[Federal Register Volume 64, Number 133 (Tuesday, July 13, 1999)]
[Proposed Rules]
[Pages 37708-37713]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 99-17679]


=======================================================================
-----------------------------------------------------------------------

FEDERAL RESERVE SYSTEM

12 CFR Part 229

[Regulation CC; Docket No. R-1031]


Availability of Funds and Collection of Checks

AGENCY: Board of Governors of the Federal Reserve System.

ACTION: Advance notice of proposed rulemaking; withdrawal.

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

SUMMARY: In December 1998, the Board issued an advance notice of 
proposed rulemaking requesting comment on the potential benefits and 
drawbacks of a modification to its Regulation CC that would shorten the 
maximum hold for many nonlocal checks. This modification would shorten 
the availability schedule for nonlocal checks from five to four 
business days, except that a depositary bank could retain a five-day 
availability schedule for subcategories of nonlocal checks for which it 
certifies that it does not receive a sufficient proportion of returned 
checks within four business days. This proposal was one of several 
possible alternatives for defining subcategories of nonlocal checks 
that would be subject to a shortened availability schedule. The Board 
has concluded that return times for nonlocal checks do not support a 
reduced availability schedule for nonlocal checks in the aggregate at 
this time. The Board has also determined that the costs and potential 
risks would outweigh the likely benefits of establishing subcategories 
of nonlocal checks for availability purposes at this time. Therefore, 
the Board has decided not to propose any specific regulatory changes at 
this time to reduce the nonlocal check availability schedule.

FOR FURTHER INFORMATION CONTACT: Jack K. Walton II, Manager, Check 
Payments Section (202/452-2660) or Michele Braun, Project Leader (202/
452-2819), Division of Reserve Bank Operations and Payment Systems; 
Stephanie Martin, Managing Senior Counsel (202-452-3198), Legal 
Division. For the hearing impaired only, contact Diane Jenkins, 
Telecommunications Device for the Deaf (TDD) (202/452-3544).

SUPPLEMENTARY INFORMATION:

Background

    As a result of concerns about some banks' practice of delaying 
funds availability by placing holds on the proceeds of checks deposited 
into customers' transaction accounts, Congress passed the Expedited 
Funds Availability Act (EFAA) in 1987.1 The EFAA specifies 
maximum time limits on the holds that banks may place on funds 
deposited into transaction accounts.
---------------------------------------------------------------------------

    \1\ 12 U.S.C. 4001-4010. As used in this notice and in 
Regulation CC, the term bank includes commercial banks, savings 
institutions, and credit unions. Depositary bank refers to the bank 
of first deposit (see 12 CFR 229.2(e) and (o)).
---------------------------------------------------------------------------

    The EFAA funds availability schedules attempt to balance banks' 
concerns about managing their risk with consumers' concerns about the 
availability of their funds. Congress recognized that banks would be 
exposed to risks if they were required to make funds from a check 
available before they had a reasonable opportunity to learn that the 
check was returned unpaid. To balance depositors' interest in receiving 
prompt access to their funds with banks' ability to manage their risks, 
the EFAA directed the Board to consider improvements to the check 
processing system that would speed the collection and return of 
checks.2 In addition, the EFAA required the Board to reduce 
the statutory funds availability schedules to as short a time as 
possible and equal to the period achievable under the improved check 
clearing system for a depositary bank to reasonably expect to learn of 
the nonpayment of most items for each category of checks.3
---------------------------------------------------------------------------

    \2\ 12 U.S.C. 4008(b) and (c).
    \3\ 12 U.S.C. 4002(d)(1).
---------------------------------------------------------------------------

    The Board's Regulation CC (12 CFR part 229), which implements the 
EFAA, includes maximum availability schedules for funds deposited into 
transaction accounts as well as provisions designed to accelerate the 
check return system. The regulation's availability schedules 
incorporate several provisions in the EFAA where Congress deemed that, 
in certain cases, a longer time was necessary to provide a reasonable 
amount of time for a

[[Page 37709]]

depositary bank to learn of a returned check before having to make the 
funds from that check available for withdrawal. For example, the 
schedules provide for a one-day schedule extension for checks deposited 
in Alaska, Hawaii, Puerto Rico, and the Virgin Islands that are payable 
by a bank located in another state or territory.4 Similarly, 
the EFAA provides a one-day extension for making certain funds 
available for withdrawal by cash or similar means.5
---------------------------------------------------------------------------

    \4\ 12 U.S.C. 4002(d)(2); 12 CFR 229.12(e).
    \5\ Under the ``cash-withdrawal rule,'' the depositary bank may 
extend the holds on local and nonlocal checks for purposes of making 
funds available for withdrawal by cash or similar means. The 
depositary bank may extend the hold until 5:00 p.m. on the normal 
availability day for the first $400 of the deposit and until the 
following business day for the remainder of the deposit. See, 12 
U.S.C. 4002(b)(3); 12 CFR 229.12(d).
---------------------------------------------------------------------------

    The availability schedules in the EFAA and Regulation CC apply to 
three broad categories of checks. Certain ``low-risk'' checks drawn or 
guaranteed by credit-worthy institutions, such as cashier's checks, 
teller's checks, certified checks, and government checks, generally 
must be made available for withdrawal on the next business day 
following the banking day of deposit. Local checks (checks payable by 
banks located in the same check processing region as the depositary 
bank) generally must be made available for withdrawal within two 
business days. Nonlocal checks (checks payable by banks located in 
different check processing regions than the depositary bank) generally 
must be made available for withdrawal within five business 
days.6
---------------------------------------------------------------------------

    \6\ Under Regulation CC's temporary availability schedule, which 
was in effect from September 1, 1988, through August 31, 1990, funds 
deposited by most local checks had to be made available for 
withdrawal within three business days, and nonlocal checks had to be 
made available for withdrawal within seven business days. Other than 
the change from the temporary to the current permanent schedule, the 
EFAA's local and nonlocal check availability schedules have not been 
modified since the EFAA was enacted. During this period, the Federal 
Reserve has consolidated several of its check processing regions, 
listed in Regulation CC's appendix A, which has resulted in some 
checks being reclassified from nonlocal to local. Thus, the 
availability that must be accorded to some deposits has improved.
---------------------------------------------------------------------------

    Pursuant to the EFAA's direction to reduce the statutory schedules 
when banks can reasonably learn of the return of most items in a 
category of checks within a faster time, the Board adopted Appendix B 
of Regulation CC. Appendix B requires depositary banks within certain 
check processing regions to make certain nonlocal checks available 
within a faster time period than that required by the EFAA. Current 
appendix B requires depositary banks in the Utica, Nashville, and 
Kansas City check processing regions to make selected nonlocal checks 
available for withdrawal within three, rather than five, business 
days.7 The Board formulated these reduced schedules in 1988 
through a relatively informal process in which each Federal Reserve 
Bank check processing office estimated which nonlocal checks that were 
deposited in banks in its region could be collected and returned faster 
than the prescribed EFAA maximum hold period. These estimates were 
based on the Reserve Bank's knowledge of geographic proximity between 
certain banks or robust transportation networks and projected 
improvements in return times that would result from requirements 
intended to speed the return of unpaid checks.
---------------------------------------------------------------------------

    \7\ A more extensive set of reduced schedules for nonlocal 
checks was in effect during the temporary schedule period from 
September 1, 1988, to August 31, 1990.
---------------------------------------------------------------------------

1998 Proposal

    After a decade of experience with the post-Regulation CC check 
collection and return system, the Board undertook a study of whether a 
more rigorous approach to reducing nonlocal check schedules would be 
appropriate and what the relative costs and benefits of such an 
approach would be. For guidance on the conditions under which it would 
be appropriate to reduce the availability schedules, the Board looked 
to the 1987 Conference Report on the EFAA.8
---------------------------------------------------------------------------

    \8\ H.R. Conf. Rep. No. 100-261 (1987).
---------------------------------------------------------------------------

    The Conference Report tied availability schedules to banks' ability 
to reasonably expect to learn of the nonpayment of a significant number 
of checks. The Report suggested that if improvements in the check 
clearing system make it possible for two-thirds of the items in a 
category of checks to meet this test in a shorter period of time, then 
the Board must shorten the schedules accordingly.9 The 
Report also recognized that geographic proximity or transportation 
arrangements between check processing regions would permit the Federal 
Reserve to provide shorter times than the general schedule for nonlocal 
checks would require. The Report noted that shorter times would be 
possible for checks transported between such nearby territories as New 
York City and Jericho, Long Island, and for checks transported between 
banks in cities with Federal Reserve check processing offices, such as 
banks in Boston and San Francisco.10
---------------------------------------------------------------------------

    \9\ Id. at 179.
    \10\ Id.
---------------------------------------------------------------------------

    In considering whether nonlocal checks overall met the Conference 
Report's ``two-thirds'' test, the Board drew on data from four surveys 
conducted by the Reserve Banks between 1990 and 1997. In general, the 
more recent surveys showed that over 80 percent of nonlocal unpaid 
checks were returned to the depositary bank within five business days 
after the day of deposit, and 60 to 65 percent of unpaid nonlocal 
checks were returned within four business days. (The surveys are 
discussed in more detail below.)
    In addition to examining nonlocal checks as a single broad 
category, the Board also began investigating whether it would be 
appropriate to define subcategories for various types of nonlocal 
checks and specify maximum availability schedules for these 
subcategories. One means of establishing subcategories of nonlocal 
checks would be for the Board to make subcategory determinations by 
regulation. These determinations would require a trade-off between 
precision in subcategory definition and the practical limitations of 
the data collection needed to support the categorization. Identifying a 
large number of subcategories of nonlocal checks should increase the 
likelihood that the checks are accurately categorized based on when 
they are returned. The greater accuracy afforded by a large number of 
subcategories would lower the risk that funds from a particular check 
would have to be made available by the depositary bank before it would 
normally be returned. Similarly, a higher degree of accuracy would 
increase the probability that customers would receive faster 
availability for those checks for which the depositary bank learns of 
the return before making funds available for withdrawal. Thus, a large 
number of subcategories of nonlocal checks should provide a better 
balance, as sought by Congress, between banks' needs to manage their 
fraud-loss risk and their customers' interests in having as early 
access to their funds as possible.
    The Board explored alternative approaches for defining 
subcategories of nonlocal checks that should receive earlier 
availability. These approaches ranged from categorizing the almost 
2,000 possible pairs of check processing regions to a more aggregated 
approach that would group nonlocal checks into only three categories 
nationwide based on the availability zone (city, RCPC, or country) of 
the paying bank.11 Each

[[Page 37710]]

approach recognized the roles of geographic proximity and 
transportation arrangements in the check clearing and return cycle. It 
was not clear, however, what would be the most reasonable and cost-
effective way to identify those subcategories of nonlocal checks that 
should receive earlier availability. Collecting data to support a valid 
analysis of return cycles for nonlocal checks becomes increasingly 
expensive and, in some cases, impractical as the number of 
subcategories increases.
---------------------------------------------------------------------------

    \11\ In general, nonlocal checks payable by banks located 
closest to Federal Reserve check processing offices are returned 
fastest. Nonlocal checks payable by banks located further away 
require somewhat more time. The locations are organized roughly in 
concentric circles. City checks are payable by banks located 
relatively close to a Federal Reserve office, RCPC checks are 
payable by banks located somewhat further from a Federal Reserve 
office, and country checks are payable by banks even more 
geographically remote. Only eight of forty-four check processing 
regions have country availability zones.
---------------------------------------------------------------------------

    The Board requested comment on an alternative approach for 
establishing nonlocal check subcategories. Specifically, the Board 
considered a self-certification system under which the general nonlocal 
check availability schedule would be reduced to four business days, and 
depositary banks could conduct their own surveys, if they believed it 
would be cost-effective to do so, to determine the subcategories of 
nonlocal checks that would be subject to five-day availability 
schedules. This approach would match the bank's actual return 
experience with availability schedules more precisely than any approach 
that relies on data that the Reserve Banks could collect. Permitting a 
bank to certify that it qualifies to use five-day availability 
schedules for some subcategories of nonlocal checks could give it the 
flexibility to weigh (1) the costs of collecting data with which to 
certify that it should be permitted to hold certain subcategories of 
nonlocal checks for five days, (2) the fraud risk associated with its 
hold policy, and (3) the customer benefits of that policy.
    The Board noted, however, the difficulty of obtaining a sufficient 
sample to validate several of the available options for defining such 
subcategories of nonlocal checks. If a bank determined that the 
administrative cost associated with demonstrating that certain 
subcategories of nonlocal checks should be subject to five-day 
availability and the resulting increased complexity of its availability 
schedules outweighs the incremental fraud protection, then it could 
adopt a four-day or shorter schedule for all of its nonlocal check 
deposits.

    The Board requested comment on this self-certification approach in 
an advance notice of proposed rule-making, issued in December 
1998.12 The notice noted that the Board was also considering 
other methods for defining categories of nonlocal checks that might 
reasonably meet the congressional mandate.
---------------------------------------------------------------------------

    \12\ 63 FR 69027, December 15, 1998.
---------------------------------------------------------------------------

Summary of Comments

General Comments

    The Board received one hundred twenty-five comment letters in 
response to the December 1998 advance notice of proposed rulemaking. 
The following table shows the number of comments by the category of 
commenter:

------------------------------------------------------------------------
                                                               Number of
                    Category of commenter                      responses
------------------------------------------------------------------------
Banks and bank holding companies............................          99
Clearinghouses and associations representing banks..........          21
Check processors............................................           1
Federal Reserve Banks.......................................           4
                                                             -----------
    Total...................................................         125
------------------------------------------------------------------------

    One hundred sixteen commenters opposed shortening nonlocal hold 
periods. One commenter stated that it would support any reduction in 
the hold period as a move to improve the image of banks in general. The 
eight other commenters did not address the length of the nonlocal 
availability schedule, but did comment on specific questions posed by 
the Board about implementing the proposed self-certification process.
    Eighty-two commenters cited increased risk of fraud loss as their 
reason for opposing any proposal to shorten nonlocal hold periods. Many 
commenters also stated that banks frequently maintain availability 
policies that make funds available sooner than required by Regulation 
CC.13 These commenters stated that although their banks 
generally make funds available earlier than required, on a case-by-case 
basis they withhold funds for the maximum permissible period. Several 
of these commenters further stated that the checks for which they use 
case-by-case holds are the ones with greatest risk for loss, so that 
shortening the hold period by even one day could increase the risk of 
loss dramatically. Other banks stated that they use the maximum 
permissible hold period and that shortening the permitted hold period 
would expose them to a potentially significant increase in check fraud 
losses.
---------------------------------------------------------------------------

    \13\ These statements are consistent with findings reported in 
studies conducted by the Board and the American Bankers Association. 
In these studies, 70 and 86 percent of responding banks, 
respectively, reported that they do not hold nonlocal checks for the 
full period permitted under Regulation CC. Board of Governors of the 
Federal Reserve System, Report to the Congress on Funds Availability 
Schedules and Check Fraud at Depository Institutions (Board of 
Governors, 1996), p. 36, and American Bankers Association, ABA 1998 
Check Fraud Survey Report, (1998), p. 19.
---------------------------------------------------------------------------

    Eighteen commenters also stated that shortening nonlocal holds by 
one day would provide little benefit to consumers, either because banks 
already make most funds available more quickly than required or because 
banks that use four-or five-day holds may release funds early if the 
customer so requests and the banks can verify payment by contacting the 
paying bank. Thus, these commenters argued, shortening the nonlocal 
hold period by one day would not benefit many depositors.

Reasonable Time To Learn of Nonpayment

    As noted above, the Board included in its notice data regarding 
return times for nonlocal checks from four surveys. In 1996, the 
Board's comprehensive survey of check-fraud losses at banks asked 
respondents to indicate the proportion of returned checks that they 
typically received on each business day following the initial deposit 
of a check (1996 bank survey). In conjunction with that check-fraud 
study, Federal Reserve staff also collected detailed data from a sample 
of checks processed during one week through the Federal Reserve Banks 
(1996 Reserve Bank survey).14 In 1997, Federal Reserve staff 
repeated the Reserve Bank survey for six weeks and thereby increased 
the number of nonlocal returned checks sampled compared with the prior 
survey (1997 Reserve Bank survey).15 The results of the 1997 
survey were generally consistent with those of the 1996 survey. For 
historical comparison, the Board also reviewed a survey of checks 
returned through the Reserve Banks conducted shortly after the 
implementation of Regulation CC (1990 Reserve Bank 
survey).16 The table below summarizes the average nonlocal 
return

[[Page 37711]]

cycles observed in the 1990, 1996, and 1997 surveys.

    \14\ Board of Governors of the Federal Reserve System, Report to 
the Congress on Funds Availability Schedules and Check Fraud at 
Depository Institutions (Board of Governors, 1996).
    \15\ The 1997 survey was designed to provide a sufficient number 
of checks to estimate the proportion of nonlocal checks returned 
within four and five days nationwide. The sample was not intended to 
provide statistically valid results between each possible pairing of 
check processing regions throughout the country. (63 FR 69027, 
December 15, 1998).
    \16\ Board of Governors of the Federal Reserve System, The 1990 
Report to Congress Under the Expedited Funds Availability Act (Board 
of Governors, 1990).

                Cumulative Percentage of Nonlocal Checks Returned Within Number of Business Days
----------------------------------------------------------------------------------------------------------------
                                                   1997          1996                       1990       Percent
                                               Reserve Bank  Reserve Bank   1996  bank    Reserve    improvement
                                                survey \1\    survey \1\      survey    Bank survey    1990-97
----------------------------------------------------------------------------------------------------------------
3 business days..............................          27.8          33.3         32.0         21.0         32.4
4 business days..............................          59.9          64.1         64.9         47.0         27.5
5 business days..............................          82.8          83.3         84.3         73.0         13.4
Number of nonlocal checks sampled............      31,646         5,707        \2\ 773           NA           NA
----------------------------------------------------------------------------------------------------------------
\1\ Excludes outlier observations defined as nonlocal checks that exceed 15 business days. For example, the 1997
  survey data exclude 1.6 percent of nonlocal checks sampled.
\2\ Reflects the number of commercial banks, savings institutions, and credit unions sampled.
Source: Board of Governors of the Federal Reserve System. See text notes 17, 18, and 19 for sources of data.

    Twenty-nine commenters stated that, in the aggregate, the return 
periods reported in the Board's notice indicated that banks did not yet 
receive two-thirds of their returns within four business days. Fifteen 
commenters stated that the average return cycle was shorter in 1996 
than in 1997 and requested that the Board defer shortening the maximum 
hold periods until the data showed that return times for nonlocal 
checks, in the aggregate, clearly exceeded at least the two-thirds 
threshold.
    The way in which the Board presented the data in the 1998 notice 
suggested that a nonlocal check that was returned to the depositary 
bank on the fifth business day after the day of deposit afforded the 
depositary bank a reasonable time in which to learn that the check was 
returned before making funds available for withdrawal on that day. 
Similarly, the notice could be read as assuming that if a nonlocal 
check is returned on the fourth business day after deposit, it may be 
appropriate to reduce the availability schedule applicable to that 
check to four business days.
    Several commenters, including the American Bankers Association, 
argued that all banks need the ability to hold funds for one day beyond 
the day on which they receive returned checks. These commenters noted 
that Regulation CC requires that funds be made available at the start 
of the business day on which the depositor must have access to funds 
pursuant to the schedules, but unpaid checks typically are not 
delivered until mid-day. As a result, the depositor might be permitted 
to withdraw cash several hours before the bank knows that the check was 
being returned.17 Therefore, they argued, banks should be 
able to hold checks for one day beyond the day they can ``reasonably 
expect to learn of the nonpayment of most items.'' Under this theory, 
nonlocal schedules should not be reduced to four business days unless 
two-thirds of nonlocal checks can be returned to the depositary bank by 
the third business day after the banking day of deposit. The American 
Bankers Association further argued that the extra day permitted for 
cash withdrawals does not ameliorate this problem because the attendant 
requirement that cash and check withdrawals be tracked separately is 
not operationally feasible for most banks.
---------------------------------------------------------------------------

    \17\ 12 CFR 229.19(b). Specifically, funds must be made 
available for withdrawal by the later of 9:00 a.m. (local time of 
the depositary bank) or the time the depositary bank's teller 
facilities (including ATMs) are available for customer-account 
withdrawals.
---------------------------------------------------------------------------

    In addition, seven commenters argued that the two-thirds threshold 
suggested by the legislative history was inadequate and that receiving 
as many as one-third of returned checks back after the maximum 
permissible hold period would expose banks to more risk than they 
considered acceptable. One commenter cited the statutory language that 
requires shorter schedules where the depositary bank can reasonably 
expect to learn of the nonpayment of ``most'' items and argued that, as 
interpreted by several courts in other contexts, ``most'' means an 
amount more significant that two-thirds, and the Conference Report 
language should not be considered controlling.
    Twenty-four commenters provided data on their return experiences. 
Some commenters provided explicit surveys of their return items, while 
others asserted that some items took six or more days to be returned 
and, therefore, they opposed reducing the permissible hold period. 
These banks also noted the difficulty they had collecting 
representative data and explained that this was an expensive, labor-
intensive, manual process.
    Several smaller institutions pointed out that they use one or more 
intermediaries to process and collect checks, which tends to add at 
least one day to the collection process, and that they would be 
particularly disadvantaged by shortened hold periods for nonlocal 
checks. The National Association of Federal Credit Unions stated that 
shorter nonlocal hold periods would have a disproportionately negative 
effect on credit unions because credit unions are less likely than 
commercial banks to receive returned checks within four business days.

Subcategories of Nonlocal Checks

    In general, the commenters stated that there were difficulties 
associated with an availability policy that includes subcategories of 
checks. Seven commenters stated that creating subcategories of checks 
within the categories of next-day, local, and nonlocal checks would 
greatly increase the complexity of the regulation, the cost of 
implementation, and the difficulty of adequately disclosing banks' 
availability policies to consumers. These commenters also stated that 
they could not reliably collect data on check return patterns beyond 
the existing categories of checks.
    Some of these commenters further stated that the EFAA established 
the check categories and does not require the Board to further 
subdivide those categories. The American Bankers Association stated 
that the most obvious meaning of ``category of checks'' in the EFAA is 
provided by the statute (that is, next-day, local, and nonlocal 
checks), on which the statutory funds availability schedules are based.
    Many commenters stated that it would be important to disclose 
availability policies to depositors thoroughly. The American Bankers 
Association stated, however, that creating a more complex, changeable 
system would confuse consumers when one of the main purposes of the 
statute was to inform consumers. Most commenters expressed similar 
views,

[[Page 37712]]

stating that existing availability schedules were complicated to 
explain to depositors and that policies that differentiate among 
categories of nonlocal checks would be more confusing.
    Forty-two respondents stated that administering holds for 
subcategories, implementing the proposed self-certification process, 
making any change to availability schedules, and training clerical 
staff on current hold and disclosure policies would be excessively 
costly. Fifty-two commenters stated that Regulation CC was already very 
complex and that training staff to properly administer the regulation 
presented a continuing problem. These commenters argued against any 
changes that might increase the cost or complexity of implementing or 
explaining the Regulation's provisions.
    Thirty-four commenters commented on the proposed self-certification 
procedure. Generally, commenters indicated that banks would be unlikely 
to use the self-certification option because of its complexity and 
implementation cost, and that they would simply use shorter hold 
periods or shorter case-by-case holds despite the potentially increased 
risk.

Conclusions

Reasonable Time to Learn of Nonpayment.

    Although the EFAA requires the Board to reduce availability 
schedules based on improvements in the check collection process, the 
EFAA states that such reductions should be made when depositary banks 
can reasonably expect to learn of the nonpayment of most items subject 
to the reduced schedules. Other provisions of the EFAA, such as the 
extended schedules allowed for cash withdrawal purposes and for certain 
checks deposited outside the continental U.S., indicate that Congress 
meant to protect depositary banks from undue risk that might accompany 
the EFAA's maximum availability schedules.18 Thus, the EFAA 
attempts to balance the interests of depositors in receiving prompt 
availability of funds against the risks to depositary banks of making 
funds available before learning that checks have not been paid.
---------------------------------------------------------------------------

    \18\ The Act also permits depositary banks to extend holds under 
certain exception situations, such as when a deposit is over $5,000 
or when a bank has reasonable cause to doubt the collectibility of a 
check. These exception holds are not based on the amount of time it 
takes to collect and return the particular check, indicating that 
Congress may have presumed that, in the normal course of events, a 
significant number of checks would not be returned in time to 
provide the depositary bank with protection within the regular 
availability schedules, making these exception holds necessary.
---------------------------------------------------------------------------

    Although the discussion of the survey data in the Board's December 
1998 notice was based on the premise that a depositary bank should be 
able to make funds available from a check on the day it would normally 
receive the return of that check, the Board has reconsidered that 
reasoning. The Board believes that the depositary bank can reasonably 
expect to learn of the nonpayment of most items only if it learns of 
the returned checks in time to take action before funds are required to 
be available for withdrawal. Generally, banks receive returned checks 
around midday. Banks require time to process the unpaid checks and post 
entries to depositors' accounts. Under the EFAA, $400 in cash must be 
made available not later than 5:00 p.m. on the day that funds are to be 
made available for other purposes. While banks are permitted to delay 
by one more day the withdrawal of additional amounts by cash or similar 
means, it is costly and perhaps operationally not feasible for banks to 
treat cash and check withdrawals differently. Accordingly, banks appear 
to make funds available for withdrawal by cash at the opening of 
business on the same day on which they make funds available for other 
purposes. If the schedule is shortened so that the depositary bank is 
required to make funds available at the opening of business on the day 
that it receives the returned check, it may need to make funds 
available several hours prior to receipt of the check and before it is 
able to post the returned check to the depositor's account. 
Accordingly, the Board has reconsidered the time frame within which the 
``two-thirds'' test is relevant. The Board believes that, before 
availability schedules are reduced for a category (or subcategory) of 
checks, a depositary bank should be able to learn of the return of most 
checks in that category in time to prevent depositors from withdrawing 
funds from the checks. The data from the surveys shows that, on an 
aggregate basis for nonlocal checks, the proportion of nonlocal checks 
returned to banks within three business days was well below the two-
thirds envisioned by Congress. In addition, although the proportion of 
nonlocal checks returned within four business days after deposit was 
close to two-thirds, it remained slightly below that threshold. The 
Board has concluded, therefore, that it would not be appropriate to 
reduce the general availability schedule for nonlocal checks five to 
four days at this time. This determination does not foreclose the 
possibility that improvements in the check return system or in posting 
systems might lead to a shortening of the general availability schedule 
in the future.

Subcategories of Nonlocal Checks

    After reviewing the comments on the difficulties of implementing 
differing availability schedules for subcategories of nonlocal checks, 
the Board has determined that the costs and difficulties of 
establishing such subcategories, in addition to those already specified 
in appendix B of Regulation CC, would outweigh the likely benefits. As 
stated by the commenters, creation of subcategories of nonlocal checks 
would increase depositary bank costs significantly, particularly in the 
area of employee training and operations changes. These costs would 
increase regardless of whether the subcategories were established by 
regulation or by a self-certification process, although depositary 
banks would bear additional costs under the latter 
process.19 In addition, the commenters expressed concern 
that increasing the complexity of the availability schedules would also 
increase confusion for depositors and bank employees. The Board also 
notes that most banks do not appear to impose the maximum permissible 
hold periods, thus reducing the apparent potential benefits to 
depositors of reducing the nonlocal hold period.
---------------------------------------------------------------------------

    \19\ Moreover, as some commenters stated, it is not clear that 
the EFAA requires reduced schedules for subcategories of checks. 
Although the EFAA does not define ``categories of checks,'' some 
commenters argued that the Board should rely on the categories of 
checks delineated in the EFAA and that the EFAA does not direct the 
Board to define additional check categories. The 1987 Conference 
Report, however, provided an example using subcategories of nonlocal 
checks.
---------------------------------------------------------------------------

    Furthermore, it would be difficult to determine specific categories 
of nonlocal checks that should be subject to a shortened availability 
schedule. While Reserve Bank estimates based on geographic proximity or 
robust transportation networks formed the basis for including specific 
categories of checks in appendix B, the basis for determining 
additional categories of nonlocal checks subject to shortened 
availability schedules in a comprehensive way would be more complex. 
Optimally, statistical sampling of data from returned checks would 
provide a valid estimate of the number of returned checks with 
reasonable confidence intervals around the estimates. Collecting such 
an optimum sample of returned nonlocal checks, however, is not simple. 
First, because most checks are local, the sample size

[[Page 37713]]

would have to be very large to obtain a sufficient number of nonlocal 
checks. Second, as the number of subcategories of nonlocal checks 
increases, the number of checks that need to be sampled increases as 
well. It may be virtually impossible to collect a sufficient number of 
checks between certain regions of the country owing to the limited 
number of checks returned between them. Further, collecting the data is 
a costly and time-consuming process. The information on returned checks 
needed for a survey must be collected manually from the back of checks 
and is often overprinted, lightly printed, and otherwise difficult to 
read. Data collection is further complicated by processing schedules. 
Returned checks become available to be sampled during peak processing 
periods in the middle of the night when bank staff have very limited 
time to collect the required data without slowing the return of those 
checks to the depositary bank.
    Accordingly, the Board has decided not to establish different 
maximum availability schedules for additional subcategories of nonlocal 
checks. Although the Board has decided not to propose any specific 
regulatory changes at this time to reduce the nonlocal check 
availability schedule, the Board will continue to monitor the time 
periods needed to return unpaid nonlocal checks and may consider 
further action if return times improve significantly.

    By order of the Board of Governors of the Federal Reserve 
System, July 7, 1999.
Jennifer J. Johnson,
Secretary of the Board.
[FR Doc. 99-17679 Filed 7-12-99; 8:45 am]
BILLING CODE 6210-01-P