CHASE O
October 28,2014
Kevin Watters
CEO • Mortgage Banking

Sent by U.S. Mail and via
FederalRegis terComments @
cfpb.gov
Ms. Monica Jackson
Office of the Executive Secretary
Consumer Financial Protection Bureau
1700 G Street, NW
Washington, DC 20552
Re: Proposed Amendments to Regulations Implementing the Home Mortgage D isclosure Act
(Regulation C) (Docket No. CFPB-20 14-0019 or RIN 31 70-AA 10)
Dear Ms. Jackson:
JPMorgan Chase & Co. ("Chase," "we," "our," or "us") appreciates this opportunity to share
with the Bureau of Consumer Financial Protection ("Bureau") our comments on certain aspects
of the above-referen ced proposal ("Proposal") to amend the Bureau's Regulation C.
Chase's comments are set forth below. For the most part, our comments focus on the issues that
are of the greatest concern to us. However, the proposed regulation is complex and detailed, and
many of those details could pose significant compliance challenges to financial institutions. We
would be pleased to provide further detail on any of our comments, as well as information on
more technical points, if that will be helpful to the Bureau. Chase has been working closely with
two of its trade groups, the Consumer Bankers Association and the Mortgage Bankers
Association, and we endorse the comments that those trade groups will be separately submitting
to the Bureau.

1.

Background and Summary

Chase is currently the second largest originator and second largest servicer of residential
mortgage loans in the United States. In 20 13, Chase originated and purchased 851 ,463
Regulation C-reportable mortgage loans nationwide.
The Bureau's proposed modifications to Regulation C represent the most significant revisions to
that regulation since the enactment of the Home Mortgage Disclosure Act ("HMDA") nearly 40
years ago. Chase recognizes that the availability of Regulation C data plays a critical role in
meeting HMDA/Regu lation C's stated purposes- providing citizens and public officials with
loan data to help determine whether financial institutions are serving the housing needs of their
communities, assisting public officials in distributing public sector investments so as to attract
private investment to areas where it is needed, and assisting in identifying possible
discriminator y lending patterns and enforcing antidiscrimina tion statutes. We fully endorse

 CHASE O
those critical purposes, and understand and support the necessity of capturing and making
available accurate HMDA data to meet those purposes. We also recognize that the types of data
fields that are needed to meet those purposes have evolved over time, as evidenced by the
number of new data fields that were mandated by Section 1094 of the Dodd-Frank Wall Street
Reform and Consumer Protection Act (" Dodd-Frank"). At the same time, Chase believes that
where the Bureau exercises the considerable discretion granted by Section l 094, it is necessary
to balance the actual benefits expected to be realized against the additional costs and burdens
imposed on reporting financial institutions as well as the consequences of requiring the
additional reporting. This need to seek balance is at the core of a number of our comments
below.
Our comments relating to the Proposal are briefly summarized as follows:

•

Effective Date. The amended Regulation C should apply to mortgage applications that are
received and boarded on and after the first January 1st that occurs after the 241h month
following the Bureau' s issuance of the final rule.
• Quarterly Reporting for Larger Originators. The perceived benefits of quarterly reporting
are unfounded and are not justified by the additional costs and burdens. lfthe Bureau
determines that more frequent reporting is necessary for larger financial institutions, Chase
urges the Bureau to require semiannual reporting rather than quarterly reporting. If the
Bureau decides to impose quarterly (or semiannual) reporting, that reporting requirement
should be delayed for 24 months fo llowing the effective date of the amended regulation
generally.
• Protection of Borrower Privacy and Data Security. The proposed test to balance the
protection of consumer privacy interests with the benefits of making HMDA data public is
too narrow, and needs to be expanded to sufficiently protect consumers. Financial
institutions should continue to release on the modified Loan Application Register ("LAR")
only those data fields that are currently released under existing Regulation C. Extensive
protections should be put into place before any additional HMDA data is made public. Some
commenters may suggest a bifurcated approach, under which financial institutions would
release on the modified LAR only those data fields that are currently released under existing
Regulation C, but where certain select members of the public - such as academics or persons
affiliated with research institutes- would have access to some or all of the new HMDA data
fields. Chase would oppose this approach unless and until there is assurance that consumers
will be fully protected against certain risks.
• Data Integrity Standards. The Bureau should revert to the reporting standards that were in
effect prior to the issuance of its 2013 guidance on HMDA/Regulation C compliance
management, resubmissions, and HMDA enforcement. If quarterly (or semiannual)
reporting is imposed on larger financial institutions, a "good faith effort" standard for their
quarterly (or semiannual) reports should be implemented. At the very least, the Bureau
should adopt a more relaxed standard for the first year that the revised regulation is in effect,
and, if applicable, for the first year that quarterly (or semiannual) reporting is in effect for
larger originators.
• The Number ofNew Data Fields. The addition of data fields in addition to those required
by Dodd-Frank is unnecessary and not justified by the associated burden and expense.

2

 CHASE O
•

Commercial Loans. The Bureau should exempt all commercial loans from HMDA reporting
(i.e. , loans, both open-end and closed-end alike, that are primarily for business or commercial
purposes in which a dwelling is taken as additional security). If, nonetheless, the Bureau
determines that it will subject commercial loans to HMDA reporting, Chase recommends that
reporting be limited to commercial loans with the three purposes set forth in current
Regulation C - that is, home purchase, home improvement, or refinancing. Further, we
suggest that the Bureau's final rule exempt commercial loans from certain additional data
fields, as discussed in Part 7, below.
• Multifamily Loans. As in the case of commercial loans, multifamily loans should be
exempted from HMDA reporting altogether. If, nonetheless, the Bureau determines that it
will subject multifamily loans to HMDA reporting, Chase recommends that reporting be
limited to multifamily loans with the three purposes set forth in current Regulation C - home
purchase, home improvement, or refinancing. We further suggest that the Bureau's final rule
exempt multifamily loans from the same additional data fields we have recommended for
commercial loans.
• Mixed-Use Properties. A bright-line test should be established to determine whether a
mixed-use property is a "dwelling" for purposes of Regulation C. To this end, if more than
50% of the square footage or income of a mixed-use property consists of, or is derived from,
a non-residential use, the property should not be treated as a dwelling. Any other mixed-use
property should be treated as a dwelling.
• Home Equity Lines of Credit and Other Open-End Lines of Credit. HMDA reporting of
consumer home equity lines of credit should remain optional. As noted above, the Bureau
should end HMDA reporting for all commercial loans (as defined above). Alternatively, if
the final rule continues to subject commercial loans generally to HMDA reporting, the rule
nevertheless should exempt commercial open-end credit plans.
• Repurchased Loans. Loans that are bought back for breaches of representations and
warranties or other conditions are being repurchased solely because the repurchasing
financial institution is contractually bound to do so. Reporting these involuntary buy-backs
as purchases is inconsistent with the purposes of HMDA and distorts the HMDA data of
reporting financial institutions. These involuntary repurchases should be excluded from
HMDA reporting altogether.
• Unsecured Home Improvement Loans. Chase agrees with the Bureau·s proposed exclusion
of unsecured home improvement loans from HMDA reporting.
• Loans in Which Real Property Collateral is Taken in an Abundance of Caution. Where a
dwelling is taken by a financial institution as security for a loan in an abundance of caution,
Regulation C should be amended to exclude the loan from HMDA reporting.
• Exclusion of Transitory Residences from the Definition of a 1'Dwelling." Proposed
Comment 2(f)-2 states that a "dwelling" excludes "transitory residences such as hotels,
hospitals, and college dormitories." Proposed Comment 4(a)(6)-4, which provides an
example of a corporation-owned property used to house employees that would be reported as
an investment property, should be clarified to exclude a property that is used as a transitory
residence for employees, inasmuch as such a property will be excluded from a "dwelling" in
the first instance in accordance with proposed Comment 2(1)-2. Further, proposed Comment
2(t)-2 should be revised to identify other types of transitory residences, including skilled
nursing care faci lities, group homes, prisons, and continuing care retirement facilities.
3

 CHAS E O
•

2.

Geocoding. The Bureau should not take responsibility for providing geocoding information
to reporting financial institutions. If the Bureau determines that it will provide geocoding
information, financial institutions should retain the option of determining and reporting the
geocoding information on their own. Further, if the Bureau provides geocoding information
to financial institutions, those institutions should not be responsible for the Bureau's own
geocoding errors or for correcting those errors.

Effective Date of the Amended Regulation

Dodd-Frank authorizes the Bureau to exercise considerable flexibility in establishing an effective
date for its Regulation C amendments, and Chase urges the Bureau to use that authority to
provide the time that financial institutions legitimately will need to implement the new
provisions properi y. Additional time will be required for the larger financial institutions, such as
Chase, that will be subjected to the quarterly reporting requirement if that requirement is
included in the final regulation.

Section 1094's Requirements. The Bureau has the leeway under Section 1094 to establish a
realistic effective date that is consistent with the actual burdens associated with the
implementation process. In this regard, Section 1094 of Dodd-Frank establishes a minimum
time period before most of the new data fields must be reported, but does not establish a
maximum time period for this reporting. More specifically, while Section 1094 states that
financial institutions cannot be required to report the new data fields under Section 304(b)(5) or
(6) ofHMDA before the first January 1st that occurs after the end of the 9-month period
following the Bureau's issuance ofthe final regulation, Section 1094 imposes no limitations on
the maximum period between the issuance of a final regulation and the effective date of that
regulation.
Necessity for a Minimum 24-Month Implementation Period. The challenges that all fmancial
institutions will face in implementing the regulation as proposed cannot be overstated. New
policies and procedures must be established; new systems must be developed to capture and
report the new data fields; intellectual technology must be created to support the new regulatory
requirements; new personnel must be hired, and both new and existing personnel must be
trained; legal, compliance, internal audit, and other groups must coordinate and develop
processes to support the new regulation; and the entire implementation effort must be monitored
and tested extensively. Although the Bureau has estimated the time and expense associated with
the implementation effort, experience has shown that these kinds of assessments prove to be
underestimated once the actual work of implementation begins.
It must be recognized that financial institutions already will have their personnel fully engaged
with other regulatory compliance efforts at the same time that the final Regulation C
amendments are issued by the Bureau. In particular, fmancial institutions will be implementing
the new TILA/RESPA disclosure regulations, will be implementing the federal risk retention
regulation, and will still be fine tuning their compliance programs relating to other Dodd-Frank
mortgage regulations that already are in effect, including the ability-to-repay rule. The stark
reality is that there are severe limitations to the ability of financial institutions to absorb yet
4

 CHASE O
another massive compliance effort- regardless of how many additional people they hire - unless
there is a reasonable implementation period. Moreover, the Congressional purpose behind
Section 1094 will not be met if an unduly short implementation period results in a high level of
inaccurate reporting. Under the circumstances, Chase believes that a minimum 24-month
implementation period following issuance of the final rule will be tight, but is realistic. More
specifically, we urge the Bureau to apply the amended regulation to mortgage applications that
are received and boarded on and after the first January 1st that occurs after the 24th month
following the issuance of the final rule.
Two items support the appropriateness of our proposed minimum 24-month implementation
period. First, the Securities and Exchange Commission's ("SEC ' s") recently revised Regulation
AB - which also requires the gathering and reporting of extensive loan-level data for residential
mortgages underlying certain publicly issued asset-backed securities -has a two-year and 60 day
implementation period. Second, when the last major revision to Regulation C was made, the
implementation period was just under 24 months. (The Federal Reserve Board ("Board" )
approved the Regulation C amendments on January 23, 2002, the original effective date was
January 1, 2003, and the effective date was later extended to January 1, 2004 due to a recognition
by the Board that the "2003 deadline does not afford institutions adequate time to take the steps
necessary to ensure full compliance with the new rules.") Thus, the Board provided nearly 24
months for financial institutions to implement Regulation C amendments that, by any measure,
were far less extensive than the amendments that the Bureau is proposing to establish.
Consistent with the approaches taken by the SEC and the Board, Chase urges the Bureau to
establish the minimum 24-month implementatio n period described above for the Regulation C
amendments.

3.

The Quarterly Reporting Requirement

Appropriateness of Quarterly Reporting. The Proposal requires financial institutions that
originate more than a threshold level of mortgages to report on a quarterly, rather than an annual,
basis. The Bureau states that this element of the Proposal "is based on considerations relating to
the timeliness of HMDA data submitted, the quality of the data submitted, and the Bureau's
desire to make annual HMDA data available to the public earlier than they are currently made
available." On closer examination, Chase does not believe that these considerations hold up, and
that the perceived benefits of quarterly reporting are not justified by the additional costs and
burdens.
The Bureau states that the current rule allows HMDA data to be reported as many as 14 months
after final action is taken on an application or loan, and that "this delay impairs the ability of the
Bureau and the appropriate agencies to use HMDA data to effectuate the purposes of the statute
in a timely manner." As the Bureau is well aware, the 14-month period represents the maximum
delay, and most HMDA data is reported far sooner. To Chase' s knowledge, the delay in
reporting has never impaired either the efforts of the Bureau itself or, in the past, the Board or
other agencies, to discharge their responsibilities to effectuate the purposes ofHMDA in a timely
manner. In particular, even where an agency preliminarily identifies problematic HMDA data,
there is usually a delay of many months - if not a year or longer- before the issues are sorted out
5

 CHASE O
and a decision to take action is made. In addition, the quarterly data currently submitted by large
financial institutions - although preliminary in nature, and not yet "cleaned upH- can serve as an
early warning system, providing the Bureau with information about possible trends that it may
wish to monitor without increasing the compliance burden and enforcement risk faced by those
institutions. Thus, in practice, we do not believe that the current annual reporting requirement
adversely impacts the purposes of HMDA or the taking of enforcement action where this is
appropriate.
The Bureau also states that quarterly reporting would require "financial institutions with larger
transaction volumes to review and edit smaller batches of reportable data several times
throughout the year," and that the Bureau believes that "quarterly reporting would facilitate and
enhance compliance with HMDA, reduce reporting errors, and improve the quality of HMDA
data." In fact, our experience indicates that the exact opposite is true. Contrary to the Bureau's
suggestion, large fmancial institutions that would be subjected to the quarterly reporting
requirement do not engage in the review and editing process only once a year. Rather, these
institutions engage in a qualitative review process throughout the year and continuing through
the end of the 60-day period fo llowing year-end until the annual report must be made. If
quarterly reporting is imposed, these institutions will have less time to complete their data
validation processes to ensure accurate data is submitted. Moreover, most financial institutions
establish technology service queues on an annual basis, and if problems need to be corrected to
ensure accurate HMDA reporting it will not be feasible - or, at best, it will be extremely
disruptive- to fix these problems in sufficient time to meet a quarterly reporting requirement. In
short, the Bureau's Proposal would establish artificial deadlines for large financial institutions'
year-round qualitative review process, which will certainly not result in enhanced compliance
and, more realistically, can be expected to result in a higher error rate. If quarterly reporting is
adopted, Chase would support a cumulative reporting system instead of treating each quarterly
LAR as a final, standalone document. Each quarter' s filing would add to and upgrade previous
quarters, working toward a final year end LAR. Data should be published, and financial
institutions should be examined, only after the LAR for the entire calendar year has been
completed.
Although the Bureau acknowledges that it does not currently anticipate that HMDA data would
be released to the public more frequently than annually, it believes that quarterly reporting may
allow the Bureau and the Federal Financial Institutions Examination Council ("FFIEC") "to
expedite disclosures of annual HMDA data to the public and better serve the public disclosure
goals of the statute." This belief appears to be speculative at best. The FFIEC' s annual
disclosure ofHMDA data to the public is in the hands of the FFIEC, not the Bureau, and there is
no assurance that quarterly reporting by larger institutions will, in practice, result in a
meaningfully earlier publication of that data by the FFIEC.
It is necessary to weigh the speculative benefits of quarterly reporting against the very real costs
and burdens. As noted above, qualitative review by larger financial institutions is already a yearround process, and it will be operationally difficult to meet the quarterly reporting mandate,
particularly following quarters where market conditions result in a surge of new mortgage
applications and loan purchases. Larger institutions will need to increase their staffmg
considerably, at substantial expense, to meet the quarterly reporting requirement. Even with
6

 CHASE O
increased staffing, we are concerned that quarterly reporting - particularly in the short tenn will result in reduced, not enhanced, accuracy. Accordingly, we urge the Bureau to put aside the
quarterly reporting concept at this time.
If, notwithstanding the difficulties and drawbacks described above, the Bureau determines that
more frequent reporting is necessary for larger financial institutions, Chase urges the Bureau to
require semiannual reporting rather than quarterly reporting. While semirumuaJ reporting would
still present the same kinds of problems, the reporting of data twice a year would reduce the
pressure on larger financial institutions' data validation processes. As a result, more accurate
data will be reported if semiannual, rather than quarterly, reporting is mandated. However, if the
Bureau wishes to obtain the most accurate data, it will retain the current annual reporting
requirement.

Delay for the Quarterly Reporting Requirement. The Bureau has solicited feedback on the
appropriate delay for the effective date of the quarterly reporting requirement (that is, the delay
beyond the effective date for the new Regulation C amendments generally) if quarterly reporting
is imposed for larger financial institutions. The Bureau stated that it is considering a delay of at
least one year for the quarterly reporting requirement.
Chase appreciates the Bureau's recognition of the additional challenges that will be faced by
financial institutions that are subjected to the quarterly reporting requirement. As stated above,
qualitative review by larger financial institutions is already a year-round process, and it will be
operationally difficult to meet the quarterly reporting mandate. If the Bureau determines that it
will impose a quarterly (or semiannual) reporting requirement for larger financial institutions, we
ask the Bureau to delay that reporting requirement for 24 months following the effective date of
the amended regulation generally. We see this as the timeframe that will be necessary to
implement a reasonably accurate quarterly (or semiannual) reporting process. In addition, we
recommend that the Bureau modify its data integrity standards to reflect the adverse impact that
a quarterly (or semiannual) requirement will have on reporting accuracy, as more fully discussed
in Part 5, below.

4.

Protection of Borrower Privacy and Data Security

The significant increase in the number and scope of data fields contemplated by the Proposal
raises several important consumer privacy issues with respect to financial institutions' disclosures
of HMDA data to the public in the modified LAR. The Bureau has recognized the privacy risks
associated with the Proposal, stating that "privacy interests arise where the data's disclosure may
both substantially facilitate the identification of an applicant or borrower and disclose
information about the applicant or borrower that is not otherwise public and may be harmful or
sensitive." Accordingly, the Bureau has called for the use of a balancing test "to determine
whether and how HMDA data should be modified prior to its public release in order to protect
applicant and borrower privacy while also fulfilling the public disclosure purposes of the
statute." Applying this balancing test, the Bureau is proposing that financial institutions include
on the modified LAR only those data fields that are currently released under existing Regulation
C. The Bureau has solicited comment on various aspects of the privacy issue, including the
7

 CHASE O
appropriateness of the balancing test and the privacy risks created by and disclosure benefits of
the regulation's data fields.
Chase believes that the consumer privacy issue is one of the most significant concerns raised by
the Proposal, and offers several comments below.

The Balancing Test. The Bureau's balancing test states that "public HMDA data [should] be
modified only when the release of the unmodified data creates risks to applicant and borrower
privacy interests that are not justified by the benefits of such release to the public for the
statutory purposes." We believe that this balancing test is too narrow, and needs to be expanded
to sufficiently protect consumers. First, in weighing the benefits against the risk of HMDA data
disclosure, the Bureau should also take into consideration the availability of data from other
sources (e.g., data brokers, public record sources, and the Internet), and consider how all
available data - whether legally or illegally obtained - can be combined to identify and harm
consumers. For example, Chase's Decision Sciences group has determined that simply making
public three data points - the dollar amount of the loan (even rounded to the nearest $1,000), the
application date and the census tract - when combined with publicly available information, will
enable the accurate identification of 95% of borrowers. Second, we believe that it is not
sufficient to consider the potential invasion of a consumer's privacy interests; rather, that it is
essential to consider how consumers might otherwise be harmed. Chase is particularly
concerned that HMDA data will be used not only to identify consumers, but to subject
consumers to identity theft and other serious crimes. Those crimes have the potential to harm
consumers far more than their loss of privacy.
Release of Data Fields on the Modified LAR. As noted above, the Bureau is proposing that
financial institutions include on the modified LAR only those data fields that are currently
released under existing Regulation C. Chase fully agrees that the new data fields should not be
disclosed on the modified LAR. The new HMDA data fields contain highly personal and
sensitive consumer financial information. As discussed above, the release of HMDA data could
expose consumers to a loss of privacy and serious crimes such as identity theft. Release of
HMDA data in addition to what is currently made available, especially sensitive items such as
credit scores and denial reasons, could lead to the exploitation of the Nation's most vulnerable
consumers, such as by way of targeted marketing of potentially predatory loan products. If this
were to occur, it would also expose the Bureau and financial institutions alike to reputation risk
and potential liability, and this could undermine consumers' faith in the mortgage process.
Moreover, we are concerned that competitor financial institutions could use public HMDA data
to reverse-engineer our proprietary underwriting systems, thereby harming our competitive
position in the mortgage marketplace. Accordingly, Chase urges the Bureau to proceed with the
utmost of caution before requiring the release of any of the new data fields. To this end, the
Bureau should not require the release of any new HMDA data fields unless and until the Bureau
has confirmed that this data (when combined with other available data) cannot be used to
identify consumers, invade their privacy, or subject them to identity theft and other serious
crimes. Given the sophistication of hackers and similar criminals, it may be necessary for the
Bureau to engage the services of outside experts who can determine whether the release of any
new HMDA data fields can be reverse-engineered or otherwise used to harm consumers.
Moreover, financial institutions should not be subjected to liability associated with the loss of
8

 CHASE O
consumer privacy or consumer crimes by reason of their compliance with a Bureau regulation.
Accordingly, at the same time, if ever, that the Bureau requires the release of any new HMDA
data fields, the Bureau should amend Regulation C to make clear that financial institutions will
have no liability for making public the data fields required by the regulation. If it is not clear that
financial institutions can be so protected by Regulation C, the Bureau should request Congress to
enact the necessary protective legislation and should not require the release of new HMDA data
fields until Congress has done so. It would be particularly appropriate for the Bureau to
approach Congress with such a request inasmuch as Section 1094 of Dodd-Frank specifically
contemplated that Congress might need to enact further legislation to implement the purposes of
HMDA. More specifically, Section 1094 enacted a new Section 307(b) ofHMDA (12 U.S.C.
§ 2806(b)), which reads, in part, as follows: "The Director of [the Bureau] shall recommend to
the Committee on Banking, Housing, and Urban Affairs of the Senate and the Committee on
Financial Services of the House of Representatives, such additional legislation as the Director of
[the Bureau] deems appropriate to carry out the purposes of this title."

Selective Release of New HMDA Data. Some commenters may suggest a bifurcated approach,
under which financial institutions would include on the modified LAR only those data fields that
are currently released under existing Regulation C, but where certain select members of the
public - such as academics or persons affiliated with research institutes- would have access to
some or all of the new HMDA data fie lds. These commenters may argue that this approach
would allow these researchers to enhance their investigation of fair lending and other legitimate
issues. Even assuming that a process could be developed to identify legitimate and objective
researchers (which may be questionable), Chase would oppose this approach unless and until
there is assurance that consumers will be fully protected against certain risks.
Chase is concerned that a bifurcated approach would expose consumers to two categories of risk
-the leakage of HMDA data by these researchers to unauthorized persons, and the unlawful
access ofHMDA data by unauthorized persons. Either the leakage or the unauthorized access of
this data could, once again, expose consumers to the loss of privacy and to identity theft and
other serious crimes. To protect against leakage, the Bureau and researchers will need to enter
into stringent confidentiality agreements that have strong enforcement provisions. To protect
against unlawful access, the Bureau must impose state of the art security standards on the
researchers, and must confirm the effectiveness of those security standards. If a bifurcated
approach to the public disclosure of the new HMDA data is authorized by the Bureau's final
rule, both the confidentiality agreement and security standard requirements must be incorporated
into Regulation C. Absent such an approach, the resulting risk to consumers is unacceptable and
Chase would oppose the bifurcated approach.

5.

Data Integrity Standards

While Chase acknowledges that financial institutions' LARs need to be accurate, we believe that
it is appropriate for the Bureau to restore its previous HMDA data integrity standards, and to
provide further leniency in the event that quarterly (or semiannual) reporting is required for
larger financial institutions.

9

 C:HASE O
In 2013, the Bureau issued its guidance on HMDA/Regulation C compliance management,
resubmissions, and HMDA enforcement. See CFPB Bulletin 2013- 11 (October 9, 2013). This
guidance established a significantly tighter standard for the triggering of LAR resubrnissions.
Resubmissions generally will be triggered based on a 4% sample error rate for financial
institutions reporting 100,000 or more HMDA LAR entries, and based on a I 0% sample error
rate for institutions reporting fewer than 100,000 HMDA LAR entries. Even lower error
thresholds may trigger LAR resubmissions in some instances. For example, with larger
institutions error rates below 2% in an individual field may call for resubmission if errors prevent
an accurate analysis of the institution's lending. While no specific data integrity standards are
identified in this guidance for the taking of a public enforcement action, the guidance sends an
unambiguous message that the tolerance for errors has been reduced.
Given the large number of new data fields proposed by the Bureau, the challenges that financial
institutions will face in implementing the revised Regulation C, and the time that will be needed
to work through the unforeseen operational problems that invariably will surface following
implementation, it will be appropriate to establish a more relaxed data integrity standard. The
standards for a public enforcement action should similarly be adjusted. To this end, we
recommend that the Bureau revert to the standards that were in effect prior to the issuance of its
2013 guidance.
As more fully discussed in Part 3, above, it will be extremely difficult for larger financial
institutions to comply with a quarterly (or semiannual) reporting requirement. Qualitative
review by larger financial institutions is already a year-round process, and converting that
process into a shorter timeframe is likely to result in less accuracy. That being the case, it is
appropriate for the Bureau' s data integrity standards to be modified in a manner that realistically
recognizes the operational hurdles that will be triggered by a quarterly (or semiannual) reporting
requirement. To this end, Chase recommends that Section 1003.6 of Regulation C be modified
to establish a "good faith effort" standard for the quarterly (or semiannual) reports submitted by
larger financial institutions. Under this standard, any errors or omissions in a financial
institution's quarterly (or semiannual) reports will not be treated as a violation ofHMDA or
Regulation C so long as the financial institution made a good faith effort to record and submit the
data fields in question.
At the very least, we urge the Bureau to adopt a more flexible standard for the first year that the
revised regulation is in effect, and for the first year that quarterly (or semiannual) reporting, if
applicable, is in effect for larger originators.

6.

The Number of New Data Fields

Regulation C currently requires financial institutions to report 17 data fie lds. Section 1094 of
Dodd-Frank requires the Bureau to amend Regulation C to require the reporting of 17 new data
fields, and the Bureau has proposed to require the reporting of20 additional data fields (" Bureauoptional data fields"). In fact, the number of new reportable data fields may be considerably
larger because some data fields require the insertion of specific information where "other" is
applicable (e.g., the reason for denial of an application.) Some data fields are not to be reported
10

 CHASE O
for certain loans (e.g., the rate-spread disclosure applies only to covered loans that are subject to
the Bureau' s Regulation Z).
Chase believes that the number of new data fields is excessive. For many financial institutions
(particularly those offering open-end lines of credit), the Proposal will more than double the line
items that they will be required to report. While we certainly acknowledge that the Bureau must
require the reporting of the data fields mandated by Dodd-Frank, the Bureau is not required by
Section I 094 to add yet an additional 20 data fields. Any governmental agency can always find
a reason - and perhaps what appears to be a good reason - to require the reporting of more
information, but the question is whether the benefits of the additional information reporting
justify the additional cost and burden as well as the risks associated with the exposure of
consumer information. Chase believes that the purposes of HMDA will be fu lly met by the
addition of the data fields mandated by Dodd-Frank. The marginal benefit of reporting the
Bureau-optional data fields will be modest. This modest benefit must be weighed against the
considerable cost and burden of building the infrastructure necessary to capture and report these
items (particularly those data fields, such as the debt-to-income ratio for certain commercial
loans subject to HMDA, that are not currently captured). Chase estimates that it will incur
significant additional initial and annual expense to report the Bureau-optional data fields.
Smaller financial institutions presumably will incur lesser costs, but those costs will represent a
larger portion of their annual budgets. Moreover, the possible release of more consumer
information raises the privacy and data security risk issues discussed in Part 4 of this letter,
above. On balance, we think the scale tips in favor of not mandating the reporting ofthe Bureauoptional data fields at this time. We recommend that the Bureau delete the Bureau-optional data
fie lds from the final rule and that it critically reevaluate the situation after the new rule has been
in place for a reasonable period of time. At that point, the Bureau will be in a stronger position
to determine whether the reporting of additional data fields is truly necessary to meet the
purposes of HMDA.
Some commenters may argue that the Bureau-optional data fields will not be burdensome when
compared to the more than 270 data fields that issuers of publicly-issued residential mortgagebacked securities ("RMBS") will be required to capture and disclose under the SEC's Regulation
AB. While such an argument may have some surface appeal, it does not hold up to scrutiny.
First, Regulation AB addresses an entirely different statutory purpose than Regulation C. To
meet the purposes of the federal securities laws, Regulation AB requires the disclosure of loan
data to potential investors in publicly-issued RMBS so that they will have the ability to make
knowing investment decisions. The SEC has determined that private investors require this
wealth of data to meet federal securities law standards. In contrast, the Bureau can meet the
statutory requirements and purpose ofHMDA by requiring the reporting of only those data fields
that are specifically mandated by HMDA. Second, the Regulation AB data fields must be
disclosed only by issuers of publicly-issued RMBS. A financial institution that does not issue
publicly-issued RMBS will not be required to capture and disclose the Regulation AB data
fields. At present, the public RMBS market is virtually non-existent, which means that, in
practice, almost no financial institutions will be required to capture and disclose the Regulation
AB data fie lds. Even if the public RMBS market is resurrected in the future, the reality is that
relatively few fmancial institutions will ever likely become issuers of publicly-issued RMBS.
Accordingly, very few fmancial institutions will ever be required to capture and disclose the
11

 CHASE O
Regulation AB data fields, while all frnancial institutions will be required to capture and disclose
the Regulation C data fields. Third, the data integrity standard under Regulation AB (i.e., a
materiality standard) is potentially far less stringent than the standard that applies under
Regulation C (which, under the Bureau's Bulletin 2013-11 , generally requires LAR
resubmissions based on a 4% sample error rate for financial institutions reporting 100,000 or
more HMDA LAR entries, and LAR resubmissions based on a 10% sample error rate for
institutions reporting fewer than 100,000 HMDA LAR entries). The potentially more flexib le
data integrity standard under Regulation AB enables those few issuers who will be subject to that
regulation more leeway in their capturing and disclosure of the required data fields. For these
reasons, the more extensive data fields that must be captured and disclosed under Regulation AB
do not provide a basis for justifying the Bureau-optional data fields under the Proposal.

7.

Commercial Loans

The Bureau has solicited feedback regarding whether it would be appropriate to modify some of
the proposed data reporting requirements for commercial transactions, or whether to exclude
commercial transactions from HMDA reporting altogether. For the reasons discussed below,
Chase believes that the Bureau should exempt all commercial loans, both open-end and closedend alike, from HMDA reporting.
For purposes of this Part 7, we define "commercial loans" to mean loans that are primarily for
business or commercial purposes (as that phrase is used in§ 1026.3(a)(l) of the Bureau' s
Regulation Z and the related commentary) in which a dwelling is taken as additional security.
(Multifamily loans, in which the dwellings are the primary security, are separately addressed in
Part 8 of this letter, below.) The loans that we discuss in this Part 7 are those that are treated by
financial institutions and regulators alike as true commercial loans. Examples include loans to
provide working capital for a business, loans to finance the purchase of inventory for a business,
loans to finance accounts receivable, loans to purchase equipment for a business, and loans to
finance the purchase or expansion of an office building or factory. Where loans of this type are
made to a business, the lender will sometimes require the borrower or principal of the borrowing
entity to provide a lien on his/her home or other dwelling as additional security. This is
frequently a junior lien with relatively little equity behind it, and the lien is usually taken to
incent the borrower or principal to make the business succeed and repay the loan. Historically,
few of these loans were HMDA-reportable because the loans were not for home purchase, home
improvement, or refinancing purposes. The Proposal would change the commercial lending
landscape dramatically because all commercial loans secured by dwellings would be subject to
HMDA reporting. As more fully discussed below, we believe that the expansion ofHMDA
reporting to commercia] loans is inappropriate.

Reporting of Commercial Loans is Not Needed to Meet the Purposes of HMDA. Since its
inception, HMDA has had a home loan focus. While the purposes of HMDA, summarized
above, are broadly stated, the legislative history of that statute, and the implementation of the
statute by the Bureau (and by the Board, prior to Dodd-Frank), leave no doubt that HMDA was
intended to obtain home loan-related data so that housing- which plays such a critical role in the
Nation's economy and in the preservation of communities nationwide - would be protected.
12

 CHASE O
Commercial loans, as described above, play a virtually non-existing role in the Nation's housing.
These loans have no meaningful relationship to the housing needs of any community, and
reporting on these loans will not provide any useful information with regard to whether financial
institutions are meeting housing needs. To the contrary, when commercial loans include
dwellings as additional security, the nexus between those loans and housing - and, therefore, the
purposes of HMDA - is tangential at best.
The insignificant relationship between commercial loans and HMDA' s purposes is further
underscored by the substantial differences between the way that commercial loans and home
loans are underwritten and priced. Commercial loans are generally made based on the strength
of the business applicant, which is measured primarily by the cash flow of the business,
management of the business, the fmancial institution's assessment of the likelihood of the
business' continued success, guarantor support of the business, and the available collateral. The
loan is then structured and the terms are set based on the financial institution' s individualized
assessment of the risks posed by these criteria. This process bears almost nothing in common
with the manner in which home loans are made. Moreover, while a dwelling is typically the only
collateral provided in support of a home loan, and while that dwelling will fully collateralize the
home loan, a dwelling provided as collateral for a commercial loan will typically be relatively
insignificant to the overall structure and collateral for the loan. For example, a dwelling may be
added as security for a commercial loan in order to shore up a collateral position the financial
institution considers marginal or in an abundance of caution. In short, the dwelling component
of commercial loans is incidental in nature and wholly unrelated to the role of dwellings in home
loans.
Commercial Loan HMDA Data Will Not Be Useful. Given the tangential linkage between
commercial loans and HMDA' s purposes, as discussed above, any HMDA data reported by
financial institutions is not likely to be useful to the Bureau or the general public. Further, given
the significant differences between various types of commercial loans- let alone between
commercial loans and home loans - any comparison of the HMDA data among these loans is not
likely to be meaningful. The commercial loan underwriting process is much more complex than
the underwriting of home loans, and this process differs significantly between loan types. For
example, different types of commercial collateral (e.g., accounts receivable vs. equipment vs.
commercial real estate) are considered differently in the underwriting process, different
industries pose different risk profiles, perishable inventory is less meaningful as collateral than
non-perishable, and start-up businesses pose different risks than established companies.
Similarly, a financial institution will underwrite a commercial loan to finance inventory
differently than a commercial loan to finance the purchase of a factory. We see no viable way to
compare the HMDA data of these loans in a manner that will allow for a meaningful use of the
information for fair lending enforcement or other HMDA purposes. To the contrary, given the
significant differences between different types of commercial loans described above, we are
concerned that the reporting of HMDA data for commercial loans will have a distortive effect on
the overall information reported by financial institutions, and that this will undermine, rather
than promote, HMDA's purposes.

13

 CHASE O
There is No Suggestion that Congress lnte11ded to Expand the Scope of Regulation C to
111clude All Commercial Loa11s tltat Take Dwelli11gs as Collateral. If a financial institution
determines that a commercial loan must be secured, it will require the borrower to provide a
security interest in the borrower's accounts receivable, inventory, equipment or commercial real
estate and, as noted above, any dwelling taken as additional collateral will be secondary at best.
When Congress enacted Dodd-Frank, it surely was well aware of the current scope ofRegulation
C. In its current form, Regulation C is limited to home purchase, home improvement, and
refinancing loans. Commercial loans, as described above, were not subject to HMDA reporting.
While Section 1094 of Dodd-Frank increased the number of reportable data fields and authorized
the Bureau to establish additional data fields, this section did not change the purposes of HMDA
or direct the Bureau to expand the scope of HMDA to include nonpurchase/improvement/refinancing commercial loans. Surely, Congress would have addressed
this issue in Section 1094 if it had intended such a dramatic expansion of HMDA. It is not
appropriate for the Bureau to expand the scope ofHMDA in this manner absent a clear direction
from Congress to do so.
Commercial Loans Will Represent an lnsignificallt Portio11 ofAll Dwelli11g-Secured
Lending. Commercial loans will constitute an insignificant portion of the Nation's dwellingsecured lending. While precise numbers are not available because commercial loans have not
historically been subject to HMDA reporting, we comfortably predict that commercial loans will
represent less than 1% of Chase's LAR submissions if the final rule is promulgated as proposed.
Although we acknowledge that Chase is only one lender, as the industry's fifth largest originator
of residential loans and as a major commercial lender, we would expect that our commercial
lending experience is representative of the broader industry. Given that commercial loans will
represent such a small portion of ail dwelling-secured loans, it follows that the submission of
commercial loan data will be insignificant in terms of meeting HMDA's purposes.
Information on Commercial Loans Would Not Have Helped Public Officials Address the
Informational Shortcomings ofthe Recent Financial Crisis. In the Supplementary Information
published as part of the Proposal ("Supplementary Information"), the Bureau discussed the
HMDA informational shortcomings that were identified when the financial crises commenced in
2007. The Bureau reported that public officials needed more data to identify at-risk
neighborhoods and to develop foreclosure relief and home ownership stabi lization programs.
However, given the small portion of all dwelling-secured loans represented by commercial loans,
and the fact that commercial loans take dwellings as security only on an incidental basis (i.e.,
these loans are not housing-focused), it is almost inconceivable that commercial loan HMDA
data would have made any difference whatsoever in public officials' identification of at-risk
neighborhoods or in their development of foreclosure relief and home ownership stabilization
programs. The insignificance of commercial loan HMDA data for these purposes is further
underscored by the fact that Regulation C does not currently, and would not as proposed, require
the reporting of the rate-spread disclosure for commercial loans, and no commercial loans would
be identified as loans subject to the Home Ownership and Equity Protection Act of 1994
("HOEPA"); both of these data points would be at the heart of public officials' informational
needs. Accordingly, commercial loans will continue to be insignificant with respect to the
expanded data required by public officials to identify at-risk neighborhoods and develop

14

 CHASE O
foreclosure relief and home ownership stabilization programs in connection with a future
financial crises.

The Proposal Already' Exempts Commercial Loans from Many of Regulation C's Reporting
Requirements, Which Means that the Data that is Reportable for such Loans is of Limited
Utility for Public Officials. If the Proposal is implemented as proposed, a number of data fields
will not be applicable to commercial loans. As more fully discussed below, many other data
fields are not relevant to commercial loans and will routinely result in the reporting of"N/A" or
its equivalent. None of this is surprising given the fact that HMDA reporting has always been
structured with a residential, rather than a commercial, perspective. It is certainly appropriate to
exclude commercial loans from the data fields as proposed - those data fields simply have no
place in commercial lending. The remaining data fields for commercial loans will be of limited
utility to public oflicials if and when they wish to identify at-risk neighborhoods and develop
foreclosure relief and home ownership stabilization programs in connection with a future
financial crises.
The Cost and Burden of Commercial Loan Compliance is Not Justified by the Modest
Expected Benefits. In light of the foregoing, Chase strongly believes that the benefits ofHMDA
reporting for commercial loans will be minimal. At the same time, the cost and burden of
building out an infrastructure to capture and report required HMDA data - particularly the new
data fields - on commercial loans will be real, considerable and ongoing. This cost and burden
cannot be justified by the minimal benefits hoped to be achieved. Many financial institutions
may change their credit requirements, such as by requiring borrowers (or their guarantors) to
post collateral other than dwellings as additional security for commercial loans. This will likely
have the undesired effect of impeding access to commercial credit. For these reasons, Chase
recommends that commercial loans be excluded from HMDA reporting altogether.
If, nonetheless, the Bureau determines that it will subject commercial loans to HMDA reporting,
Chase recommends that reporting be limited to commercial loans with the three purposes set
forth in current Regulation C - that is, home purchase, home improvement, or refinancing.
These loan purposes would have a more tangible linkage to the purposes of HMDA.
Moreover, if commercial loans will continue to be subject to HMDA reporting, we suggest that
the Bureau's final rule exempt commercial loans from the following additional data fields:
§ 1003.4(a)(2) [government insured or guaranteed loan], (4) [request for preapproval], (10)
[income information 1] , (23) [monthly debt-to-income ratio], (24) [loan-to-value ratio], (34)
[NMLSR ID], (35) [name of automated underwriting system2] , (36) [reverse mortgage], (38)
[ability-to-repay and qualified mortgage], and (39) [amount of draw on a home equity line of
credit or open-end reverse mortgage at account opening]. All of these require the submission of
data that simply does not apply to the way that commercial loan applications are underwritten,
decisioned, or priced. We also suggest that commercial loans be exempted from data fields

1

Alternatively, the Bureau can expand the instruction in Appendix A, Paragraph 4(a)(l0)(ii)-l.b. to allow the entry of"N/A" for
both multifamily loans and commercial loans.
2
Other than possibly with respect to the U.S. Small Business Administration' s 7a Small Loan program.

15

 CHASE O
1003.4(a)(10)(i) [government monitoring information] and (15) [credit score information3] as we
believe the number of commercial transactions in which this information would be relevant
would be insignificant because the transaction would need to be both to an individual and
secured by a dwelling. While we acknowledge that financial institutions could simply report
"NIA" (or its equivalent) for certain of these data fields in the LAR, even this reporting will be
burdensome. More specifically, where an institution must report a data field, even one that will
always result in aN/ A response, the institution will be required incur the expense necessary to
build the infrastructure for that reporting, and must incur the ongoing expense associated with
the quality control process to ensure accurate reporting ofthat data field. There is no meaningful
benefit that the Bureau or the public will derive from the reporting of such N/A data fields on
commercial loans, which means that these expenses cannot be justified.
Further, we suggest that the Bureau's final rule exempt commercial loans from the following
additional data fields: § 1003.4(a)(5) [construction method - site built or manufactured home] ,
(29) [legal classification of manufactured home as real or personal property] and (30) [whether
land underlying a manufactured home is owned or leased]. The common thread of these three
data fields is the presence of a manufactured home on the security property. Chase believes that
it is highly unusual for a commercial loan (as defined above) to be secured by such a property
under circumstances in which the property will meet the definition of a "dwelling" under
§ 1003.2(t) and the related commentary provisions. The rarity of such properties indicates that
the benefits of reporting HMDA data for these commercial loans is not justified by the burden
and expense of that reporting.
Finally, if commercial loans will be subject to the final rule, we urge the Bureau to address how a
syndicated or participated commercial loan should be reported. In our judgment, such a loan
should be reported solely by the lead financial institution, which can always be easily identified
by a reference to the loan file. The lead institution takes full responsibility for negotiating the
loan terms, processing the loan application, and documenting the loan. The Bureau can eliminate
uncertainty relating to the reporting responsibilities for these loans, as well as inappropriate dual
reporting, by a finaJ rule that makes clear that the lead institution alone should report the
origination, denial or other applicable action.

8.

Multifamily Loans

Many of the considerations discussed above with respect to commercial loans also apply to
multifamily loans that are commercial purpose loans. For this reason, we recommend that
multifamily loans likewise be exempted from HMDA reporting altogether.

3

Many commercial lenders utilize internal scores that include information from various sources. While some information may be
taken from a report from a consumer reporting agency, other information may be based on, for example, the loan application, a
commercial bureau report, the type of collateral, and other dealings with the applicant. We do not believe that such an internal
score constitutes a ''credit score" as defined in the Proposal, even if information provided by a consumer reporting agency is
included, and request that the definition of·'credit score' for purposes of the final regulation be modified to clarify that these
types of internal scores are not included.

16

 CHASE O
As in the case of commercial loans, there are significant differences between the way that
multifamily loans and home loans are underwritten and priced (i.e., multifamily loans are
underwritten much in the same way as commercial loans, with an emphasis on each loan's
unique cash flow, collateral value, borrower strength, and credit support). Multifamily loan
HMDA data also is not likely to impact public officials' identification of at-risk neighborhoods
or in their development of foreclosure relief and home ownership stabilization programs in
connection with future financial crises. Further, multifamily loans are already excluded from a
number ofHMDA data field disclosure requirements. Accordingly, the benefits ofHMDA
reporting for multifamily loans - particularly for those that are not for home purchase, home
improvement or refinancing purposes - will be minimal. At the same time, the cost and burden
of maintaining and building out an infrastructure to capture and report HMDA data on all
multifamily loans will be considerable. We conclude that the benefits are not justified by the
additional cost and burden. For these reasons, we recommend that the final rule exempt
multifamily loans from HMDA reporting altogether.

If, nonetheless, the Bureau determines that it will subject multifamily loans to HMDA reporting,
Chase recommends that reporting be limited to multifamily loans with the three purposes set
forth in current Regulation C -home purchase, home improvement, or refinancing. Once again,
these loan purposes would have a more tangible linkage to the purposes of HMDA. In addition,
we suggest that the Bureau's fmal rule exempt multifamily loans from the same additional data
fields discussed in Part 7, above, with respect to commercial loans. Further, if multifamily loans
will be subject to HMDA reporting, the final rule should clarify that a syndicated or participated
multifamily loan should be reported solely by the lead financial institution.

9.

Mixed-Use Properties

The Proposal adds a new Comment 2(f)-3, which addresses when a property used for both
residential and commercial purposes (a "mixed-use property") will be treated as a ''dwelling"
under Regulation C. If the property's primary use is residential, then it will be treated as a
dwelling. While the comment states that a financial institution may use any reasonable standard
to determine the primary use of the property, an institution must consider a property with five or
more individual dwelling units to have a primary residential use. The Bureau has solicited
feedback on whether it would be preferable to establish a bright-line rule for mixed-use
properties. The Bureau has solicited specific feedback on whether a mixed-use property should
be reported if it includes any individual dwelling units, or whether a clear standard can be
provided for mixed-use property with a de minimis residential component to be excluded. Chase
believes that it is appropriate to exclude a mixed-use property from HMDA reporting where the
property has a primary non-residential use, and that a bright-line test can be fashioned to identify
those properties.
Regulation C requires only the reporting of a loan secured by a property that has a primary
residential use. If the property has a primary non-residential use, then it simply does not serve
the purposes ofHMDA and Regulation C to mandate reporting of the loan in the LAR. To the
contrary, the reporting of loans secured by properties with only an incidental residential use
poses the danger of distorting financial institutions' HMDA data, which will undermine the
17

 C:HASE O
purposes of the statute and regulation. Chase believes that all stakeholders- the Bureau, other
government agencies, consumer groups, and financial institutions alike - would best be served
with a bright-line test for determining when a property will be deemed to have a primary
residential versus non-residential use.
In determining a property's primary use, it is not appropriate simply to count the number of
individual dwelling units on the property. Because this approach ignores all other factors
regarding the use of the property, it poses the genuine risk of over- or under-reporting for
HMDA purposes. For example, if a financial institution makes a loan secured by an entire
college campus, the fact that there are five dwelling units somewhere on the campus for use by
the college president and other college officials does not change the fact that virtually the entire
campus is used for non-residential purposes. Similarly, if a financial institution makes a loan
secured by an extensive church property, the fact that there is a small rectory with five dwelling
units for use by the church' s clergy does not alter the fact that the vast bul k of the lender's
collateral consists of a church used for non-residential purposes. The security properties for
these loans, and other loans of a comparable nature, are not for a primary residential use by any
stretch ofthe imagination, and these loans should not be HMDA-reportable.
Given these considerations, Chase believes that a better approach will be to establish a brightline test that considers the dwelling component of the mixed-use property in the context of the
property as a whole. To this end, we propose a simple bright-line test based on square footage or
income, as determined by the reporting fmancial institution. That is, if more than 50% of the
square footage of a mixed-use property consists of non-residential use (or, where the income test
is used, if more than 50% of the income of a mixed-use property is derived from a nonresidential use), then the property will not be treated as a dwelling. Any other mixed-use
property will be treated as a dwelling. This approach makes sense in that a property whose
resources are primarily devoted to or derived from non-residential purposes cannot meaningfully
be said to be primarily for a residential use. This approach is also consistent with the Bureau's
existing commentary provisions relating to the definitions of ''home improvement loan" and
"home purchase loan," except that it specifies a minimum 50+% square footage or income
standard for the non-residential use (thereby precluding a circumvention of the purpose ofthe
bright-line test).

10.

Home Equity Lines of Credit and Other Open-End Lines of Credit

Regulation C currently provides for optional reporting ofHMDA data in the case of consumer
home equity lines of credit ("HELOCs") that are used, in whole or in part, fo r a home purchase
or home improvement purpose. For this purpose, a HELOC is defined by reference to the
definition in Regulation Z, which means that the term is limited to borrowers who are natural
persons that obtain credit primarily for a personal, family or household purpose. As the Bureau
has acknowledged, only 1% ofHELOCs are currently reported under Regulation C.
The Proposal would impose mandatory reporting ofHMDA data for all HELOCs, regardless of
purpose. Further, the Proposal would require reporting for all open-end lines of credit ("openend credit plans") as defined in§ 1026.2(a)(20) of Regulation Z that are secured by dwellings
18

 CHASE O
(except for reverse mortgages and Regulation C-excluded transactions generally), but without
regard to the purpose of the credit or whether the lender is a "creditor" under Regulation Z. In
short, this means that reporting would be required for almost all open-end credit plans, including
those for a business or commercial purpose or secured by multi-family dwellings.
Chase believes that the dramatic expansion of the coverage of open-end credit plans is
unwarranted. Because virtually no fmancial institutions currently report HMDA data on
consumer open-end credit plans, let alone on commercial open-end credit plans, it will be
necessary to build an infrastructure from the ground up to capture the existing Regulation C data
plus the newly created data fields. The costs and burdens associated with this build-out are not
justified by the benefits expected to be gained. In the consumer space, there have not been
suggestions over the years of serious fair lending or other HMDA-focused problems with openend credit plans, in contrast with the home purchase, refinance, and home improvement loan
markets. Moreover, a large percentage of consumers use their open-end credit plans as reserve
lines of credit (i.e., they use these lines as a source of"rainy day" funds), which means that these
credit products are fundamentally different than other types of mortgage loans. The vast
majority of consumer open-end credit plan lending has been by financial institutions that have
substantial closed-end mortgage portfolios, and any fair lending problems associated with those
portfolios are likely to be identified by the HMDA data already submitted by those institutions.
Accordingly, HMDA reporting of consumer open-end credit plans should remain optional.
The Proposal indicates that there are substantial concerns regarding the potential abuse of seniors
who obtain reverse mortgages. At present, almost all reverse mortgages are made under HUD's
Home Equity Conversion Mortgage Insurance Program, and a significant percentage of these
mortgages are structured as open-end credit plans. That being the case, the better approach is to
limit the mandatory reporting of consumer open-end credit plans to reverse mortgages. The
Proposal to subject all consumer open-end credit plans to HMDA reporting is an overly inclusive
approach to what is otherwise a legitimate concern.
In any event, if the final rule continues to subject commercial loans (which, as noted above, we
define to mean loans that are primarily for business or commercial purposes in which a dwelling
is taken as additional security) to HMDA reporting, the rule nevertheless should exempt
commercial loans structured as open-end credit plans. Dwellings are sometimes taken as
additional security for commercial lines of credit provided to business borrowers. These are
generally underwritten as if they were not secured by the dwellings and, in practice, are not
regarded by either financial institutions or their business customers as real estate loan products.
That being the case, there simply is no significant HMDA-recognized rationale that justifies the
cost and burden of bui lding an infrastructure for the reporting ofHMDA data on such
commercial open-end credit plans.

11.

Repurchased Loans

The Proposal provides guidance on how loan repurchases should be reported on the LAR for
both annual and quarterly reporters. Following previous informal guidance from the FFIEC, the
Proposal generally treats a loan repurchase as a purchase that must be reported for HMDA
19

 CHASE O
purposes. The Bureau has solicited feedback generally on how repurchases should be treated for
purposes of Regulation C and, more specifica1ly, feedback on whether repurchases should be
reported under Regulation C in the first instance.
Chase believes that the reporting of loans that are bought back for contractual breaches of
representations and warranties or other conditions is inconsistent with the purposes of HMDA
and distorts the HMDA data of reporting financial institutions. The revision of Regulation C
provides the Bureau with an opportunity to correct the FFIEC's misguided approach on the
repurchase issue.
Loans that are bought back for breaches of representations and warranties or other conditions are
being repurchased solely because the repurchasi ng financial institution is contractually bound to
do so. In practice, these repurchases- which have greatly increased in number since the start of
the mortgage market collapse in 2007 - are very costly to the repurchasing institution, and most
institutions resist repurchase demands vigorously. Thus, by any standard, when a financial
institution actually repurchases loans the repurchase transactions are entirely involuntary. For
HMDA reporting purposes, it is inappropriate to equate these involuntary repurchases with
ordinary voluntary loan purchases in the market. The submission of data relating to an
involuntary repurchase does not achieve any of the purposes of HMDA - that is, it does not
enable the Bureau or a member of the public to determine whether financial institutions are
serving the housing needs of their communities, does not assist public officials in distributing
public-sector investment so as to attract private investment to areas where it is needed, and does
not assist in identifying possible discriminatory lending patterns and enforcing antidiscrimination
statutes. To the contrary, the reporting of these involuntary repurchases as if they were voluntary
purchases seriously distorts the HMDA data of the reporting financial institutions, potentially
disguising problematic data or overstating positive data. For example, where a financial
institution originates a loan in a low-income neighborhood and sells it to an investor, and then
subsequently repurchases the loan due to breach of a representation, the financial institution
would be recording the same low-income neighborhood loan twice in its LAR, thereby
overstating its actual low-income neighborhood lending activity. For these reasons, we urge the
Bureau to exclude these involuntary repurchases from HMDA reporting altogether.
The Bureau should not consider itself bound by the FFIEC' s 2010 informal pronouncement on
the reporting of repurchased loans. That pronouncement was not in the form of a formal
regulation or order and simply provided its guidance in a conclusory manner without any
supporting analysis of how that guidance would (or would not) meet the purposes ofHMDA.
The Bureau should take this opportunity to fix the problems associated with the FFIEC's
pronouncement.

12.

Exclusion of Unsecured Home Improvement Loans

The Bureau has proposed to exclude non-dwelling secured home improvement loans from
HMDA reporting. For all ofthe reasons stated in the Proposal, Chase supports this amendment
to Regulation C. The market for home improvement loans has changed considerably since the
enactment of HMDA. Today, we believe that financial institutions originate relatively few
20

 CHASE O
unsecured home improvement loans, and that institutions generally do not rely on unsecured
home improvement loan data for purposes of fair lending examinations. We see no other
significant benefits in continuing to capture and report data on unsecured home improvement
loans. In short, the burden of reporting HMDA data on such loans outweighs the value of the
information produced.

13.

Loans in Which Real Property Collateral is Taken in an Abundance of Caution

Financial institutions sometimes take real property collateral as security for a loan in an
abundance of caution. Chase recommends that where a dwelling is taken by a financial
institution as security for a loan in an abundance of caution, Regulation C should be amended to
exclude the loan from HMDA reporting.
Abundance of caution loans are generally underwritten as if there was no dwelling collateral, and
they are regarded by financial institutions as such. The dwellings almost always serve as
additional collateral for secured commercial or multifamily loans. Abundance of caution loans
have been made since time immemorial, and the federal supervisory agencies and financial
institutions alike are comfortable in recognizing these loans for what they are - fundamentally
non-dwelling-secured loans in which the dwellings provide some further assurance of repayment.
The appraisal regulations of the federal banking agencies exempt abundance of caution loans
from the requirement of an appraisal or evaluation. See, e.g., 12 C.F.R. § 34.43(a)(2) and the
Final Interagency Appraisal and Evaluation Guidelines issued by five fed eral banking agencies
on December 2, 20 10. In addition, the FFIEC's Call Report Instructions (Glossary, at A-58a)
defines a "loan secured by real estate" to exclude certain loans where a lien or liens on real
property have been taken as collateral solely through an abundance of caution.
In the interest of regulatory consistency, we urge the Bureau to exclude abundance of caution
loans from HMDA reporting. Because, in substance, these are non-dwelling-secured loans, the
purposes ofHMDA are not met by reporting HMDA data on these loans. Indeed, ifHMDA data
continues to be reported by financial institutions on such loans (i.e. , as if they were dwellingsecured), this could distort the institutions' HMDA data. This will be particularly of concern
going forward ifthe Bureau's final rule includes all of the new data fields described in the
Proposal.

14.

Exclusion of Transitory Residences from the Definition of a "Dwelling"

Proposed Comment 2(t)-2 states that a "dwelling" excludes "transitory residences such as hotels,
hospitals, and college dormitories." We wish to raise two issues relating to this provision.
Business-Owned Temporary Housingfor Employees. Chase believes that it would be
appropriate to add a clarification to proposed Comment 4(a)(6)-4 with respect to a property
owned by a corporation or other business to house its employees on a temporary basis.

21

 C:HASE O
Corporations and other businesses frequently own residential properties to provide lodging for
their employees on a short-term basis. For example, a corporation may own a building near one
of its offices, and use the building to house employees who are brought to the office to work on
special projects for a few days or weeks at a time. Once the proj ects are completed, the
employees vacate the building and return to their homes. The corporation may also temporarily
house an employee who is being permanently relocated to that office until he/she can obtain
permanent housing of his/her own. As noted above, Comment 2(f)-2 states that "transitory
residences, such as hotels, hospitals, and college dormitories" are excluded from the definition of
a "dwelling," which means that loans secured by such properties are excluded from HMDA
reporting. By any interpretation, a business-owned property used for the type of temporary
employee housing described above should be treated as a "transitory residence" under Comment
2(f)-2, which means that it should be excluded from the definition of "dwelling" for HMDAreporting purposes. Comment 4(a)(6)-4, which addresses whether a HMDA-reportable property
should be identified as a "principal residence," a "second residence," or an "investment
property," discusses the case of a corporation that buys a property which is a dwelling that it
does not occupy, and that it uses to house its employees (who do not pay rent). The comment
states that such a property should be reported as an investment property. There should not be a
conflict between Comment 2(f)-2 and Comment 4(a)(6)-4 since the example used in Comment
4(a)(6)-4 is, by its terms, limited to a property that is a "dwelling." However, to avoid any
ambiguity, we ask the Bureau to revise Comment 4(a)(6)-4 to clarify that the example used in
that comment would not include a property that is used as a transitory residence for employees,
inasmuch as such a property will be excluded from a "dwelling" in the first instance in
accordance with Comment 2(f)-2.
Skilled Nursing Care Facilities, Group Homes, Prisons, and Continuing Care Retirement
Facilities. As discussed above, proposed Comment 2(f)-2 states that a "dwelling" excludes
"transitory residences such as hotels, hospitals, and college dormitories." The Bureau has
solicited feedback on "whether the proposed revisions provide institutions with sufficient clarity
to identify transactions that must be reported and whether any additional exclusions or examples
would be appropriate." Chase recommends that this comment be expanded to identify other
examples of transitory residences, including skilled nursing care facilities, group homes (e.g. , for
troubled youths, individuals with certain disabilities, or individuals recovering from substance
abuse), prisons, and continuing care retirement facilities. The Proposal indicates that, in
determining which types of properties will be excluded as ''transitory residences," it is necessary
to focus on the "transitory nature of such structures." The Bureau is certainly correct in
identifying hotels, hospitals and college dormitories as examples of properties that have an
inherently transitory nature (even though students frequently live in college dormitories for four
or more years, and even hotels and hospitals can have individuals who are occupants for
indefinite periods of time). Similar to hotels, hospitals, and college dormitories, we believe that
skilled nursing care facilities, group homes, prisons, and continuing care retirement facilities many of which receive government and/or charitable funding assistance, reflecting the public
benefits provided by these facilities - are essentially transitory in nature. While it is true that
some occupants of these types of facilities may reside for extended periods of time, all of these
facilities tend to house people who are in a transitory period of their lives. Accordingly, we ask
the Bureau to revise Comment 2(f)-2 to identify other types of transitory residences, including
ski lled nursing care facilities, group homes, prisons, and continuing care retirement facilities.
22

 CHASE O
15.

Geocoding

In its Supplementary Information, the Bureau stated that it "is exploring operational changes that
it may achieve using the reported postal address that would reduce the burden associated with
geocoding" (i.e. , providing the census tract, MSA or MD, county and State of a property).
Noting that financial institutions report frequent examination errors relating to geocoding, the
Bureau stated that it may create a system where a financial institution will report only the postal
address of the property and the Bureau will then provide the fmancial institution with the census
tract, county, MSA or MD, and State. For the reasons discussed below, Chase does not believe
that the Bureau should be responsible for providing geocoding information to reporting financial
institutions. If the Bureau determines that it will provide geocoding information, financial
institutions should retain the option of determining and reporting the geocoding information on
their own. Further, if the Bureau provides geocoding information to financial institutions, those
institutions should not be responsible for the Bureau' s own geocoding errors or for correcting
those errors.
While the Bureau has indicated that many financial institutions experience freq uent examination
errors relating to geocoding, other financial institutions- including several small entity
representatives on the Bureau's Small Business Review Panel- do not find geocoding to be
problematic. Regulation C has mandated the reporting of geocoding informat ion for many years,
and a great number of financial institutions do not find it difficult to report this information on a
consistently accurate basis. Thus, the possible rationale for the Bureau' s providing of geocoding
information - the reduction of the burden associated with geocoding - simply is inapplicable to
many financial institutions.
Moreover, the shifting of geocoding responsibilities to the Bureau would deprive financial
institutions of a critical source of information. Chase (and, we assume, many other financial
institutions as well) accesses the geocoding information built into their HMDA reporting
infrastructure on a real-time basis for a variety of essential business functions. For example, we
use this information in making decisions relating to our sales of mortgage loan pools throughout
the year, so that we can know if these transactions fall within our low and moderate income
communities, which is essential for our Community Reinvestment Act compliance program. If
geocoding is no longer a part of our HMDA reporting infrastructure, it will not be possible for us
to access this information on a real-time basis. The delay in obtaining geocoding information
from the Bureau would be so significant that we would not be able to use that information in a
meaningful fashion in making these business decisions. As a consequence, we would be
required to rebuild this geocoding capability outside our HMDA reporting infrastructure at a
considerable additional expense.
For both of these reasons, we recommend that the Bureau's final rule continue to assign
geocoding responsibilities to the reporting financial institutions. However, we wish to be
sensitive to the needs of other financial institutions that acknowledge ongoing difficulty in
reporting accurate geocoding information. Accordingly, ifthe Bureau determines that it will
provide geocoding information, the final rule should give financial institutions the option of
23

 C:HASE O
determining and reporting the geocoding information on their own. In this way, the needs of all
financial institutions can be fairly met.
In its Supplementary Information, the Bureau acknowledged the potential questions that would
arise if it assumed responsibility for providing geocoding information to reporting financial
institutions. These "questions include whether a financial institution would be responsible for
the accuracy of the information provided by the Bureau and whether a financial institution would
be responsible for geocoding an entry if the Bureau's geocoding system returned an error." The
Bureau solicited feedback on whether such an operational change would alleviate burden and on
whether such an operational change is appropriate generally.
If the Bureau decides to provide geocoding information, Chase believes that it would be
inappropriate to hold financial institutions responsible for the accuracy of the information
provided by the Bureau. First, basic principles of fairness make clear that each entity must be
responsible for its own actions, and it will be patently unfair to hold a financial institution
responsible for the Bureau's own errors. Second, holding fmancial institutions responsible for
the Bureau' s geocoding errors (including correction of the Bureau' s errors) would mean, in
practice, that financial institutions would still be required to determine geocoding for each
property so that they can be assured that the Bureau's geocoding is correct. It is evident that this
would completely undermine the supposed benefits of having the Bureau provide geocoding
information in the first instance. Consequently, if the Bureau determines that it will provide
geocoding information to financial institutions, the Bureau needs to assume responsibility for its
own actions. This means that financial institutions should be entitled to fully rely on the
Bureau's information. It fo llows that when the Bureau makes geocoding errors, financial
institutions should not be responsible for those errors or for correcting those errors.

* * *
We are pleased to have had this opportunity to provide you with our comments on the Proposal.
If you have any questions concerning this comment letter, or would like to discuss further any of
the matters that we have raised, please feel free to contact the undersigned.

24