[House Hearing, 110 Congress]
[From the U.S. Government Publishing Office]




 
                      GAO REPORT ON REGULATION B:
                     SHOULD LENDERS BE REQUIRED TO
                      COLLECT RACE AND GENDER DATA
                      OF BORROWERS FOR ALL LOANS?

=======================================================================

                                HEARING

                               BEFORE THE

                            SUBCOMMITTEE ON
                      OVERSIGHT AND INVESTIGATIONS

                                 OF THE

                    COMMITTEE ON FINANCIAL SERVICES

                     U.S. HOUSE OF REPRESENTATIVES

                       ONE HUNDRED TENTH CONGRESS

                             SECOND SESSION

                               __________

                             JULY 17, 2008

                               __________

       Printed for the use of the Committee on Financial Services

                           Serial No. 110-129


                     U.S. GOVERNMENT PRINTING OFFICE
44-902 PDF                 WASHINGTON DC:  2008
---------------------------------------------------------------------
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512ï¿½091800  
Fax: (202) 512ï¿½092104 Mail: Stop IDCC, Washington, DC 20402ï¿½090001


                 HOUSE COMMITTEE ON FINANCIAL SERVICES

                 BARNEY FRANK, Massachusetts, Chairman

PAUL E. KANJORSKI, Pennsylvania      SPENCER BACHUS, Alabama
MAXINE WATERS, California            DEBORAH PRYCE, Ohio
CAROLYN B. MALONEY, New York         MICHAEL N. CASTLE, Delaware
LUIS V. GUTIERREZ, Illinois          PETER T. KING, New York
NYDIA M. VELAZQUEZ, New York         EDWARD R. ROYCE, California
MELVIN L. WATT, North Carolina       FRANK D. LUCAS, Oklahoma
GARY L. ACKERMAN, New York           RON PAUL, Texas
BRAD SHERMAN, California             STEVEN C. LaTOURETTE, Ohio
GREGORY W. MEEKS, New York           DONALD A. MANZULLO, Illinois
DENNIS MOORE, Kansas                 WALTER B. JONES, Jr., North 
MICHAEL E. CAPUANO, Massachusetts        Carolina
RUBEN HINOJOSA, Texas                JUDY BIGGERT, Illinois
WM. LACY CLAY, Missouri              CHRISTOPHER SHAYS, Connecticut
CAROLYN McCARTHY, New York           GARY G. MILLER, California
JOE BACA, California                 SHELLEY MOORE CAPITO, West 
STEPHEN F. LYNCH, Massachusetts          Virginia
BRAD MILLER, North Carolina          TOM FEENEY, Florida
DAVID SCOTT, Georgia                 JEB HENSARLING, Texas
AL GREEN, Texas                      SCOTT GARRETT, New Jersey
EMANUEL CLEAVER, Missouri            GINNY BROWN-WAITE, Florida
MELISSA L. BEAN, Illinois            J. GRESHAM BARRETT, South Carolina
GWEN MOORE, Wisconsin,               JIM GERLACH, Pennsylvania
LINCOLN DAVIS, Tennessee             STEVAN PEARCE, New Mexico
PAUL W. HODES, New Hampshire         RANDY NEUGEBAUER, Texas
KEITH ELLISON, Minnesota             TOM PRICE, Georgia
RON KLEIN, Florida                   GEOFF DAVIS, Kentucky
TIM MAHONEY, Florida                 PATRICK T. McHENRY, North Carolina
CHARLES A. WILSON, Ohio              JOHN CAMPBELL, California
ED PERLMUTTER, Colorado              ADAM PUTNAM, Florida
CHRISTOPHER S. MURPHY, Connecticut   MICHELE BACHMANN, Minnesota
JOE DONNELLY, Indiana                PETER J. ROSKAM, Illinois
BILL FOSTER, Illinois                KENNY MARCHANT, Texas
ANDRE CARSON, Indiana                THADDEUS G. McCOTTER, Michigan
JACKIE SPEIER, California            KEVIN McCARTHY, California
DON CAZAYOUX, Louisiana              DEAN HELLER, Nevada
TRAVIS CHILDERS, Mississippi

        Jeanne M. Roslanowick, Staff Director and Chief Counsel
              Subcommittee on Oversight and Investigations

                MELVIN L. WATT, North Carolina, Chairman

LUIS V. GUTIERREZ, Illinois          GARY G. MILLER, California
MAXINE WATERS, California            PATRICK T. McHENRY, North Carolina
STEPHEN F. LYNCH, Massachusetts      EDWARD R. ROYCE, California
EMANUEL CLEAVER, Missouri            RON PAUL, Texas
MICHAEL E. CAPUANO, Massachusetts    STEVEN C. LaTOURETTE, Ohio
AL GREEN, Texas                      J. GRESHAM BARRETT, South Carolina
RON KLEIN, Florida                   MICHELE BACHMANN, Minnesota
TIM MAHONEY, Florida                 PETER J. ROSKAM, Illinois
ED PERLMUTTER, Colorado              KEVIN McCARTHY, California
JACKIE SPEIER, California


                            C O N T E N T S

                              ----------                              
                                                                   Page
Hearing held on:
    July 17, 2008................................................     1
Appendix:
    July 17, 2008................................................    43

                               WITNESSES
                        Thursday, July 17, 2008

Braunstein, Sandra F., Director, Division of Consumer and 
  Community Affairs, Board of Governors of the Federal Reserve 
  System.........................................................     7
Cavalluzzo, Ken, Research Analyst, Wisconsin Capital Management 
  LLC............................................................    19
Corralejo, Jorge C., Chairman, Latino Business Chamber of of 
  Greater Los Angeles............................................    24
Gnaizda, Robert F., General Counsel, The Greenlining Institute...    20
Himpler, Bill, Executive Vice President of Federal Affairs, 
  American Financial Services Association (AFSA).................    22
Sullivan, Ann, President, Madison Services Group, on behalf of 
  Women Impacting Public Policy..................................    25
Williams, Orice, Director, Financial Markets and Community 
  Investment, United States Government Accountability Office.....     5

                                APPENDIX

Prepared statements:
    Braunstein, Sandra F.........................................    44
    Cavalluzzo, Ken..............................................    55
    Corralejo, Jorge.............................................    65
    Gnaizda, Robert F............................................    67
    Himpler, Bill................................................    71
    Sullivan, Ann................................................    76
    Williams, Orice..............................................    80

              Additional Material Submitted for the Record

Watt, Hon. Melvin L.:
    Letter to Federal Reserve Chairman Alan Greenspan from 
      Representative Frank and 22 other Democratic Members, dated 
      February 12, 2004..........................................   101
    Response letter from Federal Reserve Chairman Alan Greenspan 
      to Representative Frank, dated March 8, 2004...............   103
    Letter to Federal Reserve Chairman Ben Bernanke from 
      Representatives Frank, Watt, Waters, Gutierrez, and Lee, 
      dated February 14, 2007....................................   105
    Response letter from Federal Reserve Chairman Ben Bernanke to 
      Representative Frank, dated March 13, 2007.................   107
    Letter to the the Government Accountability Office requesting 
      the report from Representatives Frank, Watt, and Maloney, 
      dated July 16, 2007........................................   109
    Letter to the National Credit Union Administration from 
      Representative Watt, dated July 10, 2008...................   111
    Letter to the Office of Thrift Supervision from 
      Representative Watt, dated July 10, 2008...................   113
    Letter to the Federal Deposit Insurance Corporation from 
      Representative Watt, dated July 10, 2008...................   115
    Letter to the U.S. Small Business Administration from 
      Representative Watt, dated July 10, 2008...................   117
    Letter to the U.S. Department of Justice from Representative 
      Watt, dated July 10, 2008..................................   119
    Letter to the Federal Trade Commission from Representative 
      Watt, dated July 10, 2008..................................   121
    Response letter from the Office of Thrift Supervision to 
      Representative Watt, dated July 16, 2008...................   123
    Response letter from the Office of the Comptroller of the 
      Currency to Representative Watt, dated July 15, 2008.......   127
    Response letter from the National Credit Union Administration 
      to Representative Watt, dated July 16, 2008................   129
    Response letter from the U.S. Small Business Administration 
      to Representative Watt, dated July 16, 2008................   132
    Response letter from the Federal Trade Commission to 
      Representative Watt, dated July 15, 2008...................   134
    Response letter from the Federal Deposit Insurance 
      Corporation to Representative Watt, dated July 16, 2008....   139
    Letter from the National Consumer Law Center to 
      Representative Watt, dated July 16, 2008...................   141
    Letter from the Federal Reserve Bank of Chicago to the 
      Federal Reserve Board, dated November 17, 1999.............   147
    Federal Reserve Bank of Boston, Working Paper entitled, 
      ``Credit Card Redlining,'' dated February 28, 2008.........   163
    Responses to questions submitted to Sandra Braunstein........   202
    Responses to questions submitted to Ken Cavalluzzo...........   208
    Responses to questions submitted to Bill Himpler.............   209
    Responses to questions submitted to Ann Sullivan.............   214
    Responses to questions submitted to Orice Williams...........   217


                      GAO REPORT ON REGULATION B:
                     SHOULD LENDERS BE REQUIRED TO
                      COLLECT RACE AND GENDER DATA
                      OF BORROWERS FOR ALL LOANS?

                              ----------                              


                        Thursday, July 17, 2008

             U.S. House of Representatives,
                          Subcommittee on Oversight
                                and Investigations,
                           Committee on Financial Services,
                                                   Washington, D.C.
    The subcommittee met, pursuant to notice, at 10:02 a.m., in 
room 2128, Rayburn House Office Building, Hon. Melvin L. Watt 
[chairman of the subcommittee] presiding.
    Members present: Representatives Watt, Lynch, Cleaver, 
Green; and McHenry.
    Ex officio: Representative Frank.
    Chairman Watt. Good morning, everybody.
    This hearing of the Financial Services Subcommittee on 
Oversight and Investigations will come to order.
    We will start with opening statements by the members. 
Without objection, all members' opening statements will be made 
a part of the record, and we will proceed with allowing 
whomever wants to, to make an opening statement since we have 
so few people here to do so today. I will recognize myself for 
5 minutes to kind of frame the issue for us.
    Today's hearing is entitled, ``GAO Report on Regulation B: 
Should Lenders Be Required To Collect Race and Gender Data of 
Borrowers for all Loans?'' The Financial Services Committee has 
a special interest in ensuring fair and equal access to credit, 
and our committee was instrumental in the passage of the Equal 
Credit Opportunity Act.
    The Federal Reserve issued Regulation B to implement the 
Act. Regulation B currently forbids the collection of racial, 
gender, and other personal characteristics data for loans other 
than mortgage loans, loans such as small business loans or 
automobile loans, for example.
    By contrast, the Federal Reserve's Regulation C, which 
implements the Home Mortgage Disclosure Act, commonly known as 
HMDA, requires the collection and public reporting of racial, 
gender, and personal characteristics data on applicants for 
home mortgages, and that data is regularly used by Congress, 
Federal banking regulators, researchers, and the public to 
discern patterns of lending and alert for public possible 
discrimination.
    Last year Chairman Frank, Congresswoman Maloney, and I 
requested the GAO to study Regulation B, with particular 
attention to the Federal Reserve's factual and analytical basis 
for concluding that removing Regulation B's prohibition on the 
collection of racial and gender data of applicants and 
borrowers could increase discrimination. Without objection, I 
will submit a copy of our request to the GAO and we will make 
that a part of today's hearing record.
    We are pleased that the GAO is here today to release the 
results of the study we requested, and today's hearing will 
focus on the GAO's findings and whether lenders should be 
required to collect race and gender data on applicants and 
borrowers of all loans. The GAO report reflects that serious 
data gaps now exist that impede the efforts of Federal banking 
regulators to enforce fair lending laws, especially on loans 
other than mortgage loans.
    The primary source of data for tracking patterns of non-
mortgage lending is the Federal Reserve's survey of small 
business finances, the so-called SSBF. SSBF data is collected 
from borrowers of small business loans rather than from 
lenders. The data is voluntary, self-reported, and not verified 
by the Federal Reserve. This limits the analytical value of the 
SSBF data.
    By contrast, the likelihood, or at least the possibility, 
that discrimination is occurring in mortgage lending can be 
detected from data collected under the Home Mortgage Disclosure 
Act. This data is required to be provided annually by a large 
population of lenders and can be used by Federal bank 
regulators to help facilitate fair lending examinations.
    Another limitation of SSBF data has been that the surveys 
of small business finances were conducted only about every 5 
years, instead of annually, between 1987 and 2003. 
Additionally, the Fed has been habitually slow in reporting the 
results of the surveys. The survey results have therefore often 
been stale. In fact, much of the information reviewed by the 
GAO for the report we are receiving at today's hearing is based 
on SSBF surveys conducted in 1993 and 1998, more than 10 years 
ago in some cases.
    Some of the other information in today's GAO report is 
based on the 2003 survey that was not publicly released until 
2006. Perhaps recognizing these problems, the Federal Reserve 
discontinued the SSBF in 2007. The Fed now says that elements 
of the SSBF will be incorporated into the Federal Reserve 
survey of consumer finances, but this is not due to be released 
until 2010.
    It is hard to believe, as we convene this hearing today, 
that no one in the Federal Government has access to reliable 
data about important lending patterns and the real prospect of 
disparities and discrimination in the provision of credit other 
than mortgage credit. Indeed, the GAO report indicates that one 
Federal Reserve bank has been unable to conduct thorough fair 
lending examinations and unable to review consumer complaints 
alleging discrimination by non-mortgage lenders, due to lack of 
available data. This is alarming, and, given the recent 
documented disproportionality of subprime mortgage loans to 
racial and ethnic minorities, I personally think it is 
unacceptable.
    The Federal Reserve has periodically reviewed Regulation B, 
most recently in 2003, and concluded that amending Regulation B 
to permit the collection of racial, gender, and personal 
characteristics data could increase discrimination. If that is 
still the Fed's contention, I certainly hope that it will offer 
specific evidence for that hypothesis at today's hearing.
    I am anxious to know whether the Fed thinks that the 
collection of racial and ethnic data under HMDA is contributing 
to, causing, or increasing discrimination by mortgage lenders. 
The Financial Services Committee needs to carefully consider 
the potential cost that would result from amending Regulation B 
to require the collection and public reporting of personal 
characteristics data on loans other than mortgage loans.
    For that reason, the subcommittee invited several lenders 
to testify at today's hearing about some of the potential costs 
that we have heard about, but they declined. Perhaps this 
subcommittee will attempt to get details about these costs at a 
subsequent hearing.
    We all should be attentive to holding down the cost of 
lending, but we also have a public policy obligation to ensure 
equal and fair lending to all Americans, whether they are 
shopping for mortgage loans or non-mortgage loans.
    The GAO report and today's hearing allow us to start the 
process of evaluating both the cost as well as the benefits of 
more comprehensive reporting. I encourage our subcommittee 
members to approach this hearing in that spirit, and I thank 
all of our witnesses for appearing today to assist us as we 
start this effort.
    I am now happy to recognize my colleague from North 
Carolina, who is substituting for the ranking member. Well, he 
says he is not substituting for Mr. Miller; he says he is 
standing in for Mr. Miller, for our ranking member, Gary 
Miller, who had another commitment this morning.
    My colleague, Mr. McHenry from North Carolina, is 
recognized for 5 minutes.
    Mr. McHenry. Chairman Watt, thank you, and thank you for 
hosting this hearing today.
    And I do think it is important that we follow up on the GAO 
report on the Equal Credit Opportunity Act, which is a law that 
is intended to enforce our Nation's fair lending laws. I look 
forward to hearing from the witnesses today, both this first 
panel and the second panel, regarding the efforts to eliminate 
discrimination in the credit industry.
    You know, access to credit has provided enormous 
opportunities and benefits to consumers. I think we have to 
ensure that credit is available to all, regardless of any race, 
ethnic background, or general considerations. And I think it is 
obvious and clear to all that is what we should be doing.
    We have to ensure that families have access and 
opportunities to purchase a home or an automobile, to finance 
an education, to deal with emergencies, and to purchase 
everyday goods and services. For this reason, and ensuring that 
there's non-discrimination within lending practices, the Equal 
Credit Opportunity Act, as implemented by the Federal Reserve 
by Regulation B, it prohibits lenders from collecting racial, 
ethnic, and gender information in order to make a credit 
decision.
    So we have to look at the unintended consequences of that, 
and I think it is important that we have this hearing to follow 
up on my colleague's request for this GAO report. I look 
forward to hearing the details of it, and I think it is 
important that we ensure that the detection and enforcement 
tools are there and available to us to protect consumers in 
this country and ensure that our laws our working 
appropriately.
    So with that, I thank you, Chairman Watt, for your 
leadership, and I look forward to this hearing.
    Chairman Watt. I thank the gentleman for his statement and 
for standing in for the ranking member.
    Without objection, all members' opening statements will be 
made a part of the record in their entirety, and I would be 
happy to recognize Representative Cleaver for an opening 
statement if he cares to make one.
    Mr. Cleaver. I will wait, Mr. Chairman, until a question-
and-answer period.
    Chairman Watt. Okay. Mr. Green from Texas.
    Mr. Green. Thank you, Mr. Chairman.
    I would like to make a statement, and I want to thank your 
ranking member today, Mr. McHenry. You seem to wear that seat 
quite well.
    Mr. Chairman, I thank you for hosting this hearing, because 
it is important for us to make meaningful decisions as to how 
we can eliminate discrimination, not only in lending, but 
discrimination in the main. And the most effective way to do it 
is to have empirical evidence of what is happening.
    The acquisition of empirical evidence necessitates 
collecting certain evidence, certain information. 
Unfortunately, we have not devised a system that will allow us 
to properly collect this information. My belief is that if we 
have the will, the way is readily available to us. We have to 
adopt the will to eliminate invidious discrimination. So I am 
hopeful that at this hearing this morning we can hear more 
about how we can get this done, how we can acquire the 
intelligence necessary to not only prove that the 
discrimination exists but have the intelligence such that it 
can be used in an efficacious way to deal with the actual 
problem itself. It is the will that we need. The way is 
available to us.
    And, Mr. Chairman, I thank you very much for hosting the 
hearing, and yield back the balance of my time.
    Chairman Watt. I thank the gentleman for his opening 
statement and for his presence today for this important 
hearing.
    I am now going to proceed with an abbreviated introduction 
of the witnesses. Without objection, therefore, your bios will 
be made a part of the record; and, in the interest of time, I 
won't go into an elaborate introduction, although both of them 
deserve and warrant elaborate introductions.
    I think our first panel is extremely important for two 
reasons, because one of the witnesses is testifying about the 
actual report that was requested some time ago, and by and 
large the essence of the report is about the Federal Reserve. 
So the other witness is here in fairness to hear and respond if 
the Federal Reserve desires to do so.
    Our first witness is Ms. Orice Williams. She is the 
Director of Financial Markets and Community Investment, United 
States Government Accountability Office. And our second witness 
on this panel will be Ms. Sandra Braunstein, the Director of 
the Division of Consumer and Community Affairs at the Federal 
Reserve Board.
    Without objection, your entire written statements will be 
made a part of the record, and each of you will be recognized 
for a 5-minute summary of your testimony. We tend to be very 
liberal in our assessment of 5 minutes, especially on this 
panel, where we are getting the basic information. So don't 
feel like you are under the gun, but try to be as cognizant as 
you can of the lighting system.
    The green light will be on for 4 minutes and then the 
yellow light will be on for 1 minute. And then the red light 
will come on, so be cognizant of that, but don't stop when the 
red light comes on if you are trying to finish your statement.
    Thank you all for being here.
    Ms. Williams, you are recognized for 5 minutes.

 STATEMENT OF ORICE WILLIAMS, DIRECTOR, FINANCIAL MARKETS AND 
 COMMUNITY INVESTMENT, UNITED STATES GOVERNMENT ACCOUNTABILITY 
                             OFFICE

    Ms. Williams. Thank you.
    Chairman Watt and members of the subcommittee, I am pleased 
to be here this morning to discuss our report on the Federal 
Reserve Board's Regulation B, which is being publicly released 
today.
    As you know, Regulation B implements the Equal Credit 
Opportunity Act of 1974. Reg B, as it is known, generally 
prohibits lenders from collecting certain data from loan 
applicants such as race or gender for non-mortgage loans, 
including small business or auto loans. Whether or not to 
repeal or amend this regulation has been the subject of debate 
and review for years, including a review in 1998.
    Effective in 2003, while retaining the broad prohibition, 
the Board authorized lenders to collect such data for the 
purposes of a limited self-test to evaluate their compliance 
with ECOA. This morning, I will highlight the findings from our 
report. Specifically, I will touch on three areas: One, 
available research on possible discrimination in non-mortgage 
lending and the data used; two, the Board's 2003 decision to 
retain the prohibition of voluntary collection of personal 
characteristics data; and, three, the benefits and costs of a 
data collection and reporting requirement.
    First, what the research shows: We found that most research 
suggests that while discrimination may play a role in certain 
types of non-mortgage lending, data limitations complicate 
efforts by researchers and regulators to better understand the 
role that discrimination may actually play. For example, the 
research indicates that minority-owned and African-American-
owned small businesses, in particular, are denied loans more 
often or pay higher interest rates than white-owned businesses, 
with similar risk characteristics.
    However, the primary data source for these studies, a 
periodic Board survey of small businesses, while providing 
important insights into possible discrimination, lacks the 
rigor of the Home Mortgage Disclosure Act data, commonly known 
as HMDA data.
    For example, survey data are collected from borrowers, 
rather than lenders, which limit their usefulness as a means to 
assess lending practices across institutions and the industry. 
We also found that in the absence of personal characteristics 
data for non-mortgage loans, Federal bank regulators that 
enforce fair lending laws may rely on time-consuming and less 
reliable approaches to identify possible discrimination, such 
as assuming a loan applicant is Hispanic based on his or her 
last name.
    Next, I would like to discuss the Board's rationale for 
retaining the general prohibition of voluntary data collection 
and reactions to it. We found that while views varied about the 
decision to retain the prohibition of voluntary data 
collection, there was general agreement that such voluntary 
data would have limited benefits.
    We found that the Board's final decision to retain the 
prohibition of voluntary data collection was two-fold. First, 
it said the proposal would have created an opportunity for 
lenders to use the data for discriminatory purposes; and, 
second, voluntary data would not be useful because lenders may 
use different collection approaches. While some researchers and 
others agreed with the Board's first rationale, others said 
that data collection alone would not necessarily create the 
risk for discrimination, because in some cases, such as small 
business lending, lenders may already be aware of an 
applicant's personal characteristics, given that such lending 
is often done face-to-face.
    On the other hand, a range of researchers, regulatory 
staff, and others generally agreed that voluntary data 
collection would not likely materially benefit efforts to 
better understand possible discrimination, because the data 
would be collected on an inconsistent basis. Moreover, few 
lenders, if any, would participate out of concern for 
additional regulatory scrutiny of their non-mortgage lending 
practices and the potential for litigation.
    The last issue I will address involves our analysis of the 
implications of a data collection and reporting requirement. We 
found that while requiring lenders to collect and publicly 
report data on personal characteristics for non-mortgage loan 
applicants could help address many of the current data 
limitations, it could impose additional costs on lenders that 
could be passed on to borrowers.
    While limiting a requirement to certain types of loans, 
such as small business loans, could help mitigate such costs, 
such a requirement may also involve complexities that would 
need to be carefully considered. For example, to the extent 
that small business lending is more complicated than other 
types of lending, lenders may need to collect and report 
additional information on a range of underwriting standards in 
addition to data on personal characteristics, so that informed 
judgments can be made about their lending practices.
    In closing, I would like to note that despite limitations 
with existing data, one key data source, the Survey of Small 
Business Finances, is being discontinued. While the Board plans 
to fold it into the Survey of Consumer Finances, how this 
change will impact researchers who rely on the data remains 
unclear.
    Given the limitations of voluntary data collection, now is 
the time to fully evaluate the implications of a mandatory 
reporting requirement, and this hearing is an important step.
    Thank you, and this concludes my oral comments.
    I would be happy to answer any questions that you may have.
    [The prepared statement of Ms. Williams can be found on 
page 80 of the appendix.]
    Chairman Watt. Thank you very much for the summary of your 
report.
    Ms. Braunstein, you are recognized for 5 minutes, or 
thereabouts.

   STATEMENT OF SANDRA F. BRAUNSTEIN, DIRECTOR, DIVISION OF 
   CONSUMER AND COMMUNITY AFFAIRS, BOARD OF GOVERNORS OF THE 
                     FEDERAL RESERVE SYSTEM

    Ms. Braunstein. Thank you.
    Chairman Watt, Congressman McHenry, and members of the 
subcommittee, I want to thank you for this opportunity to 
discuss whether personal characteristics data collection for 
non-mortgage loans is appropriate and, if so, the best way to 
collect it.
    There are four possible approaches to the collection of 
applicant data: A general prohibition on data collection; 
voluntary data collection; mandatory data collection without 
public disclosure; and mandatory collection with public 
disclosure. It is important to note that the Equal Credit 
Opportunity Act is silent on the question of data collection.
    The first approach, a general prohibition on collecting an 
applicant's personal data, is the approach currently followed 
for non-mortgage loans in Regulation B which implements the 
Equal Credit Opportunity Act, or ECOA. For credit that 
typically is granted using automated underwriting systems 
without face-to-face contact between the creditor and the 
consumer, this approach seems appropriate.
    The second option, voluntary collection of applicant data, 
was considered but rejected by the Board in 2003. Voluntary 
data collection does not appear to be a useful approach. A 
voluntary collection regime would not produce either reliable 
or useful market-wide data.
    Under a voluntary regime, the data would be incomplete, 
because some creditors would elect not to collect. In addition, 
the reliability of the data could not be assured, because the 
data that is collected would be done with different standards 
criteria and methods. Thus, the data would not be comparable 
from creditor to creditor.
    The third regime, mandatory collection of personal data 
without public disclosure, is the approach that the Board 
adopted in 1977 for mortgage loans to obtain information for 
monitoring purposes. This approach can provide supervisory 
agencies with additional data that can be useful in identifying 
possible discriminatory practices, however, many creditors such 
as non-bank finance companies and auto dealers are not subject 
to regular examinations for fair lending compliance. Thus, data 
collection without public disclosure may not enhance fair 
lending enforcement against creditors that are not subject to 
routine oversight.
    Mandatory collection of personal characteristics data with 
public disclosure is the approach used for the Home Mortgage 
Disclosure Act or HMDA. Public disclosure can provide 
heightened scrutiny of lender practices by entities other than 
enforcement agencies. We believe that the availability of the 
HMDA data has led mortgage lenders to review their loan 
decisions more carefully to ensure compliance with their 
lending laws.
    Although an approach which includes public disclosure would 
provide greater transparency than mandatory collection alone, 
it also raises significant public policy choices and cost 
benefit considerations. One fundamental question is the proper 
scope of any mandatory collection, reporting, and public 
disclosure requirement for non-mortgage loans. Should such a 
requirement apply to all non-mortgage loans or some subset of 
those loans such as small business loans? A requirement to 
collect, report, and publicly disclose race, ethnicity, and 
gender data for lending other than mortgages, such as small 
business, may promote fair lending enforcement.
    However, such a requirement would be challenging to 
implement and could impose significant costs on lenders. Small 
business lending is quite complex and variable. For example, 
there are many different types of small business lending, 
including credit lines, business credit cards, vehicle and 
equipment loans, mortgages, capital leases, and trade credit. 
There are also many different types of small business lenders, 
including banks, credit card companies, finance companies, and 
trade creditors. Many different types of data about business 
attributes and underwriting standards would have to be 
collected for the data to be useful as a screen for fair 
lending enforcement purposes.
    The Board is committed to addressing racial and ethnic gaps 
in the availability and affordability of credit. However, just 
as Congress required the collection, reporting, and disclosure 
of personal data in HMDA for mortgage loans, we believe that 
Congress is in the best position to make these important public 
policy and cost-benefit determinations for non-mortgage loans.
    Thank you, and I will be happy to answer any questions from 
the subcommittee.
    [The prepared statement of Ms. Braunstein can be found on 
page 44 of the appendix.]
    Chairman Watt. Thank you very much, both of you, for your 
comprehensive statements.
    Let me do a little housekeeping here for the purpose of 
making sure that all the members have a full appreciation of 
what we are trying to develop here. This has actually been 
going on for a while and I don't want the members to miss what 
has led up to today's hearing. So let me ask unanimous consent 
to submit the following documents for the record:
    A letter dated February 12, 2004, from Representative Frank 
and 22 Members of Congress to Federal Reserve Chairman 
Greenspan; a letter dated March 8, 2004, which is the response 
of Chairman Greenspan to Representative Frank's letter;
    A letter dated February 14, 2007, from Representatives 
Frank, Watt, Waters, Gutierrez, and Lee to Chairman Bernanke; a 
letter dated March 13, 2007, which is the response of Chairman 
Bernanke to the February 14th letter;
    A letter in preparation for today's hearing from Chairman 
Watt to the FDIC, the OTS, the OCC, the NCUA, the SBA, the 
Department of Justice, and the FTC asking for their response to 
a series of questions about the desirability of collecting this 
kind of data and the contents of that letter will speak for 
itself;
    Responses to the July 10th letter from the FDIC, the OTS, 
the OCC, NCUA, the SBA, and the FTC--we have not yet received 
the response from the Department of Justice and we are still 
trying to get that;
    And a letter dated July 16th to Chairman Watt from the 
National Consumer Law Center stating its views.
    Without objection, all of those items will be submitted for 
the record.
    I have previously asked for unanimous consent to submit the 
letter dated July 16, 2007, from Representatives Frank, Watt, 
and Maloney to the GAO which requested today's report. So I am 
just trying to make sure that everybody is aware that this 
hearing today is another step in a sequence of things that has 
occurred on this issue, as Ms. Williams has indicated. This 
issue has been around for a while and we need to try to address 
it. Now let me proceed with the questioning of these witnesses, 
and I will recognize myself for 5 minutes for the first 
questions.
    Ms. Williams, I take it that just about everybody has 
concluded, based on your report, that voluntary reporting and 
collection of collection and/or reporting of race, gender, and 
personal characteristics data is probably not worthwhile, 
because of various issues that you identified in your 
statement.
    You would get all kinds of different responses if it were 
voluntary as opposed to setting up a set series of things that 
people will be expected to respond to. Is that correct?
    Ms. Williams. That is correct.
    Chairman Watt. So the options we are looking at, it seems 
to me, are mandatory: either no collection, or mandatory 
collection, with us or somebody, perhaps the Fed, giving 
guidance about how that information would be structured.
    That is a fair summary of where you got to?
    Ms. Williams. Yes.
    Chairman Watt. And, Ms. Braunstein, I take it that you 
reached kind of the same conclusion. The Fed has reached the 
same conclusion, and you basically said that there would be 
costs associated with collecting the data, but that there would 
be benefits flowing from the collection of data, and it is the 
Congress' prerogative to determine whether to collect it or 
not.
    It is not the Fed's prerogative to make that determination?
    Ms. Braunstein. Correct.
    Chairman Watt. So you have basically thrown the ball back 
to us. Okay. That frames our hearing today, because it seems to 
me that Congress, at the end of whatever series of hearings we 
have on this, has to be evaluating the benefits of collecting 
the data, the detriments of collecting the data, and the cost 
of collecting the data.
    Would that be a fair assessment of what you would think 
would be the appropriate inquiries that we would be making, Ms. 
Williams?
    Ms. Williams. Yes, I think that is right.
    Chairman Watt. Okay. Both of you made some reference to 
cost, so let me get you to elaborate, if you can. You described 
some general problems associated with collecting data, but has 
anybody done any specific looks at what the actual cost would 
be, Ms. Braunstein?
    Ms. Braunstein. Yes. The best frame of reference that we 
have, of course, is the HMDA system; and one of the benefits of 
HMDA is that the Congress did set out in a statute a very 
specific framework for the collection, the public reporting of 
that data, and it involves a lot of entities. It involves costs 
on the part of the lenders, certainly, who are doing the 
collection and the reporting. It involves a lot of costs on the 
part of the supervisory agencies who collect that information 
from the lenders and then analyze it and report it back out in 
public.
    So that is our best frame of reference. I don't have 
numbers on that, but certainly, we would be willing to come up 
and talk to you and your staff, or whomever, about what costs 
are associated with HMDA. I would say they are pretty 
significant and we have a lot of resources ourselves dedicated 
to this process. Additionally, we think that collection of 
something like small business data would be even more complex 
than mortgage data, because of the nature of the loans.
    The products in small business lending are not nearly as 
homogenous or standardized as mortgage lending, so you are 
talking about a lot of different variables. So I think it would 
be the HMDA cost, probably plus an additional factor to put 
into place a robust system that would be valuable for people.
    Chairman Watt. Okay, before my time runs out, let's compare 
the cost of collecting HMDA data to the benefits of collecting 
HMDA data.
    Does the Fed have a position on that?
    Ms. Braunstein. Yes, as I have said in my testimony, we do 
think that HMDA data has been beneficial in terms of fair 
lending enforcement and in terms of providing information in 
the public that give people an understanding of mortgage 
lending. It is certainly not determinative of discrimination, 
but it is a very useful screening tool for us in our fair 
lending examinations.
    Chairman Watt. I ask unanimous consent for one additional 
minute. Would it be correct to say that the collection and 
public reporting of this data also has been a deterrent to or 
has deterred people, lenders, from engaging in discriminatory 
practices?
    Ms. Braunstein. I think that is fair to say.
    Chairman Watt. Okay. How do you square that with the 
notion, then, that you all have had that somehow collecting 
this data in non-mortgage loans would or could run the risk of 
encouraging people to discriminate?
    Ms. Braunstein. That statement was made in the context of 
voluntary collection not publicly reported. That context was 
made regarding just lifting the prohibition, which would have 
led lenders to ask people for this data without anybody 
checking it. Many of the lenders involved do not get regular 
examinations from supervisory authorities, so we would have no 
way of knowing if they were using it for bad purposes. That was 
in that context, not in the context of a public system.
    Chairman Watt. Okay. And, Ms. Williams, finally, to what 
extent did you all do any analysis of what the actual cost 
might be of collecting this kind of data in non-mortgage 
situations?
    Ms. Williams. Given that there's no current structure in 
place, the approach we took to collect what the cost would be 
was largely through conversations with regulators, HMDA 
experience, and also talking to lenders and focusing where the 
cost would be impacted.
    Chairman Watt. And is that fully reported in your report?
    Ms. Williams. It is.
    Chairman Watt. All right. I thank you.
    My time has expired, and I will recognize Mr. McHenry for 5 
minutes for his questions.
    Mr. McHenry. Thank you, Mr. Chairman. And thank you all for 
your testimony.
    You know, I would like to know from you, Ms. Braunstein, 
the Federal Reserve had a prohibition for a few decades, 
roughly, on collection of race, gender, and ethnicity, and then 
you went to this voluntary method.
    What was the thought process to move to the voluntary 
method of collection?
    Ms. Braunstein. Actually, let me clarify that. We did not 
move to voluntary collection. We proposed it and then we 
withdrew that application.
    Mr. McHenry. What was the thought process behind the 
proposal?
    Ms. Braunstein. The proposal, what we ended up with, was we 
have a little window there where lenders can collect it but 
only for purposes for doing their own self-testing for fair 
lending. But otherwise, there still is a prohibition in place 
on collecting this data.
    Mr. McHenry. It seems sort of bizarre, because in your 
testimony you said, and in answering Chairman Watt's question 
as well, that you have some concerns about ensuring that this 
data is not used for other purposes if that data is requested, 
if that is the intent of Congress and the Federal Reserve.
    So how do you allow people to voluntarily do this as a 
self-check? I mean, it seems bizarre to me that you would say 
if you collect this data you could use it for ill intent, but 
at the same time, we are going to let you voluntarily do that 
so you can self-check. It just seems sort of odd.
    Ms. Braunstein. Well, there are self-testing provisions in 
the ECOA, so that is a normal state of affairs where banks do 
their own mystery shopping and testing to make sure that their 
policies are being carried out. And so we did say that if this 
was a useful tool for them, they could do that. I will add that 
our understanding in talking to lenders is that hardly anybody 
is doing it. For many reasons they are not collecting this 
data, so I don't think anybody has taken us up on that. But we 
did allow that window.
    Mr. McHenry. Could you collect an analysis of the cost of 
collecting HMDA data, and could you submit that to us?
    Ms. Braunstein. We could put together figures, probably, on 
HMDA data. Yes, I would think that we could follow up with you.
    Mr. McHenry. Okay, just for the subcommittee to have that 
information, both the regulatory, the governmental portion of 
the cost and the private sector as well.
    Ms. Braunstein. Right. Obviously, we would have to estimate 
the private sector but we will do that.
    Mr. McHenry. Ms. Williams, you spoke generally of the cost. 
Can you go into some more detail about the cost of possibly 
implementing this?
    Ms. Williams. Our approach on this particular issue, we 
weren't able to collect specific costs for institutions, 
because there is no rule in place. So the cost would be 
dictated by the specifics of a rule if it were in place in 
terms of what information had to be kept, what was collected, 
how it was maintained, and if it was publicly reported.
    We focused our conversations with lenders and regulators in 
terms of what are the broad categories that the cost would be 
incurred in, information systems, training employees, expanding 
their technological capabilities. And views varied about how 
much it would actually cost. Would it be extremely expensive 
because all of these systems would have to be created? Would 
they be able to build on existing systems as well as, for 
example, small business lending may not occur?
    Mr. McHenry. What was your conclusion?
    Ms. Williams. Our conclusion was that there is a potential 
for increased cost. No specific dollar figure was provided.
    Mr. McHenry. Okay. Just more?
    Ms. Williams. More.
    Mr. McHenry. So perhaps we need to have another study with 
some parameters on what it would actually cost, but we would 
have to give you some specifics. Is that what you are saying?
    Ms. Williams. Yes. And historically, when we have attempted 
to quantify costs of a particular change, it is extremely 
difficult because the specifics aren't there. And also it is 
difficult to confine how the lenders go about attributing cost 
to a specific activity. Sometimes, they will lump in all the 
cost of information systems to a change without backing out the 
fact that there are certain things that they have to collect 
regardless of the change in the regulation or not.
    Mr. McHenry. Okay. Ms. Braunstein, sampling: Is there a 
possibility the Federal Reserve could do sampling in order to 
determine in this broad market of lending, you know, the racial 
breakdown, race, ethnicity, and gender breakdown in lending?
    Ms. Braunstein. The problem is that we don't have the 
contact with the borrowers. So right now there is a prohibition 
in effect for the lenders to collect that information.
    Mr. McHenry. But it is a Federal Reserve prohibition.
    Ms. Braunstein. So there were wouldn't be the implication 
that it is a sample. Do you see what I mean? The information 
would come to us. Normally, like in HMDA, the information is 
collected by the lenders, and then it comes to us by the 
lenders. Right now, the lenders can't collect this information.
    Mr. McHenry. Yes, because of your prohibition. So you are 
saying the Federal Reserve can't do it because the Federal 
Reserve prevents it from being done. So my question is, is 
there the possibility that you could through the lenders do 
sampling that is statistically sound?
    Ms. Braunstein. We would have to remove the prohibition in 
order to do that.
    Mr. McHenry. Actually, if you removed the prohibition, 
could you do that in a statistically sound way?
    Ms. Braunstein. Yes, but the concern would be that without 
putting into place a framework for collection sampling, you 
would have similar problems to voluntary collection in that 
lenders would do it differently in different cases, and you 
would have apples and oranges. So, once again, we are talking 
about the need for a framework or system so that there is 
consistent data collected. Otherwise, it is not useful.
    Chairman Watt. Thank you. The gentleman from Missouri is 
recognized for 5 minutes.
    Mr. Cleaver. Thank you, Mr. Chairman.
    Ms. Williams, on page 9 in your report, you mention in the 
first paragraph that the four Federal bank regulatory agencies 
said that the availability of HMDA data has facilitated the 
fair lending law examination process. How? Can you explain 
that?
    Ms. Williams. Yes, the existence of that data allows 
regulators to have information on a specific lender as well as 
industry-wide so they can look at the HMDA data as a baseline 
and determine if they need to look at a particular lending 
institution or certain aspect of the industry while enforcing 
their fair lending laws so they have demographic data.
    Mr. Cleaver. Okay, that brings about Ms. Braunstein, does 
the Fed share the HMDA data with any other Federal agencies?
    Ms. Braunstein. Yes, all the agencies. Well, it is publicly 
available to everybody, to the--
    Mr. Cleaver. Public.
    Ms. Braunstein. Public, yes. So, yes, it is publicly 
available. And the other agencies, the regulatory agencies, all 
use it in their fair lending examinations.
    Mr. Cleaver. This data is available, but not necessarily 
sent to State banking?
    Ms. Braunstein. Actually, we do send HMDA data every year 
to the State banking agencies.
    Mr. Cleaver. How long has that been a practice? Forever?
    Ms. Braunstein. Well, I don't know about forever, but it is 
been quite a while.
    Mr. Cleaver. In your testimony, you discuss how many 
institutions have been referred to the Department of Justice. 
Do you have any data on how those cases were disposed with the 
Department of Justice? What happened?
    I mean, can you say 12 institutions were taken to Federal 
court?
    Ms. Braunstein. I do not have that information. Certainly, 
we receive information on how those cases are handled, and, 
there are a number of ways that they are. Sometimes, there are 
settlements between Justice and the lenders. Sometimes, the 
cases are sent back to us for disposition. I don't have those 
statistics today, but we could certainly get you that 
information.
    Mr. Cleaver. Yes, I am very much interested in it, because 
you know if there are violations, I think it would help the 
committee to know what the Department of Justice is doing. I 
mean, you know, and are we getting a letter from DOJ, saying 
you have been a bad person or are there penalties that would 
discourage others. So I think that information would be very 
helpful.
    That is all, Mr. Chairman.
    Chairman Watt. I thank the gentleman, and the gentleman 
from Texas, Mr. Green, is recognized.
    Mr. Green. Thank you, Mr. Chairman, and I thank the 
witnesses for your testimony. It has been very enlightening.
    Sometimes, we can make things a little bit more complicated 
than they actually are. Dr. King spoke of the paralysis of 
analysis, that you can take the simplest thing and analyze it 
to the point that you do absolutely nothing. So let me move 
quickly to what I say is a bottom line, notwithstanding the 
cost to collect and, by the way, that cost cannot only be the 
dollar amount to collect, but it could also be the cost on 
society for failure to collect the empirical evidence 
necessary.
    You can have a cost that is associated with actually the 
physical process of collecting information. But then, if you 
don't collect, society has a cost. It is the cost of dreams not 
being fulfilled, the cost of homes not being purchased, the 
cost of wealth building not taking place.
    So there are various and sundry costs and we ought not just 
look at the dollar amount associated with the actual collection 
of data.
    Next point, notwithstanding the complications associated 
with data collection, I understand that it can be difficult. 
But I also understand that we collect an enormous amount of 
intelligence on things, and we do it most effectively and we do 
it consistently. And we do it in such a way that the 
information can be used for whatever the stated purpose is.
    So I ask you this: Do you agree that we can construct an 
effective data collection system if we have the will to do so? 
Can it be done? I would like to start with Ms. Williams.
    Ms. Williams. I think based on the experience with HMDA, 
HMDA is an important precedent to consider. So I think it is 
within the realm of possibility.
    Mr. Green. Okay. Ms. Braunstein?
    Ms. Braunstein. Yes, certainly, it could be done. Again, it 
is a cost benefit analysis.
    Mr. Green. I understand. Well, cost benefit analysis, let's 
just take that off the table just for the purpose of our 
discussion now. Assuming that we conclude that the benefits 
outweigh whatever the costs are, can a system be devised so 
that we can acquire empirical evidence indicating whether or 
not discrimination has taken place?
    Ms. Braunstein. I don't believe as with HMDA that we would 
ever be able to construct a system for small business or other 
data where it was definitive. On discrimination, I think that 
the HMDA system is an excellent screening tool and we use it 
that way. And, we could also build a system, like you say, if 
there was the will and people felt that the benefits were worth 
it.
    We could build a system for small business lending or other 
kinds of lending that would also be a screening tool, but it 
would not be definitive on discrimination.
    Mr. Green. Let me ask you this. If we have a system that 
allows us to ascertain who is involved in the process, whom it 
is that is involved in the process; and, if we can determine 
what the outcome is of the persons involved in the process, are 
you indicating that we don't have the intellect to put a system 
in place that will give us an opportunity to analyze data and 
come to conclusions?
    Ms. Braunstein. No. I'm not saying we don't have the 
intellect. I'm saying that the lending process as with HMDA is 
so complex that you need to go beyond what is reported on a 
data sheet that there would be almost an impossibility to 
collect every single factor that would go into a credit 
decision.
    That is what I am saying. I am not saying we don't have the 
intellect. I am saying that you still would need to do 
investigations into files and other policies of the institution 
and a lot of other things in order to definitively--
    Mr. Green. Would we be in a position, Ms. Braunstein, to 
collect enough information such that we can do the follow-up 
investigations that would say to the person making the inquiry, 
you need to do this follow-up investigation?
    Ms. Braunstein. Absolutely.
    Mr. Green. Could we do that?
    Ms. Braunstein. Yes, that is what I was saying. It is a 
screening tool, yes.
    Mr. Green. All right. And upon moving forward with the 
follow-up investigation, then couple that intelligence with the 
intelligence acquired initially. Then can we start to draw the 
conclusions?
    Ms. Braunstein. Yes.
    Mr. Green. Here is a point, friends. Really now, here we 
are in 2008, and we are debating whether or not it is 
appropriate to ascertain, to put in place a proper system to 
ascertain whether discrimination is taking place.
    Now, there are so many people who can tell you that it is 
taking place and you can always dispute it, but the point is 
that at some point in America, we ought to just get on with it 
and stop making excuses for what we know to be something that 
can be done. We can put this system in place if we so choose. 
It is a question of will, not way.
    Do you differ with me, Ms. Williams?
    Could you say yes or no please?
    Ms. Williams. No.
    Mr. Green. Do you differ with me, Ms. Braunstein?
    Ms. Braunstein. No.
    Mr. Green. All right. Mr. Chairman, I yield back, and I 
just beg that we get on with the business of eliminating 
invidious discrimination. It really is time.
    Thank you.
    Chairman Watt. I thank the gentleman for his questions. I 
ask unanimous consent to ask a few more questions, just to be 
clear on a couple of things. So let me ask unanimous consent 
for 3 additional minutes.
    Ms. Braunstein, it is not clear to me whether I understand 
that you think as a matter of policy it would be better for 
Congress to make the decision whether to construct a system to 
collect this data in non-mortgage situations.
    Is it also the Federal Reserve's position that you don't 
have the authority to do it now without congressional 
legislation?
    Ms. Braunstein. We are still looking at that issue. We 
think that it may be unclear as to whether we would have the 
authority to put into place the kind of framework that exists 
with HMDA, which would involve the other regulatory agencies. 
It involves more than just telling banks to collect data.
    Chairman Watt. So, clearly, you would prefer us to do it 
than you all assume that responsibility?
    Ms. Braunstein. Yes, that is a decision for Congress. Yes.
    Chairman Watt. Okay.
    Just in follow-up to Mr. Green's questions, I would like to 
submit for the record the Federal Reserve Bank of Boston's 
report on credit card redlining in which it is fairly 
unequivocal in its assessment. Credit cards and the 
availability of credit cards is one form of credit that we are 
talking about. It is pretty specific in its observations.
    I will just read one or two sentences that says the paper's 
principal observation is that remarkably in spite of identical 
scores and identical community characteristics, an individual 
in the black neighborhood receives less consumer credit than 
the individual than the white area.
    That is in spite of the fact that both have been assessed 
to have been similar risk of non-payment as determined by the 
credit score, the person living in the black area has less 
ability to access credit.
    That is the Federal Reserve Bank of Boston. That is part of 
your operation, isn't it?
    Ms. Braunstein. Yes, it is.
    Chairman Watt. Okay. And the Federal Reserve Bank of 
Chicago, that is your operation, too?
    Ms. Braunstein. Correct.
    Chairman Watt. Going all the way back to 1999, and I am 
going to ask unanimous consent to put in a letter dated 
November 17, 1999, from the Federal Reserve Bank of Chicago and 
the attachments to that letter where it is pretty aggressive, 
and its position about whether we ought to be mandatorily 
setting up the system, too.
    So just to complete the record, I am just trying to make 
sure we have a full record on this. I noticed that our 
illustrious chairman of the full committee has come in and I 
would be honored to yield to him as much time as he may 
consume.
    The Chairman. I thank the chairman.
    If I can make a statement, it was fortuitous that I came 
here. We are talking about the good work that was done by the 
Federal Reserve Bank of Boston and there is also the University 
of Massachusetts, Boston Center, that studies this. And, of 
course, my colleague Mr. Lynch is also here. We have learned a 
lot about this in Boston.
    I would just say with regard to collecting this data, I 
have seen this movie before, and it has a happy ending. In the 
late 1980's or early 1990's, my then-colleague or former Member 
of the House, Joe Kennedy, led the fight for the Home Mortgage 
Disclosure Act data, and it was very controversial. And in 
fact, it lost in this committee and was then overturned on the 
Floor. And a former close ally of Newt Gingrich, a former 
Member from Pennsylvania, Bob Walker, took the Floor on a very 
forceful speech and helped us overturn the negative vote; he 
supported it and said, ``No.''
    If this country isn't going to confront racial 
discrimination, which we all know continues to be a factor, 
then we are failing ourselves; and, there was a good deal of 
negative argument there and there were predictions that 
collecting the HMDA data on race and gender would be terribly 
disruptive.
    Today, I think there are very few people who aren't very 
glad that we have it. It has become the common currency, the 
data that has been very important for a lot of discussions; and 
I have been to presentations in the real estate industry and 
others who are so glad to have this data. I just wish people 
would look at that example, because we had many of the same 
concerns when we decided to do this with regard to HMDA.
    Joe Kennedy took the lead and was under the chairmanship, 
of course, of Henry B. Gonzalez, a great fighter for fairness, 
and it worked very well. I believe it would work very well 
here, too. So I apologize for coming in and out of the hearing, 
but we are dealing with the housing bill and that is taking 
some of my time. But this is a very important hearing. I 
appreciate the chairman of the subcommittee devoting time to 
it, and I just want to say again, I hope people will look to 
the example that we had before and how helpful it has been. I 
think that this hearing will be helpful to us in going forward 
as well.
    I thank you, Mr. Chairman.
    Chairman Watt. I thank the chairman for being here, and I 
would just say to the chairman that I think there will not be 
put together a more comprehensive record than we are trying to 
put together in this subcommittee to support whatever public 
policy the Congress decides to take on this.
    We want to try to be fair. We hope to have a follow-up 
hearing on the cost aspects of it, but I think we have this 
issue framed pretty well.
    The Chairman. If the gentleman would yield further, when 
you say I appreciate it, it is reinforcing that, because 
obviously we are getting late in this year. But this particular 
issue will be, unless things develop in November very 
differently than I think most of us now expect, I can guarantee 
people that this will be very high on this committee's agenda 
in 2009.
    Chairman Watt. I am delighted to yield to the other 
gentleman from Massachusetts, Mr. Lynch, a member of the 
subcommittee, for as much time as he may consume.
    Mr. Lynch. Thank you, Mr. Chairman.
    I appreciate that and I associate myself with the remarks 
of the full committee chairman, as well.
    Ms. Braunstein, I don't want to put you on the spot here. I 
know that as far as the GAO report that is central to our 
discussion this morning, the FRB, the Federal Reserve Board, 
did not take an official position regarding the GAO report. 
However, I would like to really get down to the essence of this 
in a couple of sentences. And I just want to ask you about this 
partially, if you would be so kind.
    The report states in the third paragraph. It says that, 
``Requiring lenders to collect and publicly report data on 
personal characteristics for non-mortgage loan applicants could 
help address current data limitations that complicate efforts 
to better assess possible discrimination.'' It does go on to 
raise concerns consistent with the chairman's remarks about 
costs and how to do this. But just on that basic assessment, 
and I think it is pretty powerful, I want to know your own 
thoughts on this and your own observations from your position.
    Is that something that could be done here fairly accurately 
and with a minimal cost being generated?
    Ms. Braunstein. First of all, we agree with the position 
that if there was a good data collection, it could be helpful. 
As to whether it could be done with minimal costs, we are not 
sure. I mean, as I have said before, our best frame of 
reference is the HMDA data, and the HMDA data is an excellent 
system, and everybody benefits from it. We believe that 
wholeheartedly, but it is not without significant costs. So I 
think again it is a cost-benefit decision that needs to be made 
here.
    Mr. Lynch. Yes. Ms. Williams?
    Ms. Williams. Our position is that it could address many of 
the limitations that exist in terms of fully enforcing all 
aspects of the fair lending laws and the costs have to be 
weighed against that. But we do raise the fact that the survey 
of small business finances that is currently used as the data 
source by researchers that do research in this area and have 
identified issues will be going away.
    So now really is the time to evaluate given that important 
data source will be going away and how it is going to be 
replaced is unclear in terms of deciding whether additional 
information will be helpful. And I think in terms of enforcing 
fair lending laws, it is critical that information be available 
to, at a minimum, the regulators.
    Mr. Lynch. Yes, I agree; and, maybe I am just providing my 
own testimony, but based on what I have read here and some of 
the reports provided by the additional witnesses, it would 
appear, just as you say, Ms. Williams, the timing here is very, 
very critical. And, also, the likely benefit in my opinion of 
getting this additional data so that we can more accurately 
measure and address the discrimination that remains I think 
outweighs.
    We will have to do this carefully, precisely, and 
accurately, working with the lending community. But I 
definitely think this is something that is tremendously 
worthwhile and beneficial to all of us, and I think the 
challenge will be just that--how to do it in a cost-effective 
way.
    But again, I want to thank you both for your testimony and 
I yield back the balance of my time.
    Chairman Watt. I also want to express my thanks to these 
two witnesses for framing this issue and giving us the context 
for evaluating it; and, particularly, I thank Ms. Williams of 
the Government Accountability Office for the excellent report 
that your Office has generated in response to our request.
    So with that, this panel is excused, and we will ask the 
next panel of witnesses to come forward.
    While everybody is getting seated, and in hopes that they 
won't realize how much I butchered their names--I think my 
staff has made it difficult by giving me a bunch of witnesses 
with complicated names to pronounce. So let me just say that 
without objection, your full bios will be made a part of the 
record. We will not engage in long introductions in the 
interest of time.
    Our first witness will be Dr. Ken Cavalluzzo, Wisconsin 
Capital Management LLC. Our second witness will be Mr. Robert 
F. Gnaizda, general counsel, The Greenlining Institute. Our 
third witness will be Mr. Bill Himpler, executive vice 
president, Federal Affairs, American Financial Services 
Association. Our fourth witness will be Mr. Jorge Corralejo, 
chairman of the Latino Chamber of Commerce of Greater Los 
Angeles. And our final witness will be Ms. Ann Sullivan, the 
president of Madison Services Group, on behalf of Women 
Impacting Public Policy.
    Without objection, your entire written statements will be 
made a part of the record, and each of you will be recognized 
for a 5-minute summary of your testimony.
    Dr. Cavalluzzo, we now recognize you for 5 minutes.

   STATEMENT OF KEN CAVALLUZZO, RESEARCH ANALYST, WISCONSIN 
                     CAPITAL MANAGEMENT LLC

    Mr. Cavalluzzo. Thank you.
    Mr. Chairman and other distinguished members of the 
subcommittee, my name is Ken Cavalluzzo. I am a research 
analyst at Wisconsin Capital Management, an investment firm 
located in Madison, Wisconsin.
    Thank you for the opportunity to speak today about whether 
lenders should be required to collect race and gender data of 
borrowers for all loans. My testimony today is based on work I 
started as a graduate student at the Wharton School of the 
University of Pennsylvania where I received my doctorate, and 
continued while on the faculty of Georgetown University's 
McDonough School of Business.
    My research was conducted in collaboration with Dr. Linda 
Cavalluzzo, a senior economist at CNA, and Dr. John Wolken, a 
senior economist at the Board of Governors of the Federal 
Reserve. Our research on this topic has been published in 
leading, peer-reviewed, academic journals sponsored by the 
University of Chicago and Ohio State University.
    The common finding across all our research is that black-
owned firms were denied credit at higher rates than white-owned 
firms. Even after controlling for relevant risk 
characteristics, black-owned firms were denied credit at almost 
twice the rate of white-owned firms. The differences are 
economically meaningful and statistically significant.
    We found that black and Hispanic firms were significantly 
more likely not to have applied for credit for fear of being 
turned down; and, we found that blacks, Hispanics, and Asians 
were significantly more likely to have unmet credit needs than 
were firms owned by white males. Our work is based on data 
obtained from the borrower, that have strengths and limitations 
relative to lender data. These limitations together with the 
differences we document on credit access among demographic 
groups strongly point to the need to collect both borrower and 
lender data.
    I believe that data should be collected from the lender on 
the personal characteristics of borrowers of non-mortgage 
credit. I do not believe that collecting such data would be 
particularly onerous or costly to lending institutions. Such 
collection would not materially heighten the likelihood of 
discrimination.
    According to our data, 78 percent of small business loans 
are on a face-to-face basis, and small business lending tends 
to be relationship-based. So the opportunity to discriminate on 
personal characteristics already exists, even in the absence of 
collection of such data.
    Collecting personal characteristics data would likely 
benefit regulators, lenders that do not discriminate, and 
borrowers. Data collection is an important step towards 
ensuring equal treatment in lending to minority-owned small 
businesses.
    I recommend that collection of personal characteristics 
data from lenders be mandatory. Data provided by volunteers are 
unlikely to be representative of behavior across the industry. 
The data should be collected and at a minimum reported to the 
appropriate Federal banking regulator. Data reporting is a more 
expensive activity than data collection, yet such costs may be 
limited to the degree that banks already codify their data. But 
I have an important caveat.
    Reporting personal characteristics data without reporting 
corresponding information on risk characteristics is a fairly 
meaningless and potentially dangerous exercise, as such 
disclosure could unfairly characterize some banks as engaging 
in discrimination. Given the importance of financial 
institutions to the funding of small businesses, to the 
important roles small businesses play in the U.S. economy, and 
the wide differences in acceptance rates found between black- 
and white-owned small businesses, I encourage Congress to 
consider mandating that all key information to the application 
and pricing decision be made public.
    In this spirit, Congress should revisit the Federal Reserve 
Board's decision to discontinue the small business data series, 
which provided data from the borrower. The Federal Reserve no 
longer collects small business data from the borrower, nor does 
the Federal Reserve collect such data from the lender. Plain 
and simple, it is difficult to learn anything without data.
    Markets would function better if applicants knew the 
information that went into the underwriting decision. 
Regulators could regulate better. Borrowers could become better 
borrowers. Lenders would probably become better lenders. 
Obviously, reporting data is more costly than simply collecting 
it; yet, given the advancements and technological developments 
for data gathering, reporting, and analyzing, historically, it 
has never been cheaper.
    I appreciate the opportunity to testify and I look forward 
to any additional questions you may have.
    [The prepared statement of Dr. Cavalluzzo can be found on 
page 55 of the appendix.]
    Chairman Watt. Thank you very much for your testimony.
    Mr. ``Gnaizda?''
    Mr. Gnaizda. ``Gnaizda.''
    Chairman Watt. I have captured your name already, so Mr. 
Gnaizda, you are recognized for 5 minutes.

     STATEMENT OF ROBERT F. GNAIZDA, GENERAL COUNSEL, THE 
                     GREENLINING INSTITUTE

    Mr. Gnaizda. Good morning, Mr. Chairman, and members of the 
subcommittee.
    Greenlining believes that this is a unique opportunity, the 
recommendations of the GAO study. It is a unique opportunity 
for a major economic stimulus far beyond any stimulus so far 
approved by this Congress, and it is a great potential for job 
creation.
    Two-thirds of all new jobs in this country come from small 
businesses. Forty-five percent of small businesses are women or 
minority-owned. There are approximately 13 million in total. 
Greenlining members include the U.S. Hispanic Chamber, the 
California Hispanic Chamber, the California Black Chambers, and 
the Asian Business Associations. All believe that the GAO study 
was long overdue.
    We have a 15-year history of addressing this matter. 
Congresswoman Maxine Waters was a leader in 1993 in supporting 
a revision for mandatory reporting. Greenlining members visited 
with Chairman Greenspan on at least 10 occasions over the last 
15 years on this matter, and with Chairman Bernanke on three 
occasions since he became chairman. We fully support the GAO 
study.
    We believe that Congressman Green's analysis is exactly 
correct. We can paralyze ourselves with over-analysis. We have 
three recommendations. The first is that the Federal Reserve 
develop a task force and all those lenders who failed to 
respond to this committee be part of that task force to develop 
common metrics and discuss costs if they wish.
    That should be done by September. The minority business 
associations will join in if requested. We believe the Federal 
Reserve should immediately eliminate its prohibitions and allow 
any financial institution that wishes to gather the data on a 
voluntary basis; and, we believe by no later than 2011 and 
hopefully by 2010 there will be mandatory reporting, and the 
reporting of course should be public. This is no different than 
HMDA and no different than SBA lending.
    These are the benefits: Number one, transparency; number 
two, this is fully consistent with Chairman Bernanke's view 
that the Federal Reserve is often operating with inadequate 
data and needs more data--he has reiterated this on many 
occasions; third, we believe this is a multi-billion-dollar 
economic stimulus and will create hundreds of thousands if not 
millions of new jobs over the next few years; fourth, we think 
it will discourage discrimination; fifth, we have the Wells 
Fargo model, and I am sorry that they declined to testify. They 
have side-stepped the Federal Reserve's rule and courageously 
with the assistance of the Latino business community, the black 
business and the Asian business community developed multi-
billion-dollar goals for small business lending for women- and 
minority-owned businesses and have achieved all of their goals, 
in many cases exceeded them.
    Lastly, something that has not been discussed is the 
effective marketing opportunities for financial institutions. 
There is no way to effectively market to the 13 million 
minority- and women-owned businesses unless you can get data on 
it and determine how successful you are. So banks are losing 
multi-billions in opportunities.
    We will be meeting with the Federal Reserve on Friday to 
discuss this task force. We would welcome any financial 
institutions joining us, and we will be meeting with Sheila 
Bair on Friday. And this afternoon we will be meeting with Mr. 
Ryan, the Undersecretary on this matter, as well as with OCC 
and OTS; and we have scheduled a meeting for November 18th with 
Chairman Bernanke and the minority business associations.
    We are open, and I am particularly open, to any questions 
regarding the Federal Reserve and discrimination or their 
contentions of it, and are open to any questions on costs, 
because we don't agree on the cost analysis by the Federal 
Reserve. And we agree with you, Congressman McHenry. There are 
many unintended consequences of not gathering this data; and, 
we agree of course with Congressman Green's position that this 
is long overdue.
    So thank you, Mr. Chairman, and members of this committee.
    [The prepared statement of Mr. Gnaizda can be found on page 
67 of the appendix.]
    Chairman Watt. Thank you very much.
    Mr. Himpler, you are recognized for your opening statement.

STATEMENT OF BILL HIMPLER, EXECUTIVE VICE PRESIDENT OF FEDERAL 
    AFFAIRS, AMERICAN FINANCIAL SERVICES ASSOCIATION (AFSA)

    Mr. Himpler. Thank you, Mr. Chairman, and, by the way, you 
pronounced my name exactly correct.
    [Laughter]
    Chairman Watt. You and Ms. Sullivan get rewards for having 
simpler-to-pronounce names.
    Mr. Himpler. It is a pleasure to be here this morning with 
you, acting Ranking Member McHenry, and the other members of 
this subcommittee.
    I am the executive vice president for the American 
Financial Services Association, AFSA. AFSA's 350 members 
include finance companies that lend to consumers and small 
businesses.
    Mr. Chairman, I commend you and your colleagues for holding 
this hearing. We recognize the importance of ensuring that all 
persons have equal access to credit and are committed to 
eliminating discrimination in lending.
    We believe that ECOA and Regulation B contain the necessary 
restrictions and enforcement tools to end discrimination and we 
do not believe that access to affordable credit will be 
enhanced by requiring non-mortgage creditors to collect race 
and gender data. On the contrary, imposing data collection 
obligations may decrease credit options available and will 
increase the cost of credit for consumers and creditors alike.
    While both government and industry strive to make credit 
application processes as color blind as possible, we believe 
that the proposed requirement being discussed today goes 
against this goal. Reg B currently prohibits creditors from 
collecting information about the applicant's personal 
characteristics including race and gender information in 
connection with non-mortgage credit. This prohibition ensures 
that decisions in non-face-to-face transactions are race 
neutral.
    For example, in the indirect finance situation, an auto 
finance company makes a decision about whether or not to 
purchase a retail installment sales contract based on the 
applicant's credit worthiness, not his or her race. The 
decision is race neutral, because the finance company does not 
typically have contact with the applicant and therefore does 
not have race information. There is scant statistical evidence 
to demonstrate that race or gender plays a role in access to or 
cost of non-mortgage credit.
    Rather, studies suggest that credit scores and related risk 
factors determine access to credit and the cost of credit. The 
Federal Reserve Board conducted a study to determine the 
relationship between credit scores and the actual credit losses 
and how those relationships vary for groups protected under 
ECOA. The Board concluded that credit scores accurately predict 
credit risk for the population as a whole and for all major 
demographic groups.
    The study revealed that on average, blacks and Hispanics 
have lower credit scores than non-Hispanic whites and Asians. 
This study suggests that if creditors were to collect data on 
race, the results would demonstrate a disparity in access to 
the pricing of credit that would be consistent with credit risk 
factors and not necessarily any discriminatory conduct by 
creditors.
    The Federal Reserve Board has already concluded the 
benefits of voluntary collection and reporting of race and 
gender data would not outweigh the potential harm. In 2003, the 
Board decided to retain the prohibition for two primary 
reasons: First, the collection of data not available before 
could create a risk of discrimination if it was made available; 
and, second, at least the voluntarily-provided data would be of 
questionable reliability.
    If voluntary data is unreliable, then the alternative would 
be mandatory data collection. From experience with HMDA 
reporting requirements we know that collection and reporting 
requirements require tremendous time and resources. We also 
know that a mere correlation between race and pricing without 
consideration of detailed creditworthiness factors cannot tell 
us whether or not illegal discrimination has occurred.
    Although collecting the data would provide little 
additional information, it will cause creditors to incur 
massive costs. These costs will inevitably be passed along at 
least in part to consumers at a time when consumers and 
creditors alike cannot afford increased cost of credit. 
Imposing mandatory data collection requirements should be 
driven by evidence that there is a lack of access to credit or 
fairness in pricing based on discriminatory factors.
    Today, most non-mortgage credit is underwritten and priced 
by creditors using objective, risk-based credit criteria 
without face-to-face interaction or any information regarding 
the applicant's race or other prohibited characteristics. These 
race-blind decisionmaking systems provide the very best 
assurances that consumers receive credit based on objective 
non-discriminatory criteria. It is hard to imagine that 
mandatory collection of racial information will improve this 
system. Collection and reporting of race and gender information 
also raises serious privacy concerns.
    Our experience with HMDA has shown that it is sometimes 
possible with the addition of other publicly available data to 
identify consumers in the HMDA loan registers. The collection 
and reporting of non-mortgage credit data significantly 
increases the risk that a consumer's sensitive personal 
information will enter the public domain. Also, it may be that 
consumers will object to being asked information about their 
race and see this as a violation of their privacy.
    In conclusion, Mr. Chairman, we believe that the Equal 
Credit Opportunity Act and Regulation B protect consumers from 
discriminatory lending practices and the current prohibition on 
data collection should be retained. Going forward, we must be 
careful not to undo the progress that has been made in creating 
a credit granting system that is race- and gender-neutral.
    That concludes my testimony. I thank you for inviting me to 
testify, and I would be happy to answer any questions.
    [The prepared statement of Mr. Himpler can be found on page 
71 of the appendix.]
    Chairman Watt. Thank you for your testimony.
    Mr. Corralejo, you are recognized for your testimony for 5 
minutes.

  STATEMENT OF JORGE C. CORRALEJO, CHAIRMAN, LATINO BUSINESS 
                 CHAMBER OF GREATER LOS ANGELES

    Mr. Corralejo. Thank you, Mr. Chairman, and members of the 
Financial Services Committee for inviting me to participate in 
this very important hearing today.
    I sit here on behalf of the Latino Business Chamber of 
Greater Los Angeles. The Chamber does its best to represent the 
interests of over 200,000 Latino-owned businesses in the Los 
Angeles area. It also represents the interests of tens of 
thousands or more Statewide, and hundreds of thousands more 
nationally.
    Because of the dramatic growth of ethnic minority 
populations and their businesses as well, lenders must and are 
taking a different and more realistic look at their future 
business client base. For example, over 50 percent of the State 
of California's population is ethnic minority. Their rate of 
growth and the development of small business is higher than the 
national average.
    In our community, the Latino community, the small business 
growth rate is 3 times the national rate. In our many 
discussions with bankers, they are often at a loss as to how to 
approach many minority business communities. Data collection 
will immensely help them. Through our various Reg B policy 
discussions within our Chamber and with other minority business 
chambers, we have universally agreed upon the need for 
collection and review of data by race, ethnicity, and gender as 
a benefit to all. This advantage would clearly be exercised by 
lenders in their marketing efforts to penetrate new and 
emerging ethnic communities. Information of this type, HMDA, 
has greatly increased the number of home loans to minority 
communities.
    It is our expectation that the number of loans to small 
businesses would increase several fold with this policy change. 
The economic contributions and growth in minority communities 
would be substantial. In the minds of Latino business owners, 
the collection of this data makes very good business sense. 
Minority small businesses depend heavily upon home equity funds 
for small business start-up and/or expansion.
    A major question in the small business arena is how great 
of an impact will the foreclosure crisis have on the small 
business community and, in this case, minority small business 
communities. Lenders need to know how this crisis impacts the 
future for small business clients and their ability to obtain 
small business loans in the future. The compilation of the data 
that we are requesting is an important component required for 
the progress of a whole picture on the national economy.
    A key instrument to this policy alternative is the 
immediate appointment of a task force which would resolve the 
foremost details and the potential cost. This should include 
all relevant government regulators, lenders, and minority small 
business leaders and associations. The dialogue and strategies 
that will transpire from these meetings will not only bring 
resolutions to the data collection policy, but inadvertently 
address other common economic development issues as well.
    I sit here today representing hundreds of thousands of 
Latino businesses in their support for legislation requiring 
the mandatory reporting of small business by race, ethnicity 
and gender by lending institutions with $1 billion or more in 
assets.
    We further support legislation that would permit all 
lenders with the opportunity to volunteer a report on this same 
data prior to the date for mandatory reporting.
    Thank you very much.
    [The prepared statement of Mr. Corralejo can be found on 
page 65 of the appendix.]
    Chairman Watt. Thank you so much for your testimony.
    Ms. Sullivan, you are recognized for your testimony.

 STATEMENT OF ANN SULLIVAN, PRESIDENT, MADISON SERVICES GROUP, 
           ON BEHALF OF WOMEN IMPACTING PUBLIC POLICY

    Ms. Sullivan. Chairman Watt and members of the 
subcommittee, thank you for inviting me to testify today.
    I am here today representing Women Impacting Public Policy, 
a bipartisan organization that represents over half-a-million 
women business owners across the country.
    I would like to address two issues today: One, the hurdles 
that women-owned businesses face with respect to access to 
capital; and, two, the need for additional data relevant to 
small women-owned businesses.
    Let me just say at the outset it was only 34 years ago with 
the passage of the Equal Credit Opportunity Act that women were 
able to obtain their own credit in their own name. In 1988, 
landmark legislation H.R. 5050 built upon that progress by 
making business loans subject to that Act. This had a 
tremendous effect on the growth of women-owned businesses, 
which now total 10.6 million.
    The Bureau of the Census began counting women-owned 
businesses in 1972 as a pilot project. The program originally 
only counted sole proprietorships. It was later expanded to 
include C corporations, so it included women-owned businesses 
with a much larger revenue stream.
    Every year, WIPP conducts an annual survey of its 
membership. In the 2008 survey, we found that women are using 
more sources of capital than in the previous 2007 survey, and 
60 percent of women business owners continue to seek outside 
funding for their businesses: 66 percent of the respondents use 
bank financing backed by home equity loans or other 
collateralized loans; 49 percent use credit card financing; 36 
percent get their funding from family and friends; 22 percent 
use SBA loans; 10 percent utilize angel investors; and 5 
percent use SBA Microloans.
    The good news from the WIPP's annual survey is that women 
appear to be making gains and obtaining credit to grow their 
businesses, but the struggles still continue. It is much more 
subtle, if discrimination in fact does exist. It is not as 
blatant as approval or denial. Rather, it is in the terms 
offered.
    While the problems I am going to mention may not be limited 
only to women-owned businesses, and are shared by other small 
businesses, let me just give you a few examples of barriers 
that they face. First, for early-stage businesses, the 
collateral requirements are high. Unless you have personal 
property to pledge against the loan, it is likely you won't 
receive any financing.
    Second, banks will not accept a signed government contract 
as collateral, which is often the most secure stream of funding 
the small business has to offer. Third, government agencies can 
prohibit small businesses from bidding by setting the bonding 
requirements artificially high. The small business cannot 
obtain that level of bonding, so they cannot bid. It is an easy 
way to keep small businesses out.
    Fourth, the ownership terms for venture funding often 
prohibit women-owned businesses from using that avenue for 
funding; and, fifth, SBA loan fees have now become a real issue 
in whether members choose to use them or not. With regard to 
SBA loans, 40 percent of all long-term capital for small 
businesses is provided through the SBA loan programs.
    I believe some important changes have taken place since 
2004 that have really had significant consequences for the 
lending programs.
    Congress stopped subsidizing the rate for small business 
loans and it lowered the guarantee. That resulted in an 
increase of lender and oversight fees.
    Those increased fees, of course, are passed on to the 
borrower. The House FY 2009 Financial Services Appropriation 
Bill included $100 million to subsidize the loan guarantee 
program and reduce the lender fees. Unfortunately, it was not 
included in the Senate. We hope that the House will insist on 
its position with regard to this funding.
    The topic of discussion in this hearing is whether lenders 
should be required to collect race and gender data of borrowers 
for small business lending. Let me just note that the SBA 
certainly tracks all of its lending. We recommend that the 
committee take a look at how they collect their data and 
perhaps use it as a model for possible expansion of data 
collection for small business loans.
    The data on women-owned businesses and small businesses in 
general is hardly robust. Very few sources of data exist. The 
Census Bureau Statistic of U.S. businesses produces data by 
NAIC's codes. Every 5 years, the Census Bureau conducts an 
economic census and that data lags 3 years behind. The Bureau 
of Labor Statistics produces employment statistics by firm 
size, which we use.
    Studies on small business lending, as you know, are very 
limited. One of the few sources that we had, which I understand 
is not going to be continued, is the Federal Reserve's survey 
of small business finances.
    We in the small business community, especially the women-
owned business community, use that survey as the basis for many 
of our statistics. With regard to whether or not the data 
should be mandated, we do not feel qualified to comment, but we 
encourage the committee to seek the most reliable method of 
data collection. With increased data collection, privacy issues 
should also be considered.
    Regulation B was amended to track minority and women 
lending. We would request assurances that this additional data 
collection includes safeguards to protect the data from 
unlawful usage. In summary, the GAO report was not really able 
to ascertain whether or not women-owned businesses faced higher 
credit denial rates than white, male businesses. But from the 
many stories we hear across the country, we know that it is a 
difficulty for women-owned businesses to obtain growth capital 
for their businesses.
    We believe that increased information on lending can be a 
very valuable tool to identify potential barriers to obtain 
that capital.
    Thank you for giving me the opportunity to testify. I am 
happy to answer any questions.
    [The prepared statement of Ms. Sullivan cn be found on page 
76 of the appendix.]
    Chairman Watt. Thank you, and thank all of the witnesses 
for your testimony.
    We will now proceed to the questions of the members, and I 
will recognize myself for 5 minutes for questions.
    One of the concerns that has been raised is the great 
variation, once you get outside the mortgage data collection, 
the great variation in kinds of loans. You have automobile 
loans. You have credit cards. You have small business loans. 
You have various and sundry other kinds of credit extended.
    Mr. Cavalluzzo, Mr. Gnaizda in particular, are there some 
categories where we should possibly looking at not collecting 
data?
    Mr. Cavalluzzo. Well, my work found large differences among 
small business lending, and, the differences that we found were 
far greater than what researchers have found in the home 
mortgage market.
    So I would definitely stress that area needs to be looked 
at. Now, as far as areas that may not need to be looked at, 
people have suggested, and whenever I raise the question of 
credit cards, because credit cards are through the mail, that 
discrimination is not an issue.
    Chairman Watt. That is what Mr. Himpler said. I think there 
are some categories where you can't get to the race data, but 
is that really true?
    Mr. Cavalluzzo. Well, that is what I was just about to say. 
You know, how do they decide who gets those mailings? Are 
certain zip codes receiving fewer mailings? I think the area 
that one would want to focus their energies would be a little 
bit different. So in credit cards, we might want to look at how 
people are deciding where these mailings get sent.
    In auto loans, a large percentage of auto loans are done at 
the dealership. The underwriter never interacts, meets with the 
borrower; however, there is the potential for issues at the 
dealership and that may be an area that could use some focus.
    With auto loans, because there is a strong incentive to 
sell the car, we might not see differences at the acceptance 
stage, but we might see it in pricing.
    Mr. Gnaizda. Mr. Chairman, our members have a slightly 
different view. They believe that the greatest consequences are 
in regards to potential discrimination and lack of marketing 
opportunities for small businesses; and, our position would be 
we would focus on that. But we have no objection; and, in fact, 
we strongly support full data collection in every area, because 
in every area there will be forms of discrimination. However, 
there may not be the will in Congress to go that far, and there 
may be increasing political opposition if we expand it beyond 
small business; and, that is why we have done it that way.
    Regarding the studies on discrimination, the Federal 
Reserve has never done a good study on small business 
discrimination; and the GAO has never done a study, although 
they have done a little analysis. We did an analysis 10 years 
ago for Los Angeles and South Central, and what we found is 
that most African-American and Latino business owners said that 
it was not discrimination that was the problem. It was the fear 
of discrimination; that is, they didn't even apply for the 
credit because they feared discrimination.
    That is consistent with Capital Management's position as 
well. We have also done a follow-up study in July of 2008, 
regarding Latino and Asian-American business owners in 
California. It is a random survey. It is small. We are going to 
turn it over to the Federal Reserve on Friday. We are going to 
ask them to follow up. It shows that 75 to 80 percent believe 
there's either discrimination or lack of interest in small, 
minority-owned businesses. So I think we have the empirical 
evidence or we could easily demonstrate it to focus on small 
business and mandatory collection.
    Chairman Watt. Would the problem of fearing rejection or 
fearing discrimination be addressed by reporting in some 
measure?
    Mr. Gnaizda. Yes, because Wachovia Bank, Bank of America, 
and Wells Fargo, for example, because I have spoken to all of 
their chairmen, would aggressively market to minority- and 
women-owned businesses. They now won't do that except for Wells 
Fargo because they can't collect the data, and therefore they 
can't tell how successful they are. No one wants to pour 
hundreds of millions of dollars into marketing without being 
able to look at the results.
    So I think almost instantly we'll be successful, and that 
is why we are going to urge the Federal Reserve immediately, as 
of January 2009, to allow any financial institution that wishes 
to replicate the Wells Fargo model.
    Chairman Watt. My time has expired. The gentleman from 
North Carolina is recognized for 5 minutes for questions.
    Mr. McHenry. Thank you, Mr. Chairman. Now, to follow up on 
your answer, Mr. Gnaizda, so what you are saying is Bank of 
America and those large institutions you mentioned currently 
discriminate because they don't have to report data?
    Mr. Gnaizda. No. I didn't say they discriminate.
    Mr. McHenry. Well, that is what I understood. You said they 
wouldn't market to them, because they don't have to disclose 
the data.
    Mr. Gnaizda. No. They won't market effectively because they 
can't measure the results. So, there's a form of inadvertent 
discrimination, but deliberately, Bank of America does not in 
my opinion want to discriminate. They recognize the enormous 
potential, particularly of minority- and women-owned 
businesses. In California, Bank of America knows that half of 
all new businesses are minority-owned.
    Mr. McHenry. So therefore, wouldn't they market generally?
    Mr. Gnaizda. Yes.
    Mr. McHenry. Across communities; isn't that being done?
    Mr. Gnaizda. No, they would use what I would call a white, 
male business approach that is tried and true, and they don't 
use any other approach except for Wells Fargo, which has its 
own form of measuring results. It is not as accurate as would 
be under mandatory reporting, so I think this will be a golden 
opportunity for every major financial institution. And, 
Greenlining has met with occasionally every one of those.
    Mr. McHenry. Let me ask this question, Mister--
    Will you say your name?
    Mr. Cavalluzzo. Cavalluzzo.
    Mr. McHenry. Cavalluzzo, thank you.
    You know, in terms of economics, if there is a vacuum 
created, someone will move into that vacuum. And based on the 
data you present, it seems to me that somebody can make a 
fortune by providing a service to a group that has been 
discriminated against.
    At least that is one way to look at it as, you know, why 
are people not jumping into that lurch? Can you give me based 
on your study in that regard?
    Mr. Cavalluzzo. We actually tried to address the point that 
you just raised by saying, ``That is right.'' Taste-based or 
prejudicial discrimination is not profit maximizing; and, in 
fact, if it exists, competition over time should try to 
mitigate it. People ought to come in and fill that background. 
So we looked at differences in different types of markets to 
see if the markets with less competition, the differences are 
more pronounced.
    We found some modest evidence that indeed was the case. 
Unfortunately, I don't have a good answer for you as to why we 
are still seeing these differences, even though I think you are 
right. Now, some forms of discrimination are economic based, 
despite being illegal, such as what's known as statistical 
discrimination. That is, if it is costly or difficult to 
measure risk characteristics of the borrower one could use race 
as a proxy or as a measure of that as opposed to collecting the 
data.
    We can't disentangle those issues as clearly as I would 
like to be able to, so it is hard to know why people haven't 
come in.
    Mr. McHenry. Okay. You know, in your study, the question I 
have is you went to the borrowers to do the study because you 
couldn't get the data from the lenders.
    Is there a potential if you had the information from the 
lenders that the conclusion would be different or perhaps more 
refined?
    Can you address it from that data set since that is what we 
are talking about?
    Mr. Cavalluzzo. Lender data has its advantages without a 
doubt. We would get a much better sense of the true 
underwriting model. We would know better what's on the 
application and would be able to construct models that perhaps 
better resemble what goes into the actual underwriting 
decision.
    We had very rich data. The small business data series 
provided very rich data. It is hard to say with any empirical 
study, even using lender data, if we saw large differences, 
that those differences would be due definitively to 
discrimination.
    I don't know that we could make much stronger statements 
than we have already made, but it would complete the picture 
more.
    We have looked at it from the borrower's side. We have 
looked at it from the lender's side and collectively here is 
what the data point to.
    Mr. McHenry. Thank you.
    Chairman Watt. The gentleman from Missouri.
    Mr. Cleaver. Thank you, Mr. Chairman.
    Mr. Himpler, I am going to ask you questions simply because 
I can pronounce your name easily, so I am not picking on you. 
Your testimony, however, is amazing. On page 3 of your 
testimony, you open and say, ``There is scant statistical 
evidence to demonstrate that race and gender play a role in 
access to or the cost of non-mortgage credit.''
    How do you respond to the GAO report, which suggests that 
discrimination does in fact play a role in non-mortgage 
lending?
    Mr. Himpler. Congressman, the GAO concluded that it may 
play a role. It doesn't say that it does. Two, the report by 
the GAO is based on reports. It is not based on statistical 
analysis.
    Mr. Cleaver. How can you be certain that there is no 
discrimination unless you already have the data and you are 
analyzing it?
    Mr. Himpler. We believe that the regulators have the data 
that they need.
    Mr. Cleaver. Based on what?
    Mr. Himpler. Based on their ability to take a look at 
individual loan files.
    Mr. Cleaver. No. No, the regulators have data. How do you 
know they have it?
    Mr. Himpler. I do know, in one instance in particular I 
will speak to, that the Fed has worked with the credit bureaus 
to get essentially the same sort of data that we would get from 
a sort that we are talking about by looking at credit 
histories, credit scores, by census track.
    You are going to achieve the same outcome so that you have 
with what we are looking at achieving.
    Mr. Cleaver. So you have analyzed it?
    Mr. Himpler. No, sir. I have not analyzed it.
    Mr. Cleaver. How do you know? If you haven't analyzed it, 
how do you know?
    Mr. Himpler. How do I know what, sir?
    Mr. Cleaver. How do you know the data Federal agencies are 
collecting would suggest that there would be no discrimination?
    Mr. Himpler. I don't know that, sir. What I am saying is 
the evidence that is out there right now does not provide us 
with empirical data to make that conclusion.
    Mr. Cleaver. So there is no discrimination?
    Mr. Himpler. I am not saying there is no discrimination, 
sir. What I am saying is that the request to collect the data 
that we are talking about will, like HMDA, not show 
discrimination. What it does is provide a flagging system for 
the regulators to do further analysis. But at the end of the 
day, regulators or researchers have to look at individual loan 
files to determine whether or not discrimination existed. There 
is no way of getting around it, and we have talked about 
comparing this to HMDA.
    I think it is important for the committee to understand. 
Ms. Braunstein made a comparison that we want to make sure we 
don't compare apples and oranges. Extending the analogy from 
HMDA to other types of lending is not like comparing apples and 
oranges. It is like comparing apples and concrete. With 
mortgage lending, you have a nationalized, standardized system.
    Using Fannie and Freddie automated underwriting, everybody 
is playing from the same deck of cards, if you will, or the 
same song sheet in terms of criteria that they use. When it 
comes to small business lending, when it comes to personal 
loans, when it comes to auto finance, you don't have any sort 
of nationalized, standardized system. So, the question was 
raised as to whether or not we could do this.
    Yes, we could collect all the information that is necessary 
to make the determination. But essentially, what that would be 
is an unusable database.
    Mr. Cleaver. Maybe I can get at it from this way.
    Mr. Himpler. Sure.
    Mr. Cleaver. In your testimony you asked that we not act 
against Regulation B so that we will not undo the progress that 
has been made in creating a credit granting system that is race 
and gender neutral.
    Tell me what progress you are speaking about.
    Mr. Himpler. Essentially what we are talking about is as 
recent as 20 years ago. I would like to make the analogy that 
credit was like an on/off switch. If you had pristine credit 
history you had access to credit. If you didn't, you were 
dealing with folks that nobody wants to deal with. With the 
technology developments and risk-based pricing, we have gone 
from an on/off switch to more of a dial.
    Creditors base the prices that they charge customers using 
the credit factors that those borrowers bring to the table. It 
is not an either/or. It is a range.
    Mr. Cleaver. So we have made significant progress, and your 
goal is to make sure we don't undo it.
    Mr. Himpler. Absolutely, I mean, the one thing Mr. Gnaizda 
mentioned was his analysis, and we would be more than happy to 
work with his organization to take his study to the regulators, 
have them assess the study in order to validate it to the 
lending institutions they are talking about, to the market, to 
the folks in every community across this great land.
    Mr. Cleaver. Yes. I mean, you are talking about progress. 
And I have been black every day of my life and I, you know, can 
sit here and talk to you about progress. And I am troubled by 
the fact that you are almost suggesting here that there is no 
problem, and we probably shouldn't even be doing this hearing.
    Mr. Himpler. That is not what I intended to convey at all. 
We share your commitment to wanting to end discrimination. We 
are just not sure that collecting the data set of the type we 
are talking about, putting race and rates and gender 
information together without any credit information or other 
credit factors gets us in that direction.
    Those data sets are already available to regulators if they 
want to access them, essentially through the credit bureaus as 
a screening tool. But at the end of the day, you still have to 
look at the individual loan files. There is no way of getting 
around it. That is what Ms. Braunstein meant.
    Mr. Cleaver. Yes, you look at an individual loan file, and 
there is plenty of information that can be extracted that can 
be used for discrimination. I mean, somebody's name is Shaft, 
or they have a zip code or Mr. T, or something, and you know if 
they went to Howard University. I mean, there are all kinds of 
ways to extract information for purposes of discrimination, and 
I don't think we ought to deny that.
    That is one of the reasons it is continuing to hang around 
is because we are denying it instead of challenging it, and I 
am just concerned. I read your report three times, because I 
wanted to conclude that I had misread it; but I didn't, and I 
am extremely concerned, you know. You even speak about the 
negative effects of data collection, as if collecting the data 
will generate greater discrimination. How do you do that?
    Mr. Himpler. Not greater discrimination; it will reduce 
access to credit. It will raise the cost for creditors. 
Ultimately, it will raise that cost and at least in part will 
be passed the law into the consumer. As costs rise, people have 
less access to credit.
    Mr. Cleaver. Of course, that is already happening.
    Mr. Himpler. You just don't want to make it worse.
    Mr. Cleaver. Worse than what happened in the subprime 
market?
    Mr. Himpler. Yes.
    Mr. Cleaver. I mean, every study shows that people were 
targeted for subprime loans. People went after certain people. 
The brokers, the real estate companies in some instances, went 
after a target group. Am I wrong?
    Mr. Himpler. I am not familiar with the studies you are 
talking about, sir.
    Mr. Cleaver. Okay, well, let's not--forget a study. Who do 
you think were the targets of the subprime loans? You don't 
know?
    Mr. Himpler. No.
    Mr. Cleaver. In almost all the newspapers, it has been 
clear, Latinos and African Americans. And I know that is 
surprising to you.
    Mr. Himpler. No, I would say a substantial number of 
subprime borrowers are Asians and non-Hispanic whites.
    Mr. Cleaver. Well, why do you think they were targeted?
    Mr. Himpler. Based on the credit factors that they brought 
to the table, sir.
    Mr. Cleaver. That had nothing to do with race?
    Mr. Himpler. I am not going to sit here and say that there 
are no instances of discrimination. I have no evidence to back 
up that claim, but we in the lending community base our credit 
decisions on objectified credit histories that we get from the 
credit bureaus. The Federal Reserve came to this conclusion 
that credit scores are predictive.
    Mr. Cleaver. My final question: Your recommendation is that 
we do nothing?
    Mr. Himpler. My recommendation is that we not do this.
    Mr. Cleaver. So what should we do?
    Mr. Himpler. One of the things is that in listening to Mr. 
Gnaizda, he has proffered a study that he thinks will move the 
ball forward in terms of encouraging major institutions to get 
more aggressive in their marketing. I will commit to you right 
here today. We would be happy to work with his organization, to 
go to the regulators, to validate their results. And once they 
get the Federal regulators' seal of approval, which I think is 
what would help gain the confidence of the financial 
institutions, and take that to the financial institutions.
    Mr. Cleaver. That is a good answer.
    Mr. Gnaizda. I don't think that is a solution of any kind 
at all. Greenlining has raised with every financial 
institution, including your members, doing such studies. They 
have always refused to do so. We have had to foot the bill 
ourselves.
    We are happy to work with the Federal Reserve and we will 
make our invitations to the major financial institutions 
independent of the association, and independent of any delay, 
because I don't want to have a delay until we have a study of 5 
million, which your members will probably request at a minimum.
    Mr. Himpler. And, all I am saying is taking your results, 
having the Federal regulators look at them as an independent 
reviewing body, and then taking those results forward.
    Chairman Watt. The gentleman's time has expired.
    We are going to go another round, because I have a few more 
questions.
    Mr. Himpler, did your organization support collection of 
HMDA data?
    Mr. Himpler. No. We did not.
    Chairman Watt. Okay, so you oppose that, too?
    Mr. Himpler. Yes.
    Chairman Watt. A part of our responsibility that we have 
undertaken is to try to look at the cost. And on page 4 of your 
testimony you cite the ``massive'' cost that creditors would 
incur if they were required to collect this data.
    Can you quantify what ``massive'' is? What information do 
you have that you can provide to the committee about the extent 
of the cost that would be associated with this?
    Mr. Himpler. Like Ms. Braunstein, it would be a back of the 
envelope estimate, Mr. Chairman.
    Chairman Watt. But we are looking for something more than a 
back of the envelope, and I am not sure that you should have it 
this morning; but if you come before this subcommittee and 
testify that there is a ``massive'' cost associated with it, 
then we need to be evaluating what that cost is. And the only 
way we can do that is to get something other than a conclusory 
word like ``massive.''
    Mr. Himpler. May I?
    Chairman Watt. So I am not suggesting that I need it this 
morning, but part of our responsibility is to look at the 
actual projected cost. And if you have information on that, we 
would welcome you to submit it in writing to us, not do it on 
the back of an envelope. Because our inquiries are a little bit 
more serious than the back of an envelope would suggest.
    Mr. Himpler. I did not mean to imply the inquiry was not 
serious. What I can tell you in terms of using the word 
``massive'' is that a substantial number of lending 
institutions, finance companies that were discussed by Ms. 
Braunstein, do not come under Federal oversight.
    Chairman Watt. I am aware of that. That was my next 
question, in fact, but the cost of doing this for them would 
be?
    Mr. Himpler. A significantly higher portion relative to 
their business size.
    Chairman Watt. What makes you conclude that?
    Mr. Himpler. A number of them are not computerized in terms 
of their lending. A lot of loan files are--
    Chairman Watt. Who is not computerized in their lending? 
Tell me somebody who is not computerized in their lending that 
is engaged in the lending business?
    Mr. Himpler. I would say Regional Finance of Mississippi.
    Chairman Watt. Regional Finance of Mississippi?
    Mr. Himpler. A number of our companies are small companies.
    Chairman Watt. I understand that, but you said you were 
going to give me the names of some people who are not 
computerized and they are issuing credit?
    Mr. Himpler. No. No, in terms of the data set in terms of 
collecting HMDA-like data, they do not have it.
    Chairman Watt. I understand that. Nobody has a system set 
up to collect this data at this moment. We understand that. I 
mean, because nobody has required them to collect it up to this 
point, but when you represent to me that somehow this is 
disproportionate for people who are regulated or unregulated as 
opposed to those who are regulated, I don't understand that.
    Mr. Himpler. This type of data collection is not something 
you can buy right off the shelf, so you have to outsource that, 
bring people in to work with your systems, work with your 
lending officers and your branches.
    Chairman Watt. Mr. Himpler, I am going to issue you an open 
invitation. We have made a commitment here this morning to have 
a hearing about the cost, because I think it is important for 
us to assess not only the benefit of collecting data, but the 
cost that would be incurred in the collection process. And, so, 
if you would in the next 30 days give us as complete 
information as you can give us about your analysis of what you 
characterize as ``massive'' costs associated with what could be 
required.
    I understand that there are costs associated. We haven't 
denied that. It was part of my opening statement if you were 
here. And I understand that it is our obligation to assess at 
some level the benefits against the cost, and the only way we 
can do that, I think, we know the benefits of doing this, 
although you argue with those, too.
    What we are interested in is getting a better handle on 
what the cost would be, and so I am issuing you an invitation 
to submit that in writing.
    Mr. McHenry.
    Mr. McHenry. Mr. Himpler, what are the privacy concerns?
    You have testified about HMDA data, last year in the prior 
Congress. I don't have the date in mind, but can you talk about 
the privacy concerns in collecting this type of data?
    Mr. Himpler. Sure. In the HMDA context, using HMDA loan 
registers and other publicly available data, currently, you can 
determine exactly who got what loan in giving census tracks--
not across-the-board--but it is possible to do.
    Our fear in this regard is that particularly with respect 
to auto finance lending, looking at census tracks and other 
available information such as title registries, that you would 
be able to do exactly the same sort of thing, redact out of the 
data collection what rating your neighbor got on his or her 
Honda Accord or Ford Escort.
    Mr. McHenry. Okay. You know, the issue here, we go back to 
HMDA data. This is instructive for me. It seemed like the large 
financial institutions were not serving the groups that we are 
concerned about here today. And I think that was brought out.
    Mr. Gnaizda, you are nodding, but I think the HMDA data 
showed that some of these large financial institutions were not 
going out and doing it. And back to my concept with Mr. 
Cavalluzzo, at least the economic notion, I don't know if it is 
reality, and that is what we are trying to hash out here today. 
I think the chairman's intent to see if this data bears that 
out, but the vacuum is filled.
    I believe in terms of the HMDA data, the data showed that 
the subprime marketplace when it actually filled that vacuum, 
some of these larger institutions were not doing this. So you 
had smaller institutions that had different pricing models that 
went in and figured out a way to do it. And Mr. Gnaizda, go 
right ahead.
    Mr. Gnaizda. Yes, we quite agree with you. That is the 
problem, that if you created an enormous vacuum the unregulated 
will engage in unscrupulous behavior. That has what has caused 
the subprime crisis.
    Mr. McHenry. Well, that is not all of it. Let's not 
simplify this. I mean we have a huge economic issue here, and 
you can't simplify it to simply say you had a bunch of legal 
operators. You can't simplify it.
    Mr. Gnaizda. You had, for example, those on the edge like 
Countrywide that dominated the market and forced the scrupulous 
with a couple of exceptions to compete with them.
    We don't want to create the lowest common denominator. We 
believe that the HMDA data you raised demonstrated the 
importance of this data. Richard Rosenberg, then president of 
the Bank of America, had told Greenlining members that Bank of 
America took advantage of every opportunity to make loans to 
African Americans and Latinos and resented the community's 
views that they did not. When HMDA data was forced upon them, 
the first year after they collected it, he asked us to come and 
meet with them. And he said, ``I was wrong. We did not maximize 
the opportunities, and now we will.''
    And that is exactly what I think will happen when we have 
Reg B being abolished and there is mandatory reporting. The 
major banks and their CEOs will come here in 5 years and 
perhaps thank you.
    Mr. McHenry. So would you say, I mean, I know you are not 
in a lending marketplace, but I would say that a car loan is 
different from a home loan and is different from just a 
personal loan. And having gotten all three, I know the hoops 
that you go through are very different and your asset 
requirements and everything else.
    So in terms of getting the depth of the pricing model and 
the risk model, how do you really do that, Mr. Cavalluzzo in 
terms of statistical analysis? Because you may have one firm, 
and you can see this with banks currently in terms of how 
strong certain banks are versus how at risk others are based on 
their own pricing model. You know, I go to one bank and they 
say, you know, heck no, we are not giving you a loan. You turn 
around to another bank and you can get it. You know, so in 
terms of the pricing model, how can you account for that in a 
statistically reliable way comparing apples to apples?
    Mr. Cavalluzzo. That is right. Small business lending is a 
very complex issue. When I went to banks to start my research, 
I found some banks told me, ``We just look at the credit score. 
That is what we use.'' Other banks wanted a full balance sheet, 
income statement, to understand the business that the firm is 
engaging in. It is a very hard problem.
    Even bank underwriting models vary by the institution. So 
in part, and that is one of the limitations to borrower data, 
that lender data could help to address--getting data from the 
lender would potentially allow a researcher to let that 
underwriting model vary.
    Mr. McHenry. Well, if the chairman will indulge me for a 
moment here, there are two elements there. There is the 
competitive nature between the institutions doing the lending, 
and you have some institutions that believe their pricing 
model, their assessment of risk, is better and sharper than the 
other guy's down the block. Therefore, they don't want to 
release that. So you have some privacy issues within 
institutions. Secondly, counter to that you have individuals as 
a consumer in trying to get a loan who don't know the different 
pricing mechanisms. So how do you on one hand let individuals 
know in the same manner allow these institutions to figure out 
their competitive advantages. I mean, how do you keep that 
data, you know, out of the competitor's hands.
    Mr. Cavalluzzo. Well, the name of the lending institution 
doesn't have to be disclosed within the data.
    Mr. McHenry. But you can pretty easily figure it out. I am 
from a small town; and, I mean, you can give me a data set, and 
I guarantee I can figure out which one of the three banks in my 
town it is.
    Mr. Cavalluzzo. If the data set were restricted to your 
town. But if the data set were for the entire country or the 
entire State, I think it would be much more difficult.
    Mr. McHenry. But in terms of lending, you can see census 
track information is what Mr. Himpler pointed out.
    How do you restrict that? You know, because we do have 
privacy concerns.
    Mr. Cavalluzzo. I think that the regulators need to take 
care with data that they actually do disclose, I don't think 
that they necessarily need to disclose information that would 
allow a researcher to identify the particular institution.
    Yet, still, a researcher can learn an enormous amount from 
that set of data. In our data we had no idea what the lending 
institutions were. In fact, regional codes were nine large 
regions across the country.
    Mr. McHenry. But it is sort of a sliding scale. It is 
tricky to do.
    Mr. Cavalluzzo. To mask certain pieces of the data?
    Mr. McHenry. Yes.
    Mr. Cavalluzzo. I don't believe it is that tricky.
    Mr. McHenry. Okay. Well, thank you for your testimony. I 
appreciate you all going through the second round.
    And, Mr. Chairman, I think the question here, and I am glad 
you are going to go for a hearing on the cost, because I think 
it is a great unknown. Everybody admits it will cost something. 
The question is what is that cost. We know the consumer will 
bear that cost, whatever it is, whether it is very small or 
very large.
    And so, Mr. Chairman, I think it is a laudable goal, and we 
have to figure out the best way of approaching it. I am 
grateful you are having this hearing.
    Chairman Watt. Mr. Cleaver.
    Mr. Cleaver. Thank you, Mr. Chairman.
    I have two questions.
    How are you today, Ms. Sullivan?
    Ms. Sullivan. You don't want me to fall asleep, right?
    [Laughter]
    Mr. Cleaver. Yes, well, your name is easy to pronounce and 
I just wanted to make sure you were included. Thank you.
    Mr. Himpler.
    Mr. Himpler. Still here, Congressman.
    Mr. Cleaver. I am stuck on ``We must be careful not to undo 
progress that has been made in creating a credit planning 
system that has race and gender neutral.'' And I am trying to 
see how we are on the verge of undoing that progress.
    Mr. Himpler. It is a very good question, first off. What I 
said in the first round of questions was that we have gone from 
a denial of access or the unavailability of credit to segments 
of our population as recently as 30 years ago to systems now 
where we can base credit approvals on objective, credit-scoring 
models that take advantage of objective criteria that does not 
involve prohibited characteristics in granting those approvals.
    Our concern was that with some of the collection: (1) it 
will increase the cost that will be passed along to the 
consumer--that increased cost may take some folks out of the 
equation in terms of affordability; and (2) it will raise 
privacy concerns because of the availability of publicly 
available data matched up with the type of collection that we 
have that will increase the fear that Mr. Gnaizda talked about 
earlier from wanting to have their private information revealed 
and would thus opt out.
    We think both cost and from privacy concerns that could 
roll us back.
    Mr. Cleaver. Roll us back to 30 years ago.
    Mr. Himpler. Not to 30 years ago, but it is backwards. It 
is not forwards.
    Mr. Cleaver. I am getting a headache.
    Mr. Himpler. I don't mean to give you a headache, 
Congressman.
    Mr. Cleaver. No, it is too late. You know, there are all 
kind of ways to find out. I mean, to look at a document and see 
red flags, if you want to call it that, that would point to a 
specific racial or ethnic group and then make decisions based 
on that. I mean, you would not think that the gentleman sitting 
next to you to your left is Irish by looking at the name. Am I 
right about it?
    Mr. Himpler. Yes, you are right, but you wouldn't think by 
looking at my name that I am half Mexican either, would you?
    Mr. Cleaver. No. No. But I could tell by your statement 
that you haven't been treated that way.
    Mr. Himpler. Not true.
    Mr. Cleaver. That is all, Mr. Chairman.
    Chairman Watt. Thank you.
    Bottom line here, our inquiry is cost versus benefit, 
benefit versus cost. So let me just ask each one of the five 
witnesses to address, first of all, Ms. Sullivan, going from 
you to Mr. Cavalluzzo.
    What is your assessment of the cost benefit analysis on 
HMDA?
    Has HMDA been more beneficial than costly?
    And what is your assessment of the projected cost-benefit 
analysis on other kinds of lending other than mortgage lending 
if we were to require it be reported mandatory, Ms. Sullivan, 
HMDA cost benefit, non-HMDA cost benefit?
    Ms. Sullivan. We are not experts on HMDA. But I would say 
that all this talk about how much everything is going to cost 
is just confusing. And maybe the data is so much more 
comprehensive that you are talking about than what the SBA 
collects; they are not direct lenders.
    Lending institutions are reporting to them. I have never 
heard of any onerous costs associated with it. I have 
participated in plenty of small business lending Roundtables in 
the House and the Senate and no one has ever raised that. It 
seems to me that it is pretty simple.
    With regard to 7(a) loans, I can see that only 4 percent of 
African Americans used the loans in FY 2001 and it increased 5 
percent in FY 2007. Clearly, the 7(a) program is not reaching 
out to the African-American population. It looks like there 
needs to be some improvement. I mean just even simple data 
collection like this is helpful to our segments of the 
industry, and it is also helpful to policymakers. It gives you 
benchmarks about where the lending is and what needs to be 
improved. I just don't see why this can't have the same 
relevance and the same kind of benefits for a larger data 
collection that you are talking about.
    Chairman Watt. Mr. Corralejo.
    Mr. Corralejo. ``Corralejo.''
    Chairman Watt. ``Corralejo.''
    Mr. Corralejo. Thank you.
    You know, I hear discussion about the ultimate costs passed 
back onto the borrower, and that could be the case. We don't 
know. But I think that there is such a potential for growth 
here in the minority communities, I don't think they need to be 
passed along to borrower. There is another option here. I think 
this is a question of growth. I think the same way that HMDA 
has attributed to greater loans to minority communities, I 
think this will be the same thing with minority business 
people. So, as banks grow and they open new branches, there are 
costs at opening branches, but the benefits far outweigh the 
costs. And I think that is the formula that we are looking 
here: does that potential really exist.
    And in our discussions with numerous bankers--we have 
numerous bankers as our partners--they are dying for a way to 
figure this out. Not that they have yet, but I think clearly 
this is one of the means and methods that we seriously need to 
take a look at. So I think in the end my assumption is that 
this is a cost of increased business and not necessarily passed 
along to the borrower.
    Chairman Watt. I take it you are not dealing with those 
banks or members of Mr. Himpler's association.
    Mr. Corralejo. But we deal with small banks, and that is 
why in our testimony we talked about banks with 1 billion in 
assets or more, because we understand there are differences. So 
these are the details that need to be worked out.
    Chairman Watt. Mr. Himpler, on the cost benefit analysis of 
HMDA, first of all, and you have already actually testified 
about the cost benefit analysis of your assessment of the cost 
benefit analysis of what may be proposed, but I am not sure I 
have you on record.
    I know your association opposed collection of HMDA data, 
but now that it is out there, it has been done for a while, 
what is your assessment of the cost benefit analysis?
    Mr. Himpler. It would be silly to say that it hasn't been 
beneficial to the regulators in terms of their ability to 
identify potential problems and do further investigation. But 
you can't get a determination of actual discrimination from 
HMDA data. There is no credit information.
    Chairman Watt. What we do want to look at is pushing the 
data fields collection of other data fields where you can make 
that determination. But that is a subject for another day.
    Mr. Himpler. Right, I will leave it alone, but that does 
come at a potential cost of harm to the business community in 
terms of comparing race and rates, or ethnicity and rates, 
without any credit information that they are guilty until 
proven innocent. And that is a cost. So there is cost and 
benefit, and there is also the cost of the collection you have 
for the first time in the last 3 and 4 years, a number of 
mortgage companies that had never collected HMDA data period. 
So there was a ramp-up to that.
    As far as the cost associated with the proposal today, I 
just want to make sure folks understand exactly what we are 
saying: that there are easier ways to do this; that we will 
achieve the same sort of outcome from data sets that are 
already out there from the bureaus with respect to research or 
that can be used by the regulators as a flagging mechanism to 
do the further analysis that they need to do on a more cost-
effective basis.
    Chairman Watt. Mr. Gnaizda.
    Mr. Gnaizda. Yes.
    Chairman Watt. Same questions.
    Mr. Gnaizda. Yes, Mr. Chairman, we have analyzed the HMDA 
data.
    The number of African Americans and Latinos who have 
received home loans at fixed rates--these are prime fixed-rates 
every year once HMDA data was available--increased 
substantially. So there are millions of African Americans and 
Latinos and Southeast Asians who own homes today at fixed rates 
and are not subject to the present crisis due to HMDA 
collection. We believe many major banks, if their CEOs came, 
would testify to that.
    Now, with regard to cost-benefit-analysis regarding the 
future and small business lending, the benefits outweigh any 
possible cost, even if exaggerated three-fold. And that is 
because the same thing that happened with HMDA, more 
homeowners, as Mr. Corralejo has said, there will be more small 
businesses that are women-owned and minority-owned that will be 
the beneficiaries of prime, small business lending by major 
financial institutions. And we believe that the Federal Reserve 
can assist in this.
    That is why we are going to meet with Ms. Braunstein on 
Friday. They should call a task force immediately, and they 
should have the 10 largest financial institutions doing small 
business lending. We have identified them, and they should 
bring them in quickly, and by September begin a task force.
    Thank you.
    Chairman Watt. You have gone over that and that is why I am 
stopping you.
    Mr. Cavalluzzo.
    Mr. Cavalluzzo. First, I am not aware of the data sets that 
Mr. Himpler refers to on this topic; and, I know the literature 
extremely well in this area. I think that the differences we 
document in our research, while I understand that there are 
costs to the lending institutions, I think the differences we 
document suggest that the cost to society are potentially quite 
large and that, I think, fair treatment needs to examine both 
the costs to the lender and the cost to society at large. The 
cost to lending institutions, though, may be less than I think 
we were being led to believe, because banks already codified 
this data.
    Many banks transfer this data to credit scoring models, so 
that these credit scoring models can be refined and improved 
upon. Banks pay credit scoring agencies to do this, so a lot of 
this data collection and transferring is already taking place.
    I think that banks could actually find this data useful, so 
that they can better understand the lending process. The 
lending process doesn't need to be a black box to consumers or 
lenders. If people can understand what it takes to improve 
their credit score, then they are better borrowers.
    I think there are benefits to many pieces of society, 
rather than just the cost to lender. Lenders may actually find 
benefits in all of this. In particular, we may find that these 
data help us fill that vacuum that is out there for these firms 
that aren't getting the credit they need.
    Chairman Watt. Your testimony has been amazingly helpful, I 
think, in helping us evaluate that.
    The Chair notes that some members may have additional 
questions for this panel, which they may wish to submit in 
writing. Mr. Himpler, for example, we will submit the request 
for your cost information in writing.
    Without objection, the hearing record will remain open for 
30 days for members to submit written questions to these 
witnesses and to place their responses in the record.
    We thank you all immensely for being here today, and for 
your interest in this subject, and unless Mr. McHenry has any 
other comments, I declare that the hearing is adjourned.
    Thank you so much.
    [Whereupon, at 12:37 p.m., the hearing was adjourned.]


                            A P P E N D I X



                             July 17, 2008


[GRAPHIC] [TIFF OMITTED] T4902.001

[GRAPHIC] [TIFF OMITTED] T4902.002

[GRAPHIC] [TIFF OMITTED] T4902.003

[GRAPHIC] [TIFF OMITTED] T4902.004

[GRAPHIC] [TIFF OMITTED] T4902.005

[GRAPHIC] [TIFF OMITTED] T4902.006

[GRAPHIC] [TIFF OMITTED] T4902.007

[GRAPHIC] [TIFF OMITTED] T4902.008

[GRAPHIC] [TIFF OMITTED] T4902.009

[GRAPHIC] [TIFF OMITTED] T4902.010

[GRAPHIC] [TIFF OMITTED] T4902.011

[GRAPHIC] [TIFF OMITTED] T4902.012

[GRAPHIC] [TIFF OMITTED] T4902.013

[GRAPHIC] [TIFF OMITTED] T4902.014

[GRAPHIC] [TIFF OMITTED] T4902.015

[GRAPHIC] [TIFF OMITTED] T4902.016

[GRAPHIC] [TIFF OMITTED] T4902.017

[GRAPHIC] [TIFF OMITTED] T4902.018

[GRAPHIC] [TIFF OMITTED] T4902.019

[GRAPHIC] [TIFF OMITTED] T4902.020

[GRAPHIC] [TIFF OMITTED] T4902.021

[GRAPHIC] [TIFF OMITTED] T4902.022

[GRAPHIC] [TIFF OMITTED] T4902.023

[GRAPHIC] [TIFF OMITTED] T4902.024

[GRAPHIC] [TIFF OMITTED] T4902.025

[GRAPHIC] [TIFF OMITTED] T4902.026

[GRAPHIC] [TIFF OMITTED] T4902.027

[GRAPHIC] [TIFF OMITTED] T4902.028

[GRAPHIC] [TIFF OMITTED] T4902.029

[GRAPHIC] [TIFF OMITTED] T4902.030

[GRAPHIC] [TIFF OMITTED] T4902.031

[GRAPHIC] [TIFF OMITTED] T4902.032

[GRAPHIC] [TIFF OMITTED] T4902.033

[GRAPHIC] [TIFF OMITTED] T4902.034

[GRAPHIC] [TIFF OMITTED] T4902.035

[GRAPHIC] [TIFF OMITTED] T4902.036

[GRAPHIC] [TIFF OMITTED] T4902.037

[GRAPHIC] [TIFF OMITTED] T4902.038

[GRAPHIC] [TIFF OMITTED] T4902.039

[GRAPHIC] [TIFF OMITTED] T4902.040

[GRAPHIC] [TIFF OMITTED] T4902.041

[GRAPHIC] [TIFF OMITTED] T4902.042

[GRAPHIC] [TIFF OMITTED] T4902.043

[GRAPHIC] [TIFF OMITTED] T4902.044

[GRAPHIC] [TIFF OMITTED] T4902.045

[GRAPHIC] [TIFF OMITTED] T4902.046

[GRAPHIC] [TIFF OMITTED] T4902.047

[GRAPHIC] [TIFF OMITTED] T4902.048

[GRAPHIC] [TIFF OMITTED] T4902.049

[GRAPHIC] [TIFF OMITTED] T4902.050

[GRAPHIC] [TIFF OMITTED] T4902.051

[GRAPHIC] [TIFF OMITTED] T4902.052

[GRAPHIC] [TIFF OMITTED] T4902.053

[GRAPHIC] [TIFF OMITTED] T4902.054

[GRAPHIC] [TIFF OMITTED] T4902.055

[GRAPHIC] [TIFF OMITTED] T4902.056

[GRAPHIC] [TIFF OMITTED] T4902.057

[GRAPHIC] [TIFF OMITTED] T4902.058

[GRAPHIC] [TIFF OMITTED] T4902.059

[GRAPHIC] [TIFF OMITTED] T4902.060

[GRAPHIC] [TIFF OMITTED] T4902.061

[GRAPHIC] [TIFF OMITTED] T4902.062

[GRAPHIC] [TIFF OMITTED] T4902.063

[GRAPHIC] [TIFF OMITTED] T4902.064

[GRAPHIC] [TIFF OMITTED] T4902.065

[GRAPHIC] [TIFF OMITTED] T4902.066

[GRAPHIC] [TIFF OMITTED] T4902.067

[GRAPHIC] [TIFF OMITTED] T4902.068

[GRAPHIC] [TIFF OMITTED] T4902.069

[GRAPHIC] [TIFF OMITTED] T4902.070

[GRAPHIC] [TIFF OMITTED] T4902.071

[GRAPHIC] [TIFF OMITTED] T4902.072

[GRAPHIC] [TIFF OMITTED] T4902.073

[GRAPHIC] [TIFF OMITTED] T4902.074

[GRAPHIC] [TIFF OMITTED] T4902.075

[GRAPHIC] [TIFF OMITTED] T4902.076

[GRAPHIC] [TIFF OMITTED] T4902.077

[GRAPHIC] [TIFF OMITTED] T4902.078

[GRAPHIC] [TIFF OMITTED] T4902.079

[GRAPHIC] [TIFF OMITTED] T4902.080

[GRAPHIC] [TIFF OMITTED] T4902.081

[GRAPHIC] [TIFF OMITTED] T4902.082

[GRAPHIC] [TIFF OMITTED] T4902.083

[GRAPHIC] [TIFF OMITTED] T4902.084

[GRAPHIC] [TIFF OMITTED] T4902.085

[GRAPHIC] [TIFF OMITTED] T4902.086

[GRAPHIC] [TIFF OMITTED] T4902.087

[GRAPHIC] [TIFF OMITTED] T4902.088

[GRAPHIC] [TIFF OMITTED] T4902.089

[GRAPHIC] [TIFF OMITTED] T4902.090

[GRAPHIC] [TIFF OMITTED] T4902.091

[GRAPHIC] [TIFF OMITTED] T4902.092

[GRAPHIC] [TIFF OMITTED] T4902.093

[GRAPHIC] [TIFF OMITTED] T4902.094

[GRAPHIC] [TIFF OMITTED] T4902.095

[GRAPHIC] [TIFF OMITTED] T4902.096

[GRAPHIC] [TIFF OMITTED] T4902.097

[GRAPHIC] [TIFF OMITTED] T4902.098

[GRAPHIC] [TIFF OMITTED] T4902.099

[GRAPHIC] [TIFF OMITTED] T4902.100

[GRAPHIC] [TIFF OMITTED] T4902.101

[GRAPHIC] [TIFF OMITTED] T4902.102

[GRAPHIC] [TIFF OMITTED] T4902.103

[GRAPHIC] [TIFF OMITTED] T4902.104

[GRAPHIC] [TIFF OMITTED] T4902.105

[GRAPHIC] [TIFF OMITTED] T4902.106

[GRAPHIC] [TIFF OMITTED] T4902.107

[GRAPHIC] [TIFF OMITTED] T4902.108

[GRAPHIC] [TIFF OMITTED] T4902.109

[GRAPHIC] [TIFF OMITTED] T4902.110

[GRAPHIC] [TIFF OMITTED] T4902.111

[GRAPHIC] [TIFF OMITTED] T4902.112

[GRAPHIC] [TIFF OMITTED] T4902.113

[GRAPHIC] [TIFF OMITTED] T4902.114

[GRAPHIC] [TIFF OMITTED] T4902.115

[GRAPHIC] [TIFF OMITTED] T4902.116

[GRAPHIC] [TIFF OMITTED] T4902.117

[GRAPHIC] [TIFF OMITTED] T4902.118

[GRAPHIC] [TIFF OMITTED] T4902.119

[GRAPHIC] [TIFF OMITTED] T4902.120

[GRAPHIC] [TIFF OMITTED] T4902.121

[GRAPHIC] [TIFF OMITTED] T4902.122

[GRAPHIC] [TIFF OMITTED] T4902.123

[GRAPHIC] [TIFF OMITTED] T4902.124

[GRAPHIC] [TIFF OMITTED] T4902.125

[GRAPHIC] [TIFF OMITTED] T4902.126

[GRAPHIC] [TIFF OMITTED] T4902.127

[GRAPHIC] [TIFF OMITTED] T4902.128

[GRAPHIC] [TIFF OMITTED] T4902.129

[GRAPHIC] [TIFF OMITTED] T4902.130

[GRAPHIC] [TIFF OMITTED] T4902.131

[GRAPHIC] [TIFF OMITTED] T4902.132

[GRAPHIC] [TIFF OMITTED] T4902.133

[GRAPHIC] [TIFF OMITTED] T4902.134

[GRAPHIC] [TIFF OMITTED] T4902.135

[GRAPHIC] [TIFF OMITTED] T4902.136

[GRAPHIC] [TIFF OMITTED] T4902.137

[GRAPHIC] [TIFF OMITTED] T4902.138

[GRAPHIC] [TIFF OMITTED] T4902.139

[GRAPHIC] [TIFF OMITTED] T4902.140

[GRAPHIC] [TIFF OMITTED] T4902.141

[GRAPHIC] [TIFF OMITTED] T4902.142

[GRAPHIC] [TIFF OMITTED] T4902.143

[GRAPHIC] [TIFF OMITTED] T4902.144

[GRAPHIC] [TIFF OMITTED] T4902.145

[GRAPHIC] [TIFF OMITTED] T4902.146

[GRAPHIC] [TIFF OMITTED] T4902.147

[GRAPHIC] [TIFF OMITTED] T4902.148

[GRAPHIC] [TIFF OMITTED] T4902.149

[GRAPHIC] [TIFF OMITTED] T4902.150

[GRAPHIC] [TIFF OMITTED] T4902.151

[GRAPHIC] [TIFF OMITTED] T4902.152

[GRAPHIC] [TIFF OMITTED] T4902.153

[GRAPHIC] [TIFF OMITTED] T4902.154

[GRAPHIC] [TIFF OMITTED] T4902.155

[GRAPHIC] [TIFF OMITTED] T4902.156

[GRAPHIC] [TIFF OMITTED] T4902.157

[GRAPHIC] [TIFF OMITTED] T4902.158

[GRAPHIC] [TIFF OMITTED] T4902.159

[GRAPHIC] [TIFF OMITTED] T4902.160

[GRAPHIC] [TIFF OMITTED] T4902.161

[GRAPHIC] [TIFF OMITTED] T4902.162

[GRAPHIC] [TIFF OMITTED] T4902.163

[GRAPHIC] [TIFF OMITTED] T4902.164

[GRAPHIC] [TIFF OMITTED] T4902.165

[GRAPHIC] [TIFF OMITTED] T4902.166

[GRAPHIC] [TIFF OMITTED] T4902.167

[GRAPHIC] [TIFF OMITTED] T4902.168

[GRAPHIC] [TIFF OMITTED] T4902.169

[GRAPHIC] [TIFF OMITTED] T4902.170

[GRAPHIC] [TIFF OMITTED] T4902.171

[GRAPHIC] [TIFF OMITTED] T4902.172

[GRAPHIC] [TIFF OMITTED] T4902.173

[GRAPHIC] [TIFF OMITTED] T4902.174

