Discussion of Comments Received

FRY14AQM_20120628_Discussion of Comments.pdf

Capital Assessment and Stress Testing

Discussion of Comments Received

OMB: 7100-0341

Document [pdf]
Download: pdf | pdf
Discussion of Federal Reserve’s Responses to
FR Y-14 A/Q/M Public Comment Letters
Overview
The FR Y-14A annually collects large BHCs’ quantitative projections of balance sheet,
income, losses, and capital across a range of macroeconomic scenarios and qualitative
information on methodologies used to develop internal projections of capital across scenarios.
The FR Y-14Q collects granular data on BHCs’ various asset classes and PPNR for the reporting
period, which are used to support supervisory stress test models and for continuous monitoring
efforts, on a quarterly basis. The new FR Y-14M will collect one loan-level collection for
Domestic First Lien Closed-End 1-4 Family Residential Mortgage data, one loan-level collection
for Domestic Home Equity Residential Mortgage data, one account- and portfolio-level
collection for Domestic Credit Card data, and one collection for Address Matching data to
supplement the two mortgage collections.
On February 22, 2012, the Federal Reserve published a notice in the Federal Register
(77 FR 10525) requesting public comment for 60 days to revise, without extension, the FR Y-14
information collection. The comment period expired on April 23, 2012. The Board received six
comment letters from four BHCs and four trade associations.1 All substantive comments are
summarized and addressed below.
The FR Y-14A/Q/M revisions proposed in the Federal Reserve’s February 2012 Federal
Register notice, effective June 30, 2012, included (1) implementing a new monthly schedule, the
FR Y-14M, which would collect data previously collected on several quarterly Retail Risk
portfolio-level worksheets (two new loan-level collections and one new loan- and portfolio-level
collection), and collecting detailed address matching data for the two loan-level collections; (2)
revising the quarterly Wholesale Risk schedule (corporate loan data collection) by adding data
items that would allow the Federal Reserve to derive an independent probability of default,
expanding the scope of loans included in the collection by moving loans from the CRE data
collection to the corporate loan data collection, clarifying definitions of existing data items, and
requesting additional detail about collateral securing a facility; (3) revising the quarterly
Wholesale Risk schedule (CRE collection) by moving loans to the corporate loan data collection,
adding a non-accrual data item, and modifying the loan status data item to include the number of
days past due; (4) implementing a new quarterly Operational Risk schedule to gather data that
would support supervisory stress test models to forecast the BHCs’ operational loss levels under
various macroeconomic conditions; and (5) expanding the respondent panel (for the FR Y-14
A/Q/M) to include large banking organizations that meet an asset threshold of $50 billion or
more in total consolidated assets (large BHCs), as defined by the Capital Plan Rule (12 CFR
225.8).
Summary of Comments
The Federal Reserve received comments from the industry by letter, email, and orally
through industry outreach calls. Most of the comments received requested clarification of the
                                                            
1

Three trade associations submitted a joint comment letter.

1 
 

instructions for the information to be reported, or were technical in nature. Response to these
comments will be addressed in the final FR Y-14 reporting instructions. The following is a
detailed discussion of aspects of the proposed FR Y-14 collection for which the Federal Reserve
received one or more substantive comments and an evaluation of, and response to, the comments
received.
A. General
In general, three trade associations expressed support for the capital adequacy and risk
review process proposed by the Federal Reserve and appreciated the opportunity for public
comment on the proposed schedules. However, the commenters noted their concerns about the
scope of the information collection increasing since the release of the Federal Reserve’s 2012
CCAR.
The trade associations commented that the time schedule proposed for submitting the
data imposes burden on the BHCs’ resources that are already fully dedicated to submitting other
regulatory reports during the same time period. The commenters requested that the FR Y-14Q
be submitted 60 days after quarter end, as opposed to 40 days (for the first, second, and third
quarters) and 45 days (for the fourth quarter) after the quarter end. The Federal Reserve is
cognizant of the extra resources BHCs must devote to prepare data submissions for the first
quarter data are being implemented. However, the Federal Reserve expects that subsequent
submissions would require fewer resources given that BHCs would likely automate much of the
process. Further, BHCs already aggregate the more granular data reported in the FR Y-14Q and
the FR Y-14M schedules for their Consolidated Financial Statements for Bank Holding
Companies (FR Y-9C; OMB No. 7100-0128) reporting. Therefore, the Federal Reserve will
allow BHCs to request approval to file a late submission following major revisions to data
schedules.
One commenter suggested that the Federal Reserve provide an acquiring bank with one
year to incorporate the assets into its systems before requiring data to be submitted for the FR Y14. The Federal Reserve has reached out to several BHCs to inquire about the challenges
associated with providing data on merged portfolios. Through those inquiries, the Federal
Reserve has learned that the amount of time required to integrate the risk information from a new
portfolio with the risk information from an existing portfolio varies greatly depending on the
similarities of the risk information in the two portfolios. The Federal Reserve will consider on a
case-by-case basis requests from BHCs to file a delayed submission for newly acquired data
(integrated with the legacy portfolio) following an acquisition.
Several commenters suggested that BHCs with assets greater than $50 billion (covered
companies) that have not previously participated in CCAR should receive extra time to make
their first submissions. While the Federal Reserve believes that the proposed data items reflected
important elements of any sound risk management system for large BHCs, the Federal Reserve
agrees that BHCs not previously subject to CCAR would benefit from additional time to build
their internal reporting systems. As a result, the Federal Reserve finalized the proposed
schedules only for the 19 BHCs that have previously participated CCAR at this time. The

2 
 

Federal Reserve may publish a separate proposal to address data requirements for the remaining
covered companies in the future.
B. FR Y-14M: Domestic First Lien Closed-End 1-4 Family Residential Mortgage (first lien),
Domestic Home Equity Residential Mortgage (home equity), Domestic Credit Cards, and
Address Matching schedules
The Federal Reserve proposed that respondents would submit loan-level monthly data
schedules for material first lien, home equity, and credit card portfolios. Several commenters
noted that the monthly proposed schedules contain a number of substantive data items that are
duplicative of data collected by other agencies, namely the Office of the Comptroller of the
Currency (OCC). The Federal Reserve has reviewed the OCC’s data collections and determined
that the collections differ in scope and substance. For example, the universe of loans proposed to
be collected in the FR Y-14M includes all loans owned or serviced by the BHCs, while the OCC
collects all loans serviced by national banks. Furthermore, there are definitional differences
between the data items collected by the OCC and those proposed by the Federal Reserve.
The agencies’ staff worked together to ensure that data items included in the final
schedules (or data items specifically suggested by the industry) aligned to the extent possible
with the OCC data collections. In response to the comments, the Federal Reserve will broaden
the scope of the monthly Domestic Credit Card schedule to include all credit cards, not just those
reported on the FR Y-9C in Schedule HC-C, Loans and Lease Financing Receivables, data items
4.a, commercial and industrial loans to U.S. addressees, and 6.a, credit cards, as originally
proposed. In particular, this schedule will now include credit card and charge card loans
included in data item 6.d, other consumer loans. In addition, the Federal Reserve is continuing to
work with the OCC and anticipates engaging other agencies in discussions to determine how best
to collect the data so as to satisfy each agency’s mandate while imposing minimal burden on the
industry. If any future revisions are required based on this more extensive review, the agencies
would publish information collection notices in the Federal Register to seek comment on any
future revisions to their collections.
One commenter requested clarity on how to report “Income Source at Origination” for
purchased loans where this information is not available. The Federal Reserve will revise the four
choices of answers to read: Individual, at origination; Household, at origination; Individual, at
acquisition; and Household, at acquisition.
Several commenters noted that the proposed first lien and home equity schedules were
not clear regarding whether only loans owned by the respondent or loans serviced by the
respondent would be reported. The Federal Reserve will collect data on both owned and
serviced portfolios. In making this making this final determination, the Federal Reserve
carefully weighed the benefits of capturing servicing data versus the burden imposed on the
respondents. During an industry outreach call, one participant indicated that the burden
associated with reporting the servicing portfolio loans would be minimal because those data are
already aggregated. The collection of the servicing portfolio loans will allow the Federal
Reserve to investigate other potential risk factors, such as factors associated from multiple loans
being collateralized by the same property.
3 
 

One commenter suggested that the data for BHC-owned loans serviced by third parties be
collected at the portfolio-level with an extended due date. The Federal Reserve will implement
the revisions as proposed, as many of the data items in the schedules will be used for ongoing
internal risk management. The Federal Reserve expects that most respondents can readily access
the information for BHC-owned loans. However, the Federal Reserve recognizes that it may be
more difficult for the BHCs to initially submit the data as requested for loans serviced by other
entities. Over time the Federal Reserve expects that BHCs will maintain the data as requested by
the Federal Reserve in their internal systems. The Federal Reserve will consider on a case-bycase basis request to file a delayed submission for portfolio loans serviced by others.
Two commenters noted that the proposed first lien and home equity schedules were
unclear as to whether real estate owned after foreclosure but before the disposition of the
property would need to be reported. The commenters suggested that many of the data items in
the first lien and home equity schedules are related to loans, and thus would not be available for
real estate owned. In light of the comments, the Federal Reserve will require the reporting of
loans through the month they were liquidated or transferred to another servicer. The Federal
Reserve believes it is important to capture data for the final month in order to understand what
happened to loans that cease to be a part of the data collection (to help determine, for example, if
the loan dropped out of the data collection due to a prepayment, liquidation, or servicer transfer).
Several commenters noted inconsistencies between the definitions for similar data items
in the first lien and home equity schedules. The commenters requested standardized data item
definitions across the two collections. The schedules were reviewed and several definitions were
identified that could be standardized (for example, Product Type, Income Documentation,
Property Type, Refreshed Property Valuation Method, Foreclosure Status, and Occupancy);
therefore, the Federal Reserve will implement these standardized definitions.
Several commenters noted that some data items in the first lien and home equity
schedules would be difficult to provide. In some cases, the commenters suggested that the data
items could be provided at a portfolio-level but not at a loan-level, while in other cases the
commenters noted that some data items are not maintained for certain loan types. The Federal
Reserve carefully considered these comments, and reviewed the reporting instructions to identify
data items that would be difficult to report at the loan-level. Based on the review, the Federal
Reserve will (1) modify these loan-level collections to include the reporting of portfolio-level
data for certain items such as purchase impairments which typically are unavailable at the loanlevel and (2) require that only certain data items be reported for loans serviced for others or loans
obtained through mergers or acquisitions.
Several commenters suggested adding a number of data items to the first lien and home
equity schedules. The Federal Reserve reviewed the data items proposed by the commenters,
and carefully weighed the value of having the items versus the additional burden on respondents.
While, in most cases, the Federal Reserve does not have sufficient evidence to justify the
inclusion of the data items, there were a few instances in which it was determined that the value
of having the data items outweighed the burden imposed on respondents.

4 
 

One commenter suggested that the address matching schedule be appended to the first
lien and home equity schedules to reduce the number of schedules. The Federal Reserve does
not believe incorporating this suggestion would reduce the overall burden on respondents and
will maintain the structure of schedules as proposed.
C. FR Y-14Q Wholesale Schedule (Corporate Loan and CRE collections):
One commenter suggested that respondents should not be required to provide a
guarantor’s tax ID number. The commenter noted that it is not relevant for the information
collection, and expressed concerns about the sensitivity of the information. The Federal Reserve
notes that while all information reported in the FR Y-14 is confidential supervisory information,
extra care should be taken to protect the privacy of individuals; therefore the Federal Reserve
will implement the instructions as proposed, which indicate that if the guarantor is a natural
person, respondents should not report the tax ID number.
One commenter noted that the Federal Reserve should not model probability of default
(PD) on corporate loans. Specifically, the commenter stated that internal bank ratings produce a
better estimate of PD than would be done based on the reported data. The commenter also noted
that the Federal Reserve should use reporter generated PDs when evaluating capital submissions.
The Federal Reserve notes that the process of mapping the internal ratings to a common scale
introduces a potential for inconsistency due to possible differences in the way BHCs’ assign
internal ratings. Collecting the data necessary to calculate the PD on corporate loans would
enhance the comparability of estimated PDs across BHCs and promote greater consistency in
supervisory stress test estimates. Therefore, the Federal Reserve will implement the revision as
proposed.
Several commenters suggested that a number of the data items requested in the corporate
loan collection may not be readily available, and some of the data items may not be available for
certain types of loans. The Federal Reserve spoke with several BHCs to determine whether the
proposed data items would be available in the BHCs’ credit risk systems. These discussions
revealed that all of these BHCs use financial spreading software and while the systems varied,
they capture all or most of the proposed data items. Though it is possible that some BHCs may
not use financial spreading software, which would result in the increased burden, the Federal
Reserve believes the benefit of the data outweighs the increased burden and recommends
implementing the revision as proposed. The Federal Reserve also recognizes that for certain
categories of borrowers the financial information may not be readily available. Therefore, the
Federal Reserve will revise the instructions to exclude certain populations of loans on which the
Obligor Financial Information may not be available.
One commenter noted that, in the corporate loan collection, the Federal Reserve should
define obligor for purposes of the Obligor Financial Information section to be the "risk unit
underwritten by the BHC for the purpose of approving the loan." The commenter further noted
that, in cases in which there are multiple obligors, it is possible that a different obligor in the
credit agreement could drive default risk. The Federal Reserve agrees with the comment. These
situations may be rare, but could lead to an overestimation of credit risk. The current structure of
the collection is based on the legal borrowing entity. The Federal Reserve believes that, for
5 
 

consistency, it is important to retain the link between the legal borrowing entity and the other
data items. However, the Federal Reserve will allow a respondent to submit the financial
information in the Obligor Financial Information section for the entity that represents the
primary source of repayment (i.e. not a guarantor). The Federal Reserve will add three
identification data items to capture information about the entity underwritten by the respondent
in cases in which that entity is different from the obligor.
D. FR Y-14Q Operational Risk schedule
A trade association provided a general comment expressing support for collecting
quarterly data on operational risk, but had concerns about implementing the proposed new
schedule on operational risk.
Several commenters requested that the Federal Reserve not collect data on legal reserves
for pending and probable litigation claims on the proposed Operational Risk schedule. The
commenters had concerns that the Federal Reserve may not be able to guarantee the
confidentiality of the information in all cases; the data could become discoverable in third-party
litigation; and should the information make its way into the public domain, it could significantly
jeopardize the BHC’s position in litigation. Based on the comments received and subsequent
discussions with commenters, the Federal Reserve’s preliminary view is that these concerns are
justified. Accordingly, the Federal Reserve will not require BHCs to submit these data as part of
the June collection. However, the Federal Reserve will implement the remainder of Operational
Risk schedule as proposed.
Furthermore, Federal Reserve will re-open the comment period on legal reserves to
facilitate feedback on methods that would enable the Federal Reserve to collect legal reserves
data in a fashion that would protect the confidentiality of the information.
One commenter suggested requiring the submission of data on loss events of $20,000 or
more. Supervisory stress testing of operational risk is focused on estimating expected loss under
certain macroeconomic scenarios. As such, the reporting of operational losses at and above the
BHC-established collection threshold, versus a fixed threshold, is necessary as the full
distribution of captured losses is important to expected loss estimation. The Federal Reserve will
implement the revision as proposed.
One commenter suggested providing flexibility with how each loss event is assigned to a
business line(s), event type(s), or accounting date(s). The Federal Reserve agrees with this
comment and will revise the reporting instructions. The instructions will no longer require single
loss events that affect multiple business lines to be “aggregated” into one business line (the
business line which incurred the largest loss amount). Instead, the revised instructions will
enable institutions to report multiple records or transactions for the same loss event and to
respectively assign those records or transactions to the appropriate business line.
One commenter also noted that providing a single “Accounting Date” per loss event
would be confusing, because BHCs do not typically have a single accounting date for every loss
event. As described above, the reporting instructions have been revised to require the quarterly
6 
 

submission of the BHCs' complete history of operational losses. This will enable BHCs to
submit multiple transactions or records for the same loss event, as long as multiple transactions
or records contain the same reference number for the respective event.
Several commenters requested that the data collection be revised to allow BHCs’ to
submit their entire operational loss databases, rather than only submitting new and amended
events every quarter. The commenters stated that separating the new and amended loss events
would be burdensome. The original proposal was written in the spirit of reducing burden on the
respondents and therefore, based on the comments, the Federal Reserve will revise the
instructions to require submission of the entire database every quarter.
One commenter suggested allowing BHCs additional time to submit their data after the
quarter end. In order to facilitate timely risk monitoring and entry to supervisory models,
operational loss data must be submitted within time schedule prescribed; therefore, the Federal
Reserve will implement the time schedule requirements as proposed.
One commenter suggested that certain BHCs be exempt from using the definitions of
Level 1 and Level 2 Business Lines as described in the instructions. The Federal Reserve
believes that having consistent definitions of business lines is critical for the comparability of
data across BHCs. BHCs should map their internal business lines as defined in the instructions.
The Federal Reserve will implement the requirements as proposed.
As mentioned above, the Federal Reserve will not require BHCs to submit legal reserves
data as part of the June collection. Instead, the Federal Reserve re-opened the public comment
period for 60 days2 and is requesting comments on collecting these data in one or more of the
following ways:
1. Collect the data on an aggregate level rather than on a granular loss event-level (for
example, the number of loss events and the average estimated reserve amount for these
events);
2. Collect data on legal reserves in an anonymous fashion such that neither the identity of
the BHC or the loss event would be known; and
3. Collect the data in a way such that BHCs would submit a combination of actual and
randomized data3, so as not to reveal how any particular data item would or could tie
back to an actual loss event for a particular BHC.
In addition, the Federal Reserve also is requesting comment on other methods that would
allow the Federal Reserve to measure, understand, and analyze these types of legal risk without
requiring a BHC to submit data on specific legal reserves.
                                                            
2

On June 4, 2012, the Federal Reserve published a final Federal Register notice (77 FR 32970) announcing the
revisions to the FR Y-14A/Q/M and re-opening the public comment period for 30 days seeking additional comments
on the legal reserves data items. This comment period was scheduled to end on July 5, 2012. In response to
industry feedback, a Federal Register notice was published on June 27, 2012 (77 FR 38289) extending the public
comment period for an additional 30 days. The comment period now expires on August 6, 2012.
3
Randomizing survey responses have been a common technique when asking sensitive questions since the mid1960’s. The Federal Reserve will provide more detail on such techniques upon request and anticipates that industry
outreach calls would be conducted if this reporting option is selected.

7 
 

The collection of these data or any new reporting requirements related to these data
would take effect no sooner than the September 30, 2012, report date.

8 
 


File Typeapplication/pdf
File TitleMicrosoft Word - FRY14AQM_20120628_Discussion of Comments.docx
Authorfrbuser
File Modified2012-06-28
File Created2012-06-28

© 2024 OMB.report | Privacy Policy