[Federal Register Volume 81, Number 130 (Thursday, July 7, 2016)]
[Notices]
[Pages 44359-44363]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2016-16109]



[[Page 44359]]

-----------------------------------------------------------------------

SECURITIES AND EXCHANGE COMMISSION

[Release No. 34-78219; File No. SR-FINRA-2016-024]


Self-Regulatory Organizations; Financial Industry Regulatory 
Authority, Inc.; Notice of Filing of a Proposed Rule Change To Amend 
FINRA Rule 7730 (Trade Reporting and Compliance Engine (TRACE))

July 1, 2016.
    Pursuant to Section 19(b)(1) of the Securities Exchange Act of 1934 
(``Act'') \1\ and Rule 19b-4 thereunder,\2\ notice is hereby given that 
on June 28, 2016, Financial Industry Regulatory Authority, Inc. 
(``FINRA'') filed with the Securities and Exchange Commission (``SEC'' 
or ``Commission'') the proposed rule change as described in Items I, 
II, and III below, which Items have been prepared by FINRA. The 
Commission is publishing this notice to solicit comments on the 
proposed rule change from interested persons.
---------------------------------------------------------------------------

    \1\ 15 U.S.C. 78s(b)(1).
    \2\ 17 CFR 240.19b-4.+
---------------------------------------------------------------------------

I. Self-Regulatory Organization's Statement of the Terms of Substance 
of the Proposed Rule Change

    FINRA is proposing to amend FINRA Rule 7730 to create a new 
Academic Corporate Bond TRACE Data product that would be available to 
institutions of higher education.
    The text of the proposed rule change is available on FINRA's Web 
site at http://www.finra.org, at the principal office of FINRA and at 
the Commission's Public Reference Room.

II. Self-Regulatory Organization's Statement of the Purpose of, and 
Statutory Basis for, the Proposed Rule Change

    In its filing with the Commission, FINRA included statements 
concerning the purpose of and basis for the proposed rule change and 
discussed any comments it received on the proposed rule change. The 
text of these statements may be examined at the places specified in 
Item IV below. FINRA has prepared summaries, set forth in sections A, 
B, and C below, of the most significant aspects of such statements.

A. Self-Regulatory Organization's Statement of the Purpose of, and the 
Statutory Basis for, the Proposed Rule Change

1. Purpose
    FINRA Rule 7730 sets forth the TRACE data products offered by 
FINRA. FINRA's data product offerings include both real-time as well as 
historic data for most TRACE-eligible securities. FINRA is proposing to 
create a new Academic Corporate Bond TRACE Data product, which would be 
made available solely to institutions of higher education and would 
include masked dealer identifiers.
    FINRA periodically receives requests from academics for access to 
TRACE data. FINRA's existing Historic TRACE Data product provides 
transaction-level data on an 18-month delayed basis for all 
transactions that have been reported to TRACE in the classes of TRACE-
eligible securities that currently are disseminated.\3\ While Historic 
TRACE Data is used by academic researchers today, it does not include 
any identifying information regarding the dealer reporting each 
transaction. Thus, where a researcher wishes to track the behavior of 
an individual dealer or group of dealers--even anonymously--the 
existing Historic TRACE Data product would not allow for this type of 
observation. As a result, academics have requested that FINRA make 
available an enhanced version of Historic TRACE Data that would include 
dealer identification.
---------------------------------------------------------------------------

    \3\ FINRA adopted the Historic TRACE Data rule and related fees 
in 2010. See Securities Exchange Act Release No. 61012 (November 16, 
2009), 74 FR 61189 (November 23, 2009) (Order Approving File No. SR-
FINRA-2007-006). See also Regulatory Notice 10-14 (March 2010).
    \4\ Rule 6710(q) generally defines ``List or Fixed Offering 
Price Transaction'' as a primary market sale transaction sold on the 
first day of trading of a security excluding a Securitized Product 
other than an Asset-Backed Security as defined in Rule 6710(cc): (i) 
By a sole underwriter, syndicate manager, syndicate member or 
selling group member at the published or stated list or fixed 
offering price, or (ii) in the case of a primary market sale 
transaction effected pursuant to Securities Act Rule 144A, by an 
initial purchaser, syndicate manager, syndicate member or selling 
group member at the published or stated fixed offering price.
---------------------------------------------------------------------------

    In response to these requests from academics, the proposed rule 
change would create a new Academic Corporate Bond TRACE Data product 
that would include transaction-level data on corporate bonds (except a 
transaction that is a List or Fixed Offering Price Transaction \4\ or a 
Takedown
Transaction \5\),\6\ including Rule 144A transactions in corporate 
bonds, with masked dealer identifiers. Masked dealer identifiers may be 
useful to academics in a variety of ways--for example, to enable 
researchers to track activity by individual dealers or group of dealers 
and observe their behaviors, and may facilitate the ability of academic 
researchers to study the impact of various events on measures such as 
intermediation costs, dealer participation and liquidity. Academic 
Corporate Bond TRACE Data would be made available only to academics 
(i.e., requests originating from an institution of higher 
education).\7\
---------------------------------------------------------------------------

    \5\ Rule 6710(r) generally defines ``Takedown Transaction'' as a 
primary market sale transaction sold on the first day of trading of 
a security excluding a Securitized Product other than an Asset-
Backed Security: (i) By a sole underwriter or syndicate manager to a 
syndicate or selling group member at a discount from the published 
or stated list or fixed offering price, or (ii) in the case of a 
primary market sale transaction effected pursuant to Securities Act 
Rule 144A, by an initial purchaser or syndicate manager to a 
syndicate or selling group member at a discount from the published 
or stated fixed offering price.
    \6\ The existing Historic TRACE Data also does not include List 
or Fixed Offering Price or Takedown Transactions.
    \7\ In addition, FINRA intends to establish a fee for the 
Academic Corporate Bond TRACE Data product prior to the effective 
date of the proposed rule change. The fee will be established 
pursuant to a separate rule filing.
---------------------------------------------------------------------------

    While FINRA understands that masked dealer identifiers may be very 
useful to academics in connection with their research activities, we 
also appreciate that firms may be concerned regarding the potential for 
reverse engineering. To address this issue, in addition to uniquely 
masking dealer identities for each academic institution, FINRA proposes 
to take mitigating steps, including to limit transactions included in 
the Academic Corporate Bond TRACE Data product to transactions that are 
at least 36 months old. In addition, FINRA would impose certain 
requirements on subscribers regarding the terms of use of the data. In 
the written agreement with subscribers to Academic Corporate Bond TRACE 
Data, among other things, FINRA will: (1) Explicitly require 
subscribers to agree that they will not attempt to reverse engineer the 
identity of any market participant; (2) prohibit the redistribution of 
data in the Academic Corporate Bond TRACE Data product; (3) require 
users to disclose each intended use of the data (including a 
description of each study being performed and the names of each 
individual who will have access to the data for the study); (4) require 
users to ensure that any data presented in work product be sufficiently 
aggregated so as

[[Page 44360]]

to prevent reverse engineering of any dealer or transaction; and (5) 
require that the data be returned or destroyed if the agreement is 
terminated.
    If the Commission approves the proposed rule change, FINRA will 
announce the effective date of the proposed rule change in a Regulatory 
Notice to be published no later than 90 days following Commission 
approval. The effective date will be no later than 270 days following 
publication of the Regulatory Notice announcing Commission approval.
2. Statutory Basis
    FINRA believes that the proposed rule change is consistent with the 
provisions of Section 15A(b)(6) of the Act,\8\ which requires, among 
other things, that FINRA rules must be designed to prevent fraudulent 
and manipulative acts and practices, to promote just and equitable 
principles of trade, and, in general, to protect investors and the 
public interest.
---------------------------------------------------------------------------

    \8\ 15 U.S.C. 78o-3(b)(6).
---------------------------------------------------------------------------

    Pursuant to the proposal, FINRA will make available to institutions 
of higher education an enhanced historic TRACE data product that will 
include transaction-level data on corporate bonds on a 36-month delayed 
basis with masked identifying information regarding the dealer 
reporting each transaction. Academic Corporate Bond TRACE Data would be 
made available only to institutions of higher education. FINRA believes 
that the additional granularity provided by this new data product will 
enable researchers to track the behavior of individual dealers or group 
of dealers and observe their behaviors, and may facilitate the ability 
of academic researchers to study the impact of various events on 
measures such as intermediation costs, dealer participation and 
liquidity, thereby enhancing understanding of the market for corporate 
bonds and the behavior of its participants. Thus, FINRA believes that 
the proposed rule change is in the public interest and consistent with 
the Act.

B. Self-Regulatory Organization's Statement on Burden on Competition

    FINRA does not believe that the proposed rule change will result in 
any burden on competition that is not necessary or appropriate in 
furtherance of the purposes of the Act.
    FINRA's existing Historic TRACE Data product provides transaction-
level data on an 18-month delayed basis for all transactions that have 
been reported to TRACE in the classes of TRACE-eligible securities that 
currently are disseminated. As detailed above, FINRA is proposing to 
create a new Academic Corporate Bond TRACE Data product, which would be 
made available solely to institutions of higher education with a 36-
month delay and would include masked dealer identifiers associated with 
individual reported transactions, which is not available in the 
existing Historic TRACE Data product.
    The proposed rule change would expand the benefits of FINRA's TRACE 
initiatives by providing additional transparency on corporate bond 
trading for academic research purposes. The analysis that can be 
conducted using masked dealer identifiers associated with individual 
reported transactions could incorporate estimates of anonymized dealer 
positions and hence potentially enhance the ability for researchers to 
analyze and understand dealer networks and liquidity provision in the 
corporate bond market.
    The proposal to create a new Academic Corporate Bond TRACE Data 
product would not impose any additional reporting requirements or costs 
on firms and, as a result, would have no direct impact on firms. 
However, FINRA considered the potential for indirect costs regarding 
possible information leakage due to the inclusion of masked dealer 
identifiers in the data. To investigate whether the dissemination of 
masked dealer identifiers pose a risk for reverse engineering of the 
data to reveal the identity of individual firms, FINRA analyzed 
15,533,134 corporate bond secondary market trades (that occurred 
between February 6, 2012 and February 5, 2016) in 21,164 unique 
corporate bonds that were issued between February 6, 2012 and February 
7, 2015.\9\
---------------------------------------------------------------------------

    \9\ The analysis is conducted from the perspective of the sell-
side in a transaction. Historic TRACE Data and the proposed Academic 
Corporate Bond TRACE Data product do not include List or Fixed 
Offering Price Transactions or Takedown Transactions. Therefore, 
these transactions are excluded from our sample.
---------------------------------------------------------------------------

    The analysis below attempts to answer the question of whether 
primary underwriter information can be reliably linked to the largest 
seller in a given CUSIP and potentially unmask the true identity of the 
firm.\10\
---------------------------------------------------------------------------

    \10\ Primary underwriter information is not a data field in 
TRACE, but is publicly available from various academic and 
commercial databases at the CUSIP level. ``Largest seller'' is 
defined as the Market Participant Identifier (``MPID'') with the 
highest number of transactions over a given number of days.
---------------------------------------------------------------------------

    Figure 1a plots the number of distinct corporate bond CUSIPs that 
are traded within the first n days after issuance (n = 0, 1, 2 . . . 
30) and the percentage of CUSIPs where the largest seller in the 
secondary market also is the primary underwriter for that issue.\11\ 
11,825 distinct corporate bond CUSIPs are traded in the secondary 
market on the day of issuance (n = 0) and the largest seller also is 
the primary underwriter for approximately 6% of those CUSIPs. Within 
the first 30 days of trading (n = 30), the number of CUSIPs traded 
increase to 15,595, and the percentage of CUSIPs where the largest 
seller also is the primary underwriter increases to 11%. Effectively, 
if one assigned the masked dealer identifier associated with the most 
sale transactions in the 30-day window to the primary underwriter, the 
assignment would be correct for about one in ten CUSIPs.
---------------------------------------------------------------------------

    \11\ For example, for n = 2, the measure would determine the 
number of unique CUSIPs where the underwriter had been the largest 
seller of the security for the previous three days.

---------------------------------------------------------------------------

[[Page 44361]]

[GRAPHIC] [TIFF OMITTED] TN07JY16.001

    Figure 1a suggests that largest seller information in a specific 
corporate bond can accurately be linked to the primary underwriter, 
unmasking the identity of the trading firm for approximately 10% of the 
CUSIPs. Alternatively, a researcher could limit its sample to those 
CUSIPs that are traded in the secondary market by a single masked 
dealer identifier within the first n days of trading and assume that 
this seller is the primary underwriter.
    For example, in Figure 1b below, on the day of issuance (n = 0), 
there are 1,835 distinct corporate bonds that are traded by a single 
MPID, of which 222 (approximately 12%) are sold by the primary 
underwriter. If one looked at the first 30 days of secondary market 
trading (n = 30), there would be 2,138 distinct CUSIPs in our sample 
with a single MPID trading the issue and 17% of those MPIDs would be 
associated with the primary underwriter.
[GRAPHIC] [TIFF OMITTED] TN07JY16.002


[[Page 44362]]


    Hence, these finding confirm that primary underwriter information 
alone is not sufficient to discover the true identity of the trading 
firm where the only other information used in the analysis is the 
information to be contained in the Academic Corporate Bond TRACE Data 
product.
    However, FINRA acknowledges the potential for reverse engineering 
of masked dealer identifiers to determine the true identities of 
individual firms, and has taken a number of measures, as discussed 
above, to reduce this risk and mitigate any potential impacts.\12\ 
FINRA believes that the potential additional research that may be 
facilitated by the availability of this new data product will enhance 
understanding of the market for TRACE-eligible securities and trending 
behavior and, therefore, should create a benefit for market 
participants.
---------------------------------------------------------------------------

    \12\ For example, other publicly-available information exists 
that may contribute to the potential for successful reverse 
engineering of dealer identities. One such dataset that can be 
obtained by academics is sold by the National Association of 
Insurance Commissioners (NAIC) and contains detailed information 
about insurance company bond transactions, including the CUSIP of 
the bond traded, the identities of insurance companies and the 
dealers between whom each trade is completed, the date of the 
transaction, the amount traded, and the price of the transaction. 
Please see description of the data in a recent paper by O'Hara et 
al. (2015) at http://papers.ssrn.com/sol3/Papers.cfm?abstract_id=2680480.
---------------------------------------------------------------------------

    FINRA may consider expanding TRACE data product offerings in the 
future to make transaction-level information with masked dealer 
identifiers available to academics for other types of TRACE-eligible 
securities. However, FINRA believes that starting with corporate bonds 
is an appropriate first step because most data requests received from 
academics have related to corporate bond data, and because corporate 
bonds generally are traded by a greater number of dealers and, 
therefore, do not present the same likelihood for accurate reverse 
engineering by academics.

C. Self-Regulatory Organization's Statement on Comments on the Proposed 
Rule Change Received From Members, Participants, or Others

    The proposed rule change was published for comment in Regulatory 
Notice 15-26.\13\ FINRA received four comments in response to the 
Regulatory Notice.\14\ A copy of the Regulatory Notice is attached as 
Exhibit 2a. A list of the commenters and copies of the comment letters 
received in response to the Regulatory Notice are attached as Exhibits 
2b and 2c, respectively.
---------------------------------------------------------------------------

    \13\ Regulatory Notice 15-26 (July 2015).
    \14\ See Letter from Michael Nicholas, Chief Executive Officer, 
Bond Dealers of America, to Marcia E. Asquith, Corporate Secretary, 
FINRA, dated August 24, 2015 (``BDA''), letter from Luis Palacios, 
Director of Research Services, The Wharton School, to Marcia E. 
Asquith, Corporate Secretary, FINRA, dated September 10, 2015 
(``Wharton''), letter from David L. Cohen, Managing Director & 
Associate General Counsel, and Sean Davy, Managing Director, 
Securities Industry and Financial Markets Association, to Marcia E. 
Asquith, Corporate Secretary, FINRA, dated September 11, 2015 
(``SIFMA''), and letter from Carrie Devorah, Founder, The Center for 
Copyrights Integrity, to Marcia E. Asquith, Corporate Secretary, 
FINRA, dated September 14, 2015 (``CCI'').
---------------------------------------------------------------------------

    SIFMA generally supports the proposal but recommends specific 
modifications to further guard against information leakage. 
Specifically, SIFMA's suggestions include that TRACE data should be 
delayed a minimum of four years prior to being included in the academic 
data product; that transactions be grouped by dealer rather than masked 
on an individual basis (excluding information on List or Fixed Offering 
Price Transactions and Takedown Transactions); \15\ and that the 
subscription agreement include restrictions around who at an academic 
institution is authorized to access the data. BDA also raised concerns 
regarding information leakage, and believes that the proposal does not 
adequately balance the risk to dealers with the benefits of academic 
research.
---------------------------------------------------------------------------

    \15\ BDA also notes that the proposal does not state that the 
masked ID used will be changed periodically. To reduce the risk of 
dealer identification, BDA recommends that dealers be grouped by 
size in the Academic TRACE Data.
---------------------------------------------------------------------------

    FINRA has considered concerns regarding information leakage due to 
masked dealer identifiers and the specific comments received. In 
response to comments, FINRA has modified the proposal in two 
significant ways. First, FINRA has modified the proposal to extend the 
data delay period to 36 months rather than the 24-month delay proposed 
in Regulatory Notice 15-26. In addition, FINRA is limiting the data to 
be included in the scope of the current proposal to transactions in 
corporate bonds, including Rule 144A transactions in corporate bonds. 
In Regulatory Notice 15-26, FINRA proposed to include all of the data 
sets currently included in the Historic TRACE Data product.\16\ 
However, because most data requests from academics relate to corporate 
bonds, and because trading may be more concentrated among a smaller 
number of dealers for other types of TRACE-eligible securities, FINRA 
believes it is appropriate to initially adopt the Academic TRACE Data 
product to include transaction information on corporate bonds only, and 
may reconsider the scope of the product in the future. FINRA believes 
that these changes to the academic data product, along with the other 
measures included in the proposal, such as the restricted scope of 
distribution limited to institutions of higher education; the 
safeguards included in the data agreement; and the use of masked 
identifiers, are sufficient in preventing and mitigating any impact 
associated with information leakage.
---------------------------------------------------------------------------

    \16\ Historic TRACE Data is transaction-level data and includes 
the following data sets: The Historic Corporate Bond Data Set, the 
Historic Agency Data Set, the Historic Securitized Product Data Set 
and the Historic Rule 144A Data Set.
---------------------------------------------------------------------------

    BDA and SIFMA also suggest using groupings, rather than masked 
individual dealer IDs, in the academic data product. FINRA has 
considered this alternative and continues to believe that transaction-
level information masked at the individual dealer level is appropriate. 
FINRA believes that groupings will reduce the utility of the data for 
academic researchers and prevent them from accurately undertaking 
studies that analyze dealer behavior, or that need to control for 
dealer-specific factors. However, FINRA notes that masked identifiers 
will be made unique per subscriber. FINRA believes that, while changing 
the masked identifier per data request as suggested by BDA would impede 
research by a single subscriber, assigning unique masked identifiers 
per subscriber may both help guard against coordinated efforts at 
attempting reverse engineering dealer identities as well as assist 
FINRA in identifying the source of conduct that violates the FINRA 
subscription agreement. FINRA may consider amending or discontinuing 
the Academic Corporate Bond TRACE Data product, as currently proposed, 
if future experience shows that anonymized dealer identifier are 
reverse engineered by academics.
    BDA states that prohibiting users from attempting to reverse 
engineer a dealer's identity will not extend to a reader of any study. 
However, FINRA notes that the user agreement also will require that any 
data presented in work product be sufficiently aggregated so as to 
prevent reverse engineering of any dealer or transaction, and believes 
that this measure would protect against reverse engineering by readers 
of published works.
    Wharton supports the proposed academic data product and states that 
the ``[a]cademic community's primary interest in having broker IDs is 
not related to the desire to determine the identities/names of 
underlying brokers, but most importantly to assess the role of brokers 
in bond market liquidity and

[[Page 44363]]

price discovery process.'' \17\ Wharton also states that it has 
received data with masked broker identities for years from data vendors 
and is unaware of any cases where this availability has led to 
successful reverse engineering and public disclosure of broker 
identities.\18\
---------------------------------------------------------------------------

    \17\ See Wharton letter.
    \18\ Wharton provides in its letter examples of vendor data that 
has been available with masked broker IDs. Specifically, Wharton 
states that ``Thomson-Reuters IBES analyst forecast and 
recommendations database is a good example as it has been providing 
masked IDs for both brokerage houses as well as individual analysts 
since the early 80's. Another example is Ancerno (Abel-Noser) high-
frequency database of institutional trades which academic 
researchers have used mainly for the reason that it contains a 
masked institution ID (e.g., Arif, Rephael and Lee, 2015; Choi and 
Sias, 2012).'' See Wharton letter.
---------------------------------------------------------------------------

    BDA and SIFMA raised concerns around the inclusion of primary 
market transaction information (for List or Fixed Offering Price 
Transactions and Takedown Transactions) in Academic TRACE Data. FINRA 
confirms that List or Fixed Offering Price Transactions and Takedown 
Transactions will not be included in the Academic Corporate Bond TRACE 
Data product.\19\
---------------------------------------------------------------------------

    \19\ See supra note 6.
---------------------------------------------------------------------------

    BDA, CCI \20\ and SIFMA raised the issue of information leakage due 
to potential data security breaches. FINRA notes that the data usage 
agreement also will address security measures. For example, FINRA 
intends that the data agreement require the use of commercially 
reasonable measures to protect the data and that users administer 
reasonable security procedures where the data is used, accessed, 
processed, stored or transmitted to ensure that the data remains secure 
from unauthorized access.
---------------------------------------------------------------------------

    \20\ CCI raises issues regarding the security of customer 
information. FINRA notes that the Academic TRACE Data would consist 
of security-focused transaction information, not customer 
information. CCI also raises other issues that are not germane to 
the instant proposal and that, therefore, are not addressed herein.
---------------------------------------------------------------------------

III. Date of Effectiveness of the Proposed Rule Change and Timing for 
Commission Action

    Within 45 days of the date of publication of this notice in the 
Federal Register or within such longer period (i) as the Commission may 
designate up to 90 days of such date if it finds such longer period to 
be appropriate and publishes its reasons for so finding or (ii) as to 
which the self-regulatory organization consents, the Commission will:
    (A) By order approve or disapprove such proposed rule change, or
    (B) institute proceedings to determine whether the proposed rule 
change should be disapproved.

IV. Solicitation of Comments

    Interested persons are invited to submit written data, views, and 
arguments concerning the foregoing, including whether the proposed rule 
change is consistent with the Act. Comments may be submitted by any of 
the following methods:

Electronic Comments

     Use the Commission's Internet comment form (http://www.sec.gov/rules/sro.shtml); or
     Send an email to [email protected]. Please include 
File Number SR-FINRA-2016-024 on the subject line.

Paper Comments

     Send paper comments in triplicate to Brent J. Fields, 
Secretary, Securities and Exchange Commission, 100 F Street NE., 
Washington, DC 20549-1090.

All submissions should refer to File Number SR-FINRA-2016-024. This 
file number should be included on the subject line if email is used. To 
help the Commission process and review your comments more efficiently, 
please use only one method. The Commission will post all comments on 
the Commission's Internet Web site (http://www.sec.gov/rules/sro.shtml). Copies of the submission, all subsequent amendments, all 
written statements with respect to the proposed rule change that are 
filed with the Commission, and all written communications relating to 
the proposed rule change between the Commission and any person, other 
than those that may be withheld from the public in accordance with the 
provisions of 5 U.S.C. 552, will be available for Web site viewing and 
printing in the Commission's Public Reference Room, 100 F Street NE., 
Washington, DC 20549 on official business days between the hours of 
10:00 a.m. and 3:00 p.m. Copies of such filing also will be available 
for inspection and copying at the principal office of FINRA. All 
comments received will be posted without change; the Commission does 
not edit personal identifying information from submissions. You should 
submit only information that you wish to make available publicly. All 
submissions should refer to File Number SR-FINRA-2016-024, and should 
be submitted on or before July 28, 2016.

    For the Commission, by the Division of Trading and Markets, 
pursuant to delegated authority.\21\
---------------------------------------------------------------------------

    \21\ 17 CFR 200.30-3(a)(12).
---------------------------------------------------------------------------

Brent J. Fields,
Secretary.
[FR Doc. 2016-16109 Filed 7-6-16; 8:45 am]
BILLING CODE 8011-01-P