increment under the adopted
amendments, and thus would not preserve price coherence. Another
commenter suggested a uniform access fee cap below 10 mils.\1821\
---------------------------------------------------------------------------
\1818\ See supra section VII.D.2.a.
\1819\ If the access fee cap was set below an exchange's net
capture rate, then its profitability would decrease because it would
no longer be able to charge fees high enough to cover any non-
negative rebate. To retain the same net capture rate, the exchange
would have to charge a negative rebate (i.e., a fee), which would
represent a major change in pricing model.
\1820\ See Citigroup Letter at 6, WFE Letter at 4.
\1821\ See IEX Letter IV at nn.14 and 21.
---------------------------------------------------------------------------
As discussed in section VII.D.2.b, exchanges have sufficient
flexibility under the adopted amendments to maintain their current net-
capture and agency business models on stocks with prices above $1. A
uniform access fee higher than 10 mils would afford exchanges more
flexibility relative to the adopted amendments. For stocks with prices
less than $1.00, an access fee percentage (of the share price) higher
than the adopted 0.1% would imply, relative to the adopted amendments,
a lower revenue loss on sub $1.00 trading.\1822\ To illustrate with an
example, if the access fee percentage were 0.15% instead of the adopted
0.10%, then exchanges' expected revenue loss on sub $1.00 trading would
be approximatively $41 million across those exchanges charging the full
0.30% under the baseline, instead of approximatively $55 million under
the adopted amendments.\1823\ The main cost of an access fee cap above
10 mils would be to increase transaction costs for stocks with economic
spreads smaller than the minimum pricing increment.\1824\ Also, an
access fee cap higher than 10 mils would allow for a greater wedge to
exist between displayed prices and the net prices that are actually
realized, potentially undermining price transparency.
---------------------------------------------------------------------------
\1822\ See supra section VII.D.2.b for a discussion of lost
revenue under the adopted amendments.
\1823\ See table 14 in supra section VII.D.2.b. In that table,
if the access fee percentage were 0.15% instead of the adopted 0.10%
then the lost revenue on sub $1.00 trading would be approximatively
$41 million across exchanges charging the full 0.30% under the
baseline.
\1824\ For stocks with wider economic spreads, the higher access
fee would most likely reduce the spread in equilibrium, implying
little or no effect on transaction costs. See supra sections VII.B.3
and VII.D.2.c
---------------------------------------------------------------------------
In contrast, if the access fee caps were set below the adopted
levels, then the effects described in the prior paragraph would all
flip. Namely, relative to the adopted amendments, transaction costs for
stocks with economic spreads smaller than the minimum pricing increment
would be lower, and there could exist a smaller wedge between displayed
prices and the net prices that are actually realized, potentially
improving price transparency. However, relative to the adopted
amendments, exchanges could no longer have sufficient flexibility to
earn their net capture on stocks with prices above $1.00, and exchanges
would incur a greater loss in revenue on sub $1.00 trading.
An access fee cap higher than the adopted 10 mils, and the
associated higher rebates, also exacerbates the potential conflict of
interest for broker-dealers who route customers' orders to the
exchanges. As discussed in section VII.D.3, fees and rebates introduce
the potential for a conflict of interest if those fees and rebates are
not fully passed through to the routing broker-dealers' customers. A
higher access fee cap would increase the potential proceeds a broker-
dealer would receive if it acted on the conflict of interest. A lower
access fee cap would decrease the differences between the fees and
rebates offered by different exchanges, which would decrease the
potential proceeds a broker-dealer would receive if it acted on the
conflict of interest.
Additionally, relative to the adopted amendments to Rule 610(c), a
lower access fee cap could hinder an exchange's ability to
differentiate itself from other exchanges on the basis of its pricing
schedule, whereas a higher access fee cap could enable more
differentiation. As discussed in section VII.E.2.b, the Commission
expects that exchanges will continue to set fees and rebates at or near
the access fee cap; therefore, a higher or lower access fee cap would
likely have minimal effect on pricing differentiation across exchanges.
For retail investors, an access fee cap different from the adopted
levels would likely have little effect on retail market quality for
reasons discussed earlier.\1825\ With regard to exchange trading versus
off-exchange trading, as discussed above,\1826\ an access fee cap lower
than the adopted level could bring more trading volume onto exchanges
by further relieving tick constraints that drive volume off-exchange. A
higher access fee cap would reverse these effects.
---------------------------------------------------------------------------
\1825\ See supra note 1480 and surrounding text.
\1826\ See supra section VII.E.2.b.
---------------------------------------------------------------------------
VIII. Paperwork Reduction Act
Certain provisions of the rules and rule amendments contain
``collection of information requirements'' within the meaning of the
Paperwork Reduction Act of 1995 (``PRA'').\1827\ The Commission
requested comment on the collection of information requirements in the
Regulation NMS Proposal and submitted relevant information to the
Office of Management and Budget (``OMB'') for review in accordance with
44 U.S.C. 3507(d) and 5 CFR 1320.11. The title of the new collection of
information is ``Odd-Lot Information Acceleration.'' An agency may not
conduct or sponsor, and a person is not required to respond to, a
collection of information unless the agency displays a currently valid
control number. The Commission has received an OMB control number
(3235-0802) for this collection of information.
---------------------------------------------------------------------------
\1827\ 44 U.S.C. 3501 et seq.
---------------------------------------------------------------------------
One commenter stated that the hourly rates for certain positions
were inconsistent across the four proposals related to separate aspects
of equity market structure and Regulation NMS.\1828\ No other comments
were received discussing the PRA. The hourly rates used to monetize
burden
[[Page 81769]]
hours differ across releases in order to account for changes in
inflation rates. Consistent with this approach, the hourly rate figures
discussed below have been updated from those cited in the Proposing
Release to reflect recent inflation rates. In addition, certain
estimates outlined in the MDI Adopting Release have been modified, as
discussed in section VII.G below, to conform to the adopted amendments.
---------------------------------------------------------------------------
\1828\ See Data Boiler Letter I at 16 (identifying what it
termed ``inconsistent rates'' for the Attorney and Compliance
Manager positions).
---------------------------------------------------------------------------
A. Summary of Collection of Information
The rule amendments include a collection of information within the
meaning of the PRA. Specifically, the amendments to Rule 603(b) require
the exclusive SIPs to collect, consolidate, and disseminate odd-lot
information, including the best odd-lot orders to buy and sell. The
exclusive SIPs are also required to disseminate indicators of the
applicable round lot size and minimum pricing increment for each NMS
stock, both of which will be provided to the exclusive SIPs by the
primary listing exchange.
B. Proposed Use of Information
The information collected under the amendments to Rule 603(b) will
be consolidated and disseminated by the exclusive SIPs to market
participants who will use this odd-lot information for trading.
Widespread availability of odd-lot information promotes fair and
efficient markets and facilitates the ability of brokers and dealers to
trade more effectively and to provide best execution to their
customers. The round lot and minimum pricing increment indicators that
will be disseminated by the exclusive SIPs will provide market
participants with information about the parameters for trading in a
particular NMS stock.
C. Respondents
The collection of information under amended Rule 603(b) will apply
to the two exclusive SIPs.
D. Total Annual Reporting and Recordkeeping Burden
1. Initial Burden Hours and Costs
The two exclusive SIPs will have to modify their systems to
collect, consolidate, and disseminate the odd-lot information,
including the best odd-lot orders to buy and sell, that they do not
currently collect, consolidate, and disseminate \1829\ and to
disseminate the round-lot and minimum pricing increment indicators
provided by the primary listing exchange. These modifications will
involve the addition of new hardware, network infrastructure, and
bandwidth, as well as programming and development costs, to take in
additional inbound odd-lot quotation messages from SROs, to calculate
odd-lot information, and to consolidate and disseminate odd-lot
information and the round lot and minimum pricing increment indicators
to subscribers.
---------------------------------------------------------------------------
\1829\ The exclusive SIPs currently disseminate odd-lot
transaction data.
---------------------------------------------------------------------------
The Commission estimates that each exclusive SIP will incur 440
initial burden hours to modify its systems to collect, calculate,
consolidate and disseminate odd-lot information and to disseminate the
round-lot and minimum pricing increment indicators \1830\ and initial
external costs of $412,500 to purchase the necessary technology to
effect such modifications.\1831\ Thus, the Commission estimates that
the total initial burden hours for two exclusive SIPs will be 880
burden hours \1832\ and that total initial external costs would be
$825,000.\1833\
---------------------------------------------------------------------------
\1830\ The Commission estimates the monetized initial burden for
this requirement to be $167,670, broken down as follows: [(Sr.
Programmer at $399/hour for 210 hours) + (Sr. Systems Analyst at
$343/hour for 180 hours) + (Compliance Manager at $373/hour for 20
hours) + (Director of Compliance at $588/hour for 10 hours) +
(Compliance Attorney at $440/hour for 20 hours)] = 440 initial
burden hours to modify its systems to comply with the requirement to
collect, calculate, and disseminate odd-lot information. The
Commission based these estimates on 10% of the initial burden hour
estimates for each exclusive SIP to become a competing consolidator
provided in the MDI Rules to account for the fact that these
amendments do not require the exclusive SIPs to calculate and
disseminate full consolidated market data (e.g., depth of book data
or auction information) as defined in the MDI Rules. See MDI
Adopting Release, supra note 10, at 18712-13. The Commission derived
the hourly rate figures from SIFMA's Management & Professional
Earnings in the Securities Industry 2013, modified to account for an
1,800-hour work-year and inflation, and multiplied by 5.35 to
account for bonuses, firm size, employee benefits, and overhead.
\1831\ The Commission arrived at this estimate by dividing the
initial external cost estimate provided in the MDI Rules for each
exclusive SIP to become a competing consolidator by three to account
for the fact that the exclusive SIPs would not need to build
aggregation systems in three separate data centers to collect,
calculate, and disseminate odd-lot information. See MDI Adopting
Release, supra note 10, at 18712-13.
\1832\ The Commission estimates the monetized initial burden for
this requirement to be $335,340, broken down as follows: [(Sr.
Programmer at $399/hour for 210 hours) + (Sr. Systems Analyst at
$343/hour for 180 hours) + (Compliance Manager at $373/hour for 20
hours) + (Director of Compliance at $588/hour for 10 hours) +
(Compliance Attorney at $440/hour for 20 hours)] x [(2 exclusive
SIPs)] = 880 total initial burden hours across the exclusive SIPs.
\1833\ The Commission estimates total initial external costs as
follows: initial external costs of $412,500 per exclusive SIP x (2
exclusive SIPs) = $825,000.
---------------------------------------------------------------------------
2. Ongoing Burden Hours and Costs
The Commission believes that the two exclusive SIPs will incur
annual ongoing burden hours and external costs to operate and maintain
their modified systems to collect, calculate, and disseminate odd-lot
information and to disseminate the round-lot and minimum pricing
increment indicators. The Commission estimates that each exclusive SIP
will incur 132 ongoing, annual burden hours \1834\ and ongoing, annual
external costs of $123,725 to operate and maintain its systems to
collect, calculate, and disseminate odd-lot information and to
disseminate the round-lot and minimum pricing increment
indicators.\1835\ Thus, the Commission estimates that the total
ongoing, annual burden hours for two exclusive SIPs will be 264 burden
hours \1836\ and that total ongoing, annual external costs would be
$247,450.\1837\
---------------------------------------------------------------------------
\1834\ The Commission estimates the monetized annual ongoing
burden for this requirement to be $50,301, broken down as follows:
[(Sr. Programmer at $399/hour for 63 hours) + (Sr. Systems Analyst
at $343/hour for 54 hours) + (Compliance Manager at $373/hour for 6
hours) + (Director of Compliance at $588/hour for 3 hours) +
(Compliance Attorney at $440/hour for 6 hours)] = 132 ongoing,
annual burden hours to operate and maintain its systems to comply
with the requirement to collect, calculate, and disseminate odd-lot
information. The Commission based these estimates on 10% of the
ongoing, annual burden hour estimates provided in the MDI Rules for
each exclusive SIP competing consolidator to operate and maintain
its systems to comply with Rules 614(d)(1) through (4) to account
for the fact that these amendments do not require the exclusive SIPs
to calculate and disseminate full consolidated market data (e.g.,
depth of book data or auction information) as defined in the MDI
Rules. See MDI Adopting Release, supra note 10, at 18712-13. The
Commission derived the hourly rate figures from SIFMA's Management &
Professional Earnings in the Securities Industry 2013, modified to
account for an 1,800-hour work-year and inflation, and multiplied by
5.35 to account for bonuses, firm size, employee benefits, and
overhead.
\1835\ The Commission arrived at this estimate by dividing by
three the ongoing, annual external cost estimate provided in the MDI
Rules for each exclusive SIP competing consolidator to operate and
maintain its systems to comply with rules 614(d)(1) through (4) to
account for the fact that the exclusive SIPs will not need to build
aggregation systems in three separate data centers to collect,
calculate, and disseminate odd-lot information. See MDI Adopting
Release, supra note 10, at 18712-13.
\1836\ The Commission estimates the monetized annual ongoing
burden for this requirement to be $100,602, broken down as follows:
[(Sr. Programmer at $399/hour for 63 hours) + (Sr. Systems Analyst
at $343/hour for 54 hours) + (Compliance Manager at $373/hour for 6
hours) + (Director of Compliance at $588/hour for 3 hours) +
(Compliance Attorney at $440/hour for 6 hours) x (2 exclusive SIPs)]
= 264 total ongoing, annual burden hours across the exclusive SIPs.
\1837\ The Commission estimates total annual ongoing external
costs as follows: annual ongoing external costs of $123,725 per
exclusive SIP x (2 exclusive SIPs) = $247,450.
---------------------------------------------------------------------------
[[Page 81770]]
E. Collection of Information Is Mandatory
The collection of information discussed above is a mandatory.
F. Confidentiality
This information collection will be public.
G. Revisions to Current MDI Rules Burden Estimates
Currently, the MDI Rules impose ``collection of information''
requirements within the meaning of the PRA. Specifically, pursuant to
Rule 603(b), SROs are required to make available all data necessary to
generate consolidated market data to competing consolidators and self-
aggregators. As explained in more detail below, the Commission is
revising the burden estimates associated with this requirement in light
of the amendments. In the MDI Rules, the Commission estimated that each
SRO will require an average of 220 initial burden hours of legal,
compliance, information technology, and business operations personnel
time to prepare and implement a system to collect the information
necessary to generate consolidated market data (for a total cost per
SRO of $70,865).\1838\ The Commission estimated that each SRO would
incur an annual average burden on an ongoing basis of 396 hours to
collect the information necessary to generate consolidated market data
required by Rule 603(b) (for a total cost per SRO of $128,064).\1839\
---------------------------------------------------------------------------
\1838\ In the MDI Adopting Release, the Commission estimated the
monetized initial burden for this requirement to be $70,865. The
Commission derived this estimate based on per hour figures from
SIFMA's Management & Professional Earnings in the Securities
Industry 2013, modified to account for an 1,800-hour work-year and
inflation, and multiplied by 5.35 to account for bonuses, firm size,
employee benefits, and overhead: [(Compliance Manager at $310 for
105 hours) + (Attorney at $417 for 70 hours) + (Sr. Systems Analyst
at $285 for 20 hours) + (Operations Specialist at $137 for 25
hours)] = 220 initial burden hours and $70,865.
\1839\ In the MDI Adopting Release, the Commission estimated the
monetized ongoing, annual burden for this requirement to be
$128,064. The Commission derived this estimate based on per hour
figures from SIFMA's Management & Professional Earnings in the
Securities Industry 2013, modified to account for an 1,800-hour
work-year and inflation, and multiplied by 5.35 to account for
bonuses, firm size, employee benefits, and overhead: [(Compliance
Manager at $310 for 192 hours) + (Attorney at $417 for 48 hours) +
(Sr. Systems Analyst at $285 for 96 hours)] = 336 initial burden
hours and $128,064.
---------------------------------------------------------------------------
As described above, the amendments to Rule 603(b) require SROs to
make available all data necessary to generate odd-lot information to
the exclusive SIPs whereas, under the decentralized consolidation model
set forth in the MDI Rules, consolidated market data would be provided
by competing consolidators and self-aggregators. The SROs already
provide certain quotation information to the exclusive SIPs, and many
SROs already provide odd-lot quotation information to customers through
their proprietary data feeds.\1840\ Nevertheless, providing the
exclusive SIPs with the data necessary to generate odd-lot information
may entail additional burdens. Specifically, technical development work
may be needed to direct odd-lot quotations to the exclusive SIPs and to
expand the capacity of the existing connections (including acquiring
the necessary hardware, network capabilities and power) through which
the SROs provide data to the exclusive SIPs to support the additional
message traffic associated with odd-lot quotations. Therefore, the
Commission is revising its burden estimates for Rule 603(b) upwards by
5% to account for the provision of the data necessary to generate odd-
lot information to the exclusive SIPs.\1841\ Specifically, the
Commission is adding 11 initial burden hours \1842\ and 19.8 annual
burden hours \1843\ to its previous estimates.
---------------------------------------------------------------------------
\1840\ See MDI Adopting Release, supra note 10, at 18599.
\1841\ The Commission believes that 5% of the initial and
ongoing, annual burden hour estimates provided in the MDI Rules for
each SRO to make the data necessary to generate consolidated market
data available to competing consolidators and self-aggregators is
appropriate because the SROs already collect the data necessary to
generate odd-lot information and this information is a subset of
consolidated market data as defined in the MDI Rules.
\1842\ The Commission estimates the monetized initial burden for
this requirement to be $4,261. The Commission derived this estimate
based on per hour figures from SIFMA's Management & Professional
Earnings in the Securities Industry 2013, modified to account for an
1,800-hour work-year and inflation, and multiplied by 5.35 to
account for bonuses, firm size, employee benefits, and overhead:
[(Compliance Manager at $373 for 5.25 hours) + (Attorney at $501 for
3.5 hours) + (Sr. Systems Analyst at $343 for 1 hour) + (Operations
Specialist at $165 for 1.25 hours)] = 11 initial burden hours and
$4,261.
\1843\ The Commission estimates the monetized ongoing, annual
burden for this requirement to be $7,646.6. The Commission derived
this estimate based on per hour figures from SIFMA's Management &
Professional Earnings in the Securities Industry 2013, modified to
account for an 1,800-hour work-year and inflation, and multiplied by
5.35 to account for bonuses, firm size, employee benefits, and
overhead: [(Compliance Manager at $373 for 10.6 hours) + (Attorney
at $501 for 3.4 hours) + (Sr. Systems Analyst at $343 for 5.8
hours)] = 19.8 annual burden hours and $7,646.6.
---------------------------------------------------------------------------
In addition, the amendments require the primary listing exchange
for each NMS stock to provide an indicator of the round lot size to the
applicable exclusive SIP for dissemination and to calculate and provide
to competing consolidators, self-aggregators, and the applicable
exclusive SIP an indicator of the applicable minimum pricing increment
for dissemination. The primary listing exchange is already required to
calculate the applicable round lot size and provide it to competing
consolidators and self-aggregators under the MDI Rules, and the
incremental burden of providing this indicator to the two exclusive
SIPs is likely to be minimal. However, calculating the applicable
minimum pricing increment and providing it to competing consolidators,
self-aggregators, and the exclusive SIPs will entail additional
burdens.
Specifically, primary listing exchanges will need to program
systems to calculate the applicable minimum pricing increment for each
NMS stock that they list semiannually based on its TWAQS and to include
this information in the data that they provide to competing
consolidators, self-aggregators, and the exclusive SIPs. Therefore, the
Commission revising its burden estimates for Rule 603(b) upwards to
account for the calculation of the applicable minimum pricing increment
and the provision of this information to competing consolidators, self-
aggregators, and the exclusive SIPs. Specifically, the Commission is
adding 50 initial burden hours \1844\ and 32 annual burden hours \1845\
for each primary listing exchange to its previous estimates and 250
total initial burden hours \1846\ and 160 total annual burden hours
\1847\ for five primary listing exchanges.
---------------------------------------------------------------------------
\1844\ The Commission estimates the monetized initial burden for
this requirement to be $19,000 per primary listing exchange.
Salaries are derived from SIFMA's Management & Professional Earnings
in the Securities Industry 2013, modified to account for an 1,800-
hour work-year and inflation, and multiplied by 5.35 to account for
bonuses, firm size, employee benefits and overhead: [(Sr. Programmer
at $368 for 25 hours) + (Sr. Systems Analyst at $316 for 10 hours) +
(Compliance Manager at $344 for 10 hours) + (Director of Compliance
at $542 for 5 hour)] [ap] $19,000 per listing exchange). See supra
notes 1644-1645 and accompanying text.
\1845\ The Commission estimates the monetized ongoing, annual
burden for this requirement to be $9,000 per primary listing
exchange. ((Compliance Attorney at $406 for 6 hours) + (Compliance
Manager at $344 for 2 hours)) x 4 tick size revisions per year] [ap]
$9,000 per listing exchange. Id.
\1846\ 50 initial burden hours per primary listing exchange x 5
primary listing exchanges = 250 total initial burden hours. The
Commission estimates the total monetized initial burden of this
requirement to be $95,000 ($19,000 per primary listing exchange x 5
primary listing exchanges = $95,000). Id.
\1847\ 32 annual burden hours per primary listing exchange x 5
primary listing exchanges = 160 total annual burden hours. The
Commission estimates the total monetized annual burden of this
requirement to be $45,000 ($9,000 per primary listing exchange x 5
primary listing exchanges = $45,000). Id.
---------------------------------------------------------------------------
In addition, the MDI Rules include a collection of information
requirement
[[Page 81771]]
under rules 614(d)(1) through (3), which require competing
consolidators to collect from the SROs quotation and transaction
information for NMS stocks, calculate and generate a consolidated
market data product, and make the consolidated market data product
available to subscribers.\1848\ As discussed above, the amended
definition of odd-lot information includes a specified best odd-lot
order to buy and best odd-lot order to sell. Since the odd-lot quotes
that a competing consolidator would use to identify and disseminate the
best odd-lot orders--if the competing consolidator offers a
consolidated market data product that includes this information--are
already included in the data necessary to generate odd-lot information,
the Commission believes that the existing burden estimates for rules
614(d)(1) through (3) account for the identification and dissemination
of the best odd-lot orders.
---------------------------------------------------------------------------
\1848\ MDI Adopting Release, supra note 10, at 18703.
---------------------------------------------------------------------------
IX. Regulatory Flexibility Act
The Regulatory Flexibility Act (``RFA'') requires the Commission,
in promulgating rules,\1849\ to consider the impact of those rules on
small entities. This Final Regulatory Flexibility Analysis has been
prepared in accordance with section 604 of the RFA.\1850\ The
Commission prepared an Initial Regulatory Flexibility Analysis and a
Regulatory Flexibility Act certification in accordance with the RFA and
included in the Proposing Release.\1851\
---------------------------------------------------------------------------
\1849\ 5 U.S.C. 553.
\1850\ See 5 U.S.C. 604.6.
\1851\ See section VIII of the Proposing Release, supra note 11.
---------------------------------------------------------------------------
In the Proposing Release, the Commission certified that the
proposed amendments to Rules 600, 603 and 610 would not have a
significant economic impact on a substantial number of small entities
for purposes of the RFA.\1852\ The Proposing Release solicited comments
on the certification. The Commission received no comments on this
certification.
---------------------------------------------------------------------------
\1852\ See supra id.
---------------------------------------------------------------------------
With respect to Rule 612, an initial Regulatory Flexibility
Analysis (``IFRA'') was prepared in accordance with the RFA and was
included in the Proposing Release.\1853\ The Commission has prepared
this Final Regulatory Flexibility Analysis (``FRFA'') in accordance
with section 604 of the RFA.\1854\ The Commission did not receive
comments on the IRFA.
---------------------------------------------------------------------------
\1853\ See section VIII of the Proposing Release, supra note 11.
See supra section VII.B.
\1854\ 5 U.S.C. 604.
---------------------------------------------------------------------------
A. Amendments to Rule 612--Final Regulatory Flexibility Analysis
1. Reasons for the Action
As discussed in section III, the Commission is adopting amendments
to Rule 612 to update and modernize the rule for the current trading
environment. As adopted, Rule 612 will reduce minimum pricing increment
for orders and quotes priced $1.00 or greater for certain NMS stocks.
2. Small Entities Subject to the Rule
Rule 612 would apply to national securities exchanges, national
securities associations, ATSs, vendors, and broker or dealers.
National securities exchanges are not small entities as defined by
Commission rules. Exchange Act rule 0-10(e) \1855\ states that the term
``small business'' when referring to an exchange means any exchange
that has been exempted from the reporting requirements of Exchange Act
rule 601 and is not affiliated with any person that is not a small
business or small organization. There is only one national securities
association, and the Commission has previously stated that it is not a
small entity as defined by 13 CFR 121.201.\1856\
---------------------------------------------------------------------------
\1855\ 17 CFR 240.0-10(e).
\1856\ See Securities Exchange Act Release No. 62174 (May 26,
2010), 75 FR 32556 (June 8, 2010) (``FINRA is not a small entity as
defined by 13 CFR 121.201.'').
---------------------------------------------------------------------------
Commission rule 0-10(c) defines a broker-dealer as a small entity
for the purpose of this section if the broker-dealer had a total
capital (net worth plus subordinated liabilities) of less than $500,000
on the date in the prior fiscal year as of which its audited financial
statements were prepared, had less than $200 million of funds and
securities in its custody of control at all times during the preceding
fiscal year, and the broker-dealer is not affiliated with any person
(other than a natural person) that is not a small entity.\1857\ The
Commission is updating the estimate from the Proposing Release and
estimates that as of December 31, 2023, there were approximately 734
Commission registered broker-dealers that would be small entities for
purposes of the statute that would be required to comply with the
amendments to Rule 612 regarding quotation in the minimum pricing
increments.\1858\ The updated estimate number is approximately 3.5%
lower and does not impact the Commission's analysis.
---------------------------------------------------------------------------
\1857\ 17 CFR 240.0-10(c).
\1858\ In the Proposing Release, the Commission estimated that
as of June 30, 2022, there were approximately 761 Commission
registered broker-dealers that would be small entities for purposes
of the statute.
---------------------------------------------------------------------------
Rule 612 applies to NMS stocks and the rule would apply to NMS
Stock ATSs. NMS Stock ATSs that are not registered as exchanges are
required to register as broker-dealers.\1859\ Accordingly, NMS Stock
ATSs would be considered small entities if they fall within the
standard for small entities that would apply to broker-dealers. The
Commission examined FOCUS data for the 33 broker-dealers that currently
operate NMS Stock ATSs and, applying the test for broker-dealers
described above, believes that none of the NMS Stock ATSs currently
trading NMS stocks were operated by a broker-dealer that is a ``small
entity.'' \1860\
---------------------------------------------------------------------------
\1859\ See rule 301(b)(1) of Regulation ATS.
\1860\ A list of NMS Stock ATSs with Form ATS on file with the
Commission is available at https://www.sec.gov/about/divisions-offices/division-trading-markets/alternative-trading-systems/form-ats-n-filings-information#ats-n. The Commission examined the list as
of January 31, 2024. The number of broker-dealers that operate NMS
Stock ATSs has not changed from the Proposing Release.
---------------------------------------------------------------------------
A vendor is defined in rule 600(b)(100) of Regulation NMS as any
SIP engaged in the business of disseminating transaction reports, last
sale data, or quotations with respect to NMS securities to brokers,
dealers, or investors on a real-time or other current and continuing
basis, whether through an electronic communications network, moving
ticker, or interrogation device.\1861\ Commission rule 0-10(g) states
that the term small business when referring to a SIP, means any SIP
that had gross revenues of less than $10 million during the preceding
year, provided service to fewer than 100 interrogation devices or
moving tickers at all times during the preceding year, and is not
affiliated with any person that is not a small business or small
organization.\1862\ The Commission estimates as of August 31, 2022,
that there are approximately 80 vendors, 13 of which would be small
entities.
---------------------------------------------------------------------------
\1861\ See 17 CFR 242.600(b)(100).
\1862\ See 17 CFR 242.0-10(g).
---------------------------------------------------------------------------
3. Reporting, Recordkeeping, and Other Compliance Requirements
Rule 612 will no impose any new reporting, recordkeeping, or other
compliance requirements on market participants that are small entities.
4. Significant Alternatives
Pursuant to section 3 of the RFA, the Commission must consider the
following types of alternatives: (a) the establishment of differing
compliance or
[[Page 81772]]
reporting requirements or timetables that take into account the
resources available to small entities; (b) the clarification,
consolidation, or simplification of compliance and reporting
requirements under the proposed rule for small entities; (c) the use of
performance rather than design standards; and (d) an exemption from
coverage of the proposed rule, or any part thereof, for small entities.
The primary goal of Rule 612 is to provide uniform minimum pricing
increments for NMS stocks. This primary goal continues with the
amendments to Rule 612. As such, imposing different compliance or
reporting requirements or possibly a different timetable for
implementing compliance or reporting requirements, for small entities,
could undermine the goal of uniformity. In addition, the Commission has
concluded similarly that it would not be consistent with the primary
goal to further clarify, consolidate, or simplify the amendments to
Rule 612 for small entities. The amendments to Rule 612 are performance
standards and do not dictate for entities of any size any particular
design standards, e.g., technology, that must be employed to achieve
the objectives of the rule. It would be inconsistent with the purposes
of the Exchange Act to specify different requirements for small
entities or to exempt broker-dealers from the amendments to Rule 612.
B. Amendments to Rule 610
The changes to Rule 610(c) would apply to trading centers as
defined in rule 600(b)(95) that impose fees for access against a
protected quotation or any other quotation of the trading center that
is the best bid or best offer of a national securities exchange or
national securities association. As discussed above, currently national
securities exchanges are the only trading centers publishing protected
quotations. Pursuant to rule 0-10(e), none of the national securities
exchanges are small entities for purposes of the RFA.\1863\
---------------------------------------------------------------------------
\1863\ See 17 CFR 240.0-10(e).
---------------------------------------------------------------------------
New Rule 610(d) will require all fees charged and rebates paid by
national securities exchanges to be determinable at the time of
execution. Pursuant to rule 1-10(e), none of the national securities
exchanges are small entities for purposes of the RFA.\1864\
---------------------------------------------------------------------------
\1864\ Id.
---------------------------------------------------------------------------
C. Amendments to Rule 603 and Definitions Odd-Lot Information and
Regulatory Data Under Rule 600
The amendments to Rule 603(b) and to the definitions of odd-lot
information and regulatory data in rule 600(b) would apply to national
securities exchanges registered with the Commission under section 6 of
the Exchange Act, national securities associations registered with the
Commission under section 15A of the Exchange Act, and the exclusive
SIPs. As stated above, pursuant to rule 0-10(e), none of the national
securities exchanges small entities for the purposes of the RFA.\1865\
There is one national securities association, and the Commission has
previously stated that it is not a small entity.\1866\ With respect to
the exclusive SIPs, neither SIAC nor Nasdaq meet the criteria for a
``small business'' or ``small organization'' when used with reference
to a securities information processor.\1867\ Thus the amendments to
rules 600(b) and 603(b) would not affect any small entities. For the
purposes of the RFA, the Commission certifies that the amendments to
Rule 603(b) and rule 600(b) would not have a significant economic
impact on a substantial number of small entities.
---------------------------------------------------------------------------
\1865\ See 17 CFR 240.0-10(e).
\1866\ See supra note 1856.
\1867\ See 17 CFR 240.0-10(g). See also Securities Exchange Act
Release No. 61595 (Feb. 26, 2010), 75 FR 11232, 11320 (Mar. 10,
2010) (determining that SIAC and Nasdaq are not small entities for
purposes of the RFA).
---------------------------------------------------------------------------
D. Certification
For the reasons described above, the Commission certifies that the
final amendments to Rules 600, 603(b) and 610 would not have a
significant economic impact on a substantial number of small entities
for purposes of the RFA.
X. Other Matters
Pursuant to the Congressional Review Act,\1868\ the Office of
Information and Regulatory Affairs has designated these rules as a
``major rule'' as defined by 5 U.S.C. 804(2). The Commission considers
the provisions of the final amendments to be severable to the fullest
extent permitted by law. ``If parts of a regulation are invalid and
other parts are not,'' courts ``set aside only the invalid parts unless
the remaining ones cannot operate by themselves or unless the agency
manifests an intent for the entire package to rise or fall together.''
Bd. of Cnty. Commissioners of Weld Cnty. v. EPA, 72 F.4th 284, 296
(D.C. Cir. 2023); see K mart Corp. v. Cartier, Inc., 486 U.S. 281, 294
(1988). ``In such an inquiry, the presumption is always in favor of
severability.'' Cmty. for Creative Non-Violence v. Turner, 893 F.2d
1387, 1394 (D.C. Cir. 1990). Consistent with these principles, while
the Commission believes that all provisions of the final amendments are
fully consistent with governing law, if any of the provisions of these
amendments, or the application thereof to any person or circumstance,
is held to be invalid, the Commission intends that such invalidity
shall not affect other provisions or application of such provisions to
other persons or circumstances that can be given effect without the
invalid provision or application. In particular, the amendments
relating to round lots, odd-lot information, Rule 610(c), and Rule
610(d) can operate independently from each other and from the
amendments related to Rule 612. Additionally, the amendments to Rule
612 can operate independently from the amendments relating to round
lots, odd-lot information, and Rule 610(d).
---------------------------------------------------------------------------
\1868\ 5 U.S.C. 801 et seq.
---------------------------------------------------------------------------
Statutory Authority and Text of Rule Amendments
Pursuant to the Exchange Act, and particularly sections 2, 3(b), 5,
6, 11, 11A, 15, 15A, 17, 19, 23(a), and 36 thereof, 15 U.S.C. 78b, 78c,
78e, 78f, 78k, 78k-1, 78o, 78o-3, 78q, 78s, 78w(a), and 78mm the
Commission is amending sections 242.600, 242.603, 242.610, and 242.612
of chapter II of title 17 of the Code of Federal Regulations.
List of Subjects in 17 CFR Part 242
Brokers, Confidential business information, Fraud, Reporting and
recordkeeping requirements, Securities.
For the reasons stated in the preamble, the Commission is amending
title 17, chapter II of the Code of Federal Regulations as follows:
PART 242--REGULATIONS M, SHO, ATS, AC, NMS, AND SBSR, AND CUSTOMER
MARGIN REQUIREMENTS FOR SECURITY FUTURES
0
1. The authority citation for part 242 continues to read as follows:
Authority: 15 U.S.C. 77g, 77q(a), 77s(a), 78b, 78c, 78c-4,
78g(c)(2), 78i(a), 78j, 78k-1(c), 78l, 78m, 78n, 78o(b), 78o(c),
78o(g), 78q(a), 78q(b), 78q(h), 78w(a), 78dd-1, 78mm, 80a-23, 80a-
29, 80a-37, and 8343.
0
2. Amend Sec. 242.600 by:
0
a. In paragraph (b)(69)(i), removing the word ``and'' from the end of
the paragraph;
0
b. In paragraph (b)(69)(ii), removing the period at the end of the
paragraph and adding the text ``; and'' in its place;
0
c. Adding paragraph (b)(69)(iii);
[[Page 81773]]
0
d. In paragraph (b)(89)(i)(D), removing the word ``and'' from the end
of the paragraph;
0
e. In paragraph (b)(89)(i)(E), removing the period from the end of the
paragraph and adding the text ``; and'' in its place;
0
f. Adding paragraphs (b)(89)(i)(F) and (89)(iv); and
0
g. Revising and republish paragraph (b)(93).
The additions and revisions read as follows:
Sec. 242.600 NMS security designation and definitions.
* * * * *
(b) * * *
(69) * * *
(iii) Best odd-lot order to buy and best odd-lot order to sell. The
best odd-lot order to buy means the highest priced odd-lot order to buy
that is priced higher than the national best bid, and the best odd-lot
order to sell means the lowest priced odd-lot order to sell that is
priced lower than the national best offer, for an NMS stock that are
calculated and disseminated on a current and continuing basis by a
competing consolidator or plan processor or calculated by a self-
aggregator; provided, that in the event two or more market centers
transmit to a competing consolidator, plan processor, or a self-
aggregator identical odd-lot buy orders or odd-lot sell orders for an
NMS stock, the highest priced odd-lot buy order or lowest priced odd-
lot sell order (as the case may be) shall be determined by ranking all
such identical odd-lot buy orders or odd-lot sell orders (as the case
may be) first by size (giving the highest ranking to the odd-lot buy
order or odd-lot sell order associated with the largest size), and then
by time (giving the highest ranking to the odd-lot buy order or odd-lot
sell order received first in time).
* * * * *
(89) * * *
(i) * * *
(F) An indicator of the applicable minimum pricing increment
required under Sec. 242.612.
* * * * *
(iv) The primary listing exchange shall also provide the
information required under paragraphs (b)(89)(i)(E) and (F) of this
section to the applicable plan processor for dissemination.
* * * * *
(93) Round lot means:
(i) For any NMS stock for which the average closing price on the
primary listing exchange during the prior Evaluation Period was:
(A) $250.00 or less per share, an order for the purchase or sale of
an NMS stock of 100 shares;
(B) $250.01 to $1,000.00 per share, an order for the purchase or
sale of an NMS stock of 40 shares;
(C) $1,000.01 to $10,000.00 per share, an order for the purchase or
sale of an NMS stock of 10 shares;
(D) $10,000.01 or more per share, an order for the purchase or sale
of an NMS stock of 1 share; and
(ii) New NMS stocks. Any security that becomes an NMS stock during
an operative period as described in paragraph (b)(93)(iv) of this
section shall be assigned a round lot of 100 shares.
(iii) For purposes of this section only, the Evaluation Period
means:
(A) All trading days in March for the round lot assigned on the
first business day of May; and
(B) All trading days in September for the round lot assigned on the
first business day of November during which the average closing price
of an NMS stock on the primary listing exchange shall be measured by
the primary listing exchange to determine the round lot for each NMS
stock.
(iv) The round lot assigned under this section shall be operative
on:
(A) The first business day of May for the March Evaluation Period
and continue through the last business day of October of the calendar
year; and
(B) The first business day of November for the September Evaluation
Period and continue through the last business day of April of the next
calendar year.
* * * * *
0
3. Amend Sec. 242.603 by revising the section heading and paragraph
(b) to read as follows:
Sec. 242.603 Distribution, consolidation, dissemination, and display
of information with respect to quotations for and transactions in NMS
stocks.
* * * * *
(b) Consolidation and dissemination of information. (1) Application
of paragraphs (b)(2) and (3) of this section:
(i) Compliance with paragraph (b)(3) of this section is required
until the date indicated by the Commission in any order approving
amendments to the effective national market system plan(s) to
effectuate a cessation of the operations of the plan processors that
disseminate consolidated information regarding NMS stocks.
(ii) Compliance with paragraph (b)(2) of this section is required
180 calendar days from the date of the Commission's approval of the
amendments, filed as required under Sec. 242.614(e), to the effective
national market system plan(s).
(2) Every national securities exchange on which an NMS stock is
traded and national securities association shall act jointly pursuant
to one or more effective national market system plans for the
dissemination of consolidated market data. Every national securities
exchange on which an NMS stock is traded and national securities
association shall make available to all competing consolidators and
self-aggregators its information with respect to quotations for and
transactions in NMS stocks, including all data necessary to generate
consolidated market data, in the same manner and using the same
methods, including all methods of access and the same format, as such
national securities exchange or national securities association makes
available any information with respect to quotations for and
transactions in NMS stocks to any person.
(3) Every national securities exchange on which an NMS stock is
traded and national securities association shall act jointly pursuant
to one or more effective national market system plans to disseminate
consolidated information, including a national best bid and national
best offer and odd-lot information, on quotations for and transactions
in NMS stocks. Such plan or plans shall provide for the dissemination
of all consolidated information for an individual NMS stock through a
single plan processor and such single plan processor must represent
quotation sizes in such consolidated information in terms of the number
of shares, rounded down to the nearest multiple of a round lot. Every
national securities exchange on which an NMS stock is traded and
national securities association shall make available to a plan
processor all data necessary to generate odd-lot information.
* * * * *
0
4. Amend Sec. 242.610 by:
0
a. Revising paragraph (c);
0
b. Redesignating paragraphs (d) and (e) as paragraphs (e) and (f); and
0
c. Adding new paragraph (d).
The revisions and addition read as follows:
Sec. 242.610 Access to quotations.
* * * * *
(c) Fees for access to quotations. A trading center shall not
impose, nor permit to be imposed, any fee or fees for the execution of
an order against a protected quotation of the trading center or against
any other quotation of the trading center that is the best bid or best
offer of a national securities exchange or the best bid or best offer
of a national securities association in an NMS stock that exceed or
accumulate to more than the following limits:
[[Page 81774]]
(1) If the price of a protected quotation or other quotation is
$1.00 or more, the fee or fees cannot exceed or accumulate to more than
$0.001 per share; or
(2) If the price of a protected quotation or other quotation is
less than $1.00, the fee or fees cannot exceed or accumulate to more
than 0.1% of the quotation price per share.
(d) Transparency of fees. A national securities exchange shall not
impose, nor permit to be imposed, any fee or fees, or provide, or
permit to be provided, any rebate or other remuneration, for the
execution of an order in an NMS stock that cannot be determined at the
time of execution.
* * * * *
0
5. Revise Sec. 242.612 to read as follows:
Sec. 242.612 Minimum pricing increment.
(a) Definitions. For purposes of this section only, the following
terms shall have the meanings set forth in this section.
(1) Evaluation Period means:
(i) The three months from January through March of a calendar year;
and
(ii) The three months from July through September of a calendar
year during which the Time Weighted Average Quoted Spread of an NMS
stock shall be measured by the primary listing exchange to determine
the minimum pricing increment for each NMS stock.
(2) Time Weighted Average Quoted Spread means the average dollar
value difference between the NBB and NBO during regular trading hours
where each instance of a unique NBB and NBO is weighted by the length
of time that the quote prevailed as the NBB or NBO.
(b) Minimum pricing increments. (1) The minimum pricing increment
under paragraph (b)(2) of this section shall be operative on:
(i) The first business day of May for the Evaluation Period from
January through March and continue through the last business day of
October of the calendar year; and
(ii) The first business day of November for the Evaluation Period
from July through September and continue through the last business day
of April of the next calendar year.
(2) No national securities exchange, national securities
association, alternative trading system, vendor, or broker or dealer
shall display, rank, or accept from any person a bid or offer, an
order, or an indication of interest in any NMS stock in an increment
smaller than required pursuant to either paragraph (b)(2)(i) or (ii) of
this section if that bid or offer, order, or indication of interest is
priced equal to or greater than $1.00 per share:
(i) $0.01, if the Time Weighted Average Quoted Spread for the NMS
stock during the Evaluation Period was greater than, $0.015; or
(ii) $0.005, if the Time Weighted Average Quoted Spread for the NMS
stock during the Evaluation Period was equal to or less than $0.015.
(3) No national securities exchange, national securities
association, alternative trading system, vendor, or broker or dealer
shall display, rank, or accept from any person a bid or offer, an
order, or an indication of interest in any NMS stock priced in an
increment smaller than $0.0001 if that bid or offer, order, or
indication of interest is priced less than $1.00 per share.
(c) New NMS Stocks. Any security that becomes an NMS Stock during
an operative period as described in paragraph (b)(1) of this section
shall be assigned a minimum pricing increment of $0.01 for bids or
offers, orders, or indications of interest priced equal to or greater
than $1.00 per share.
(d) Exemptions. The Commission, by order, may exempt from the
provisions of this section, either unconditionally or on specified
terms and conditions, any person, security, quotation, or order, or any
class or classes of persons, securities, quotations, or orders, if the
Commission determines that such exemption is necessary or appropriate
in the public interest, and is consistent with the protection of
investors.
By the Commission.
Dated: September 18, 2024.
Vanessa A. Countryman,
Secretary.
[FR Doc. 2024-21867 Filed 10-7-24; 8:45 am]
BILLING CODE 8011-01-P