This is the accessible text file for GAO report number GAO-05-223
entitled 'Credit Reporting Literacy: Consumers Understood the Basics
but Could Benefit from Targeted Educational Efforts' which was released
on March 16, 2005.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Report to Congressional Committees:
March 2005:
Credit Reporting Literacy:
Consumers Understood the Basics but Could Benefit from Targeted
Educational Efforts:
[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-05-223]:
GAO Highlights:
Highlights of GAO-05-223, a report to congressional committees:
Why GAO Did This Study:
This report responds to a mandate in the Fair and Accurate Credit
Transactions Act (FACT Act) of 2003 requiring GAO to assess consumers’
understanding of credit reporting. The FACT Act, among other things,
extended provisions governing the credit reporting system and addressed
ongoing concerns about inaccuracies in credit reports. For example, the
act expanded access to credit information by entitling consumers to one
free credit report each year. It also established the Financial
Literacy and Education Commission (FLEC) to improve consumers’
understanding of credit issues. This report examines consumers’
understanding and use of credit reports and scores and the dispute
process and looks at factors that may influence their understanding of
credit reporting.
What GAO Found:
Based on survey responses for a national sample of 1,578 consumers, GAO
found that consumers understood the basics of credit reporting and the
dispute process. For example, many consumers understood what a credit
report contained and the sources of this information, and about 60
percent had seen their credit reports. However, many consumers did not
know more detailed information, such as how long items remained on
their credit reports or the impact their credit history could have on
insurance rates and potential employment. Further, most consumers knew
what a credit score was, and approximately one-third had obtained their
credit scores, but many did not know that some behaviors—such as using
all their available credit—could negatively affect their scores.
Similarly, GAO found that most consumers knew they had the right to
dispute information on their credit reports, and a small percentage (18
percent) had disputed inaccuracies. But most consumers did not fully
understand their rights in the dispute process—for example, that there
is no cost to dispute inaccurate information or that they could contact
the Federal Trade Commission, the federal agency primarily responsible
for enforcing consumers’ rights with respect to credit reporting
agencies (CRAs), if they could not resolve a dispute with the CRAs.
GAO also found that several factors were associated with consumers’
knowledge. For instance, having less education, lower incomes, and less
experience obtaining credit were associated with lower survey scores,
while having certain types of credit experiences—such as an automobile
loan or a mortgage—were associated with higher scores. Other factors,
such as gender and living in a state where credit reports were free
prior to the FACT Act, did not have a significant effect on consumers’
knowledge. Educational efforts could potentially increase consumers’
understanding of the credit reporting process. These efforts should
target those areas in which consumers’ knowledge was weakest and those
subpopulations that did not score as well on GAO’s survey.
Consumers’ Knowledge of Credit Reporting:
What GAO Recommends:
GAO recommends that (1) the Secretary of Treasury, as Chairman of FLEC,
working with its members, take steps to improve consumers’
understanding of their rights and remedies under the FACT Act,
targeting the population groups that would most benefit; and (2) the
Chairman of the Federal Trade Commission, take steps to improve
consumers’ understanding of how credit reports and scores are used,
their right to dispute inaccurate information, and how consumers’
credit behavior could affect their credit history. Both agencies
generally agreed with the findings.
www.gao.gov/cgi-bin/getrpt?GAO-05-223.
To view the full product, including the scope and methodology, click on
the link above. For more information, contact Yvonne D. Jones, (202)
512-8678, jonesy@gao.gov.
[End of section]
Contents:
Letter:
Background:
Results in Brief:
Consumers Generally Understood Credit Reports and Had Viewed Their
Reports, but Many Lacked Specific Knowledge:
Consumers Had a Basic Understanding of Credit Scores, and About One-
Third Had Obtained Their Scores:
Consumers' Knowledge of the Dispute Process Was Limited, and Few
Consumers Had Disputed Information:
Certain Demographics and Credit Experiences Are Associated with
Consumers' Understanding of Credit Reporting Issues:
Conclusions:
Recommendations for Executive Action:
Agency Comments and Our Evaluation:
Appendixes:
Appendix I: Objectives, Scope, and Methodology:
Scope and Methodology:
Design of the Telephone Sample:
Developing the Questionnaire:
Administering the Survey:
Survey Response:
Survey Error and Data Quality:
Survey Data Analysis:
Interviews and Secondary Analysis:
Appendix II: Survey Questionnaire:
Appendix III: Review of Selected Findings from Other Studies:
Appendix IV: Certain Demographics and Credit Experiences Are Associated
with Consumers' Understanding of Credit Reporting Issues:
Those with Less Education and Lower Incomes Scored Lower on Our Survey
Race/Ethnicity Was Associated with Differences in Survey Scores:
Younger and Older Consumers Scored Lower on Our Survey:
Actively Employed Consumers Scored Slightly Higher on Our Survey:
Consumers with Credit Experiences Scored Higher on Our Survey:
Being a Victim of Identity Theft and Being from a Free Report State Had
Little Effect on Consumers' Survey Scores:
Appendix V: A Regression Model to Explain Differences in Total
Knowledge Scores:
Variables in the Model:
The Structure of the Model:
The Statistical Results:
Using the Regression to Illustrate the Simultaneous Influences of
Several Variables:
Appendix VI: Web Appendix on Credit Reporting Literacy Survey Data
Available at http://www.gao.gov/cgi-bin/getrep?GAO-05-411SP:
Appendix VII: Comments from the Department of the Treasury:
Appendix VIII: Comments from the Federal Trade Commission:
GAO Comments:
Appendix IX: GAO Contacts and Staff Acknowledgements:
GAO Contacts:
Staff Acknowledgements:
Tables:
Table 1: Selected Survey Responses by Race/Ethnicity:
Table 2: Experience with Credit-Related Products and Race/Ethnicity:
Table 3: Sample Dispositions and Response Rates:
Table 4: Review of Selected Findings from Other Studies:
Table 5: Selected Survey Responses and Educational Level:
Table 6: Selected Survey Responses of Different Household Income
Levels:
Table 7: Selected Survey Responses by Race/Ethnicity:
Table 8: Experience with Credit-Related Products and Race/Ethnicity:
Table 9: Selected Survey Responses by Age Group:
Table 10: Selected Survey Responses by Employment Group:
Table 11: Selected Survey Responses and Consumers' Credit Reporting
Experience:
Table 12: Selected Survey Responses and Experience with Automobile
Loans or Home Loans/Mortgages:
Table 13: Variables in the Statistical Model:
Table 14: A Regression Model to Explain Differences in Total Knowledge
Scores:
Table 15: Comparing Two Hypothetical Hispanics Using the Regression
Model:
Figures:
Figure 1: Selected Survey Responses to Some Key Questions about Credit
Reports:
Figure 2: Consumers' Understanding of the Information Credit Reports
Contain:
Figure 3: Reasons Consumers Viewed their Reports:
Figure 4: Factors Affecting Credit Scores:
Figure 5: Consumers' Knowledge of the Factors That Affect Credit
Scores:
Figure 6: Reasons Consumers Disputed Information on Their Credit
Reports:
Figure 7: Average Total Scores and Demographic Groups:
Figure 8: Credit Reporting Experience by Race/Ethnicity:
Figure 9: Average Total Scores and Credit Experiences:
Figure 10: Average Total Scores and Demographic Groups:
Figure 11: Credit Reporting Experience by Educational Level:
Figure 12: Credit Reporting Experience and Household Income Level:
Figure 13: Credit Reporting Experience by Race/Ethnicity:
Figure 14: Credit Reporting Experience by Age Group:
Figure 15: Credit Reporting Experience by Employment Group:
Figure 16: Average Total Scores and Credit Experience:
Figure 17: Credit Reporting Experience and Experience with Automobile
Loans or Home Loans/Mortgages:
Abbreviations:
AARP: American Association of Retired Persons:
ASEC: American Savings Education Council:
CDIA: Consumer Data Industry Association:
CFA: Consumer Federation of America:
CFTC: Commodity Futures Trading Commission:
CRA: Credit Reporting Agency:
CRC: Credit Research Center of Georgetown University:
CRS: Congressional Research Service:
EBRI: Employee Benefit Research Institute:
FACT Act: Fair and Accurate Credit Transactions Act of 2003:
Fannie Mae: Federal National Mortgage Association:
FCRA: Fair Credit Reporting Act:
FDIC: Federal Deposit Insurance Corporation:
FLEC: Financial Literacy and Education Commission:
FRB: Federal Reserve Board:
Freddie Mac: Federal Home Loan Mortgage Corporation:
FTC: Federal Trade Commission:
La Raza: National Council of La Raza:
NCUA: National Credit Union Administration:
OCC: Office of the Comptroller of the Currency:
OTS: Office of Thrift Supervision:
SEC: Securities and Exchange Commission:
SSA: Social Security Administration:
U.S. PIRG: U.S. Public Interest Research Group (also known as the
National: Association of State PIRGs):
Letter March 16, 2005:
The Honorable Richard C. Shelby:
Chairman:
The Honorable Paul S. Sarbanes:
Ranking Minority Member:
Committee on Banking, Housing, and Urban Affairs:
United States Senate:
The Honorable Michael G. Oxley:
Chairman:
The Honorable Barney Frank:
Ranking Minority Member:
Committee on Financial Services:
House of Representatives:
Credit reports detailing personal credit histories and the credit
scores derived from these reports affect many aspects of consumers'
lives. Both credit reports and credit scores can influence lenders'
decisions to grant credit and may affect individuals' ability to obtain
jobs, insurance, and rental housing. They can also affect interest
rates offered on automobile, mortgage, and other consumer loans. For
example, having a low credit score could add almost $370 dollars to a
monthly mortgage payment on a 30-year home mortgage of $150,000
compared with a payment calculated using a higher credit
score.[Footnote 1] Because of the importance of credit reports and
credit scores, consumers need to know and understand what their reports
contain, and many credit experts suggest that it is prudent practice
for consumers to check the accuracy and completeness of their credit
report information periodically.
As GAO has previously reported, inaccurate credit report data could
have important implications for both consumers and creditors in today's
sophisticated credit markets.[Footnote 2] Equifax, Experian, and
TransUnion--the three national credit reporting agencies
(CRAs)[Footnote 3]--each maintain an estimated 200 million credit files
and enter more than 4 billion pieces of data into these files
monthly.[Footnote 4] While CRAs take measures to minimize errors in
data files, the volume of data they handle increases the possibility of
inaccuracies on credit reports. Further, personal credit histories can
be damaged by identity theft--the use of another person's identity to
obtain credit cards, access bank accounts, and commit other fraudulent
acts. Identity theft is considered a fast-growing, white-collar crime
and poses a direct threat to the accuracy and integrity of credit
report data.
CRAs have been subject to specific federal regulation since the passage
of the Fair Credit Reporting Act (FCRA) in 1970.[Footnote 5] FCRA gave
the Federal Trade Commission (FTC) responsibility for enforcing CRAs'
compliance with the act.[Footnote 6] In December 2003, the Fair and
Accurate Credit Transactions Act (FACT Act) was enacted, which amended
FCRA. Among other things, the act included provisions to improve the
accuracy of personal information assembled by CRAs and better provide
for the fair use of and consumer access to such information.[Footnote
7] In addition, the act created the Financial Literacy and Education
Commission (FLEC) to improve financial literacy and education through
the development of a national strategy, a Web site, and a toll-free
hotline and designated the Secretary of the Treasury as the chairperson
of FLEC.[Footnote 8] The act directs FLEC to emphasize basic personal
income and household money management and planning skills, including
increased awareness of the importance of (1) credit reports and credit
scores in obtaining credit and on the terms of credit, (2) accuracy in
credit reports and scores, (3) correcting inaccuracies, and (4) the
effects common financial decisions can have on credit scores. In
addition, Congress mandated that the Secretary of the Treasury, after
reviewing FLEC's recommendations, implement and conduct a national
public service multimedia campaign to improve financial literacy and to
publicize FLEC's Web site and toll-free hotline. The FACT Act also
mandated that GAO evaluate consumers' knowledge and experience with
credit reporting.[Footnote 9] As agreed with your offices, this report
responds to the FACT Act mandate by examining the extent to which
consumers (1) understand and review their credit reports, (2)
understand and review their credit scores, and (3) know how to dispute
information on their credit reports and actually do so. This report
also discusses some of the factors that are associated with consumers'
understanding of these issues.
To meet these objectives, we contracted with a survey research firm to
conduct a telephone survey of randomly selected adults in the United
States. We designed the survey questions to assess respondents'
knowledge of and experience with credit reports, credit scores, and the
dispute resolution process. This effort resulted in 1,578 completed
interviews, with a response rate of 48 percent and a cooperation rate
of 59 percent (see appendix I for a discussion of response
rates).[Footnote 10] The survey was conducted in both English and
Spanish. The results can be generalized to the population of U.S.
adults aged 18 and older, and we refer to respondents as "consumers"
throughout the report. The survey fieldwork was conducted from late
July to early October 2004. Using the survey results, we applied two
methodologies--cross-tabulations and regression--to analyze the effects
that selected factors, such as specific experiences and demographic
factors, had on credit reporting knowledge. The results of the two
methods were generally consistent with one another. Appendix IV
discusses the results of the cross-tabulations analysis and appendix V
discusses the regression analysis results.
In developing the survey questions and to provide context, we conducted
a literature search to identify other surveys and documents related to
credit reporting issues. Appendix III contains our review of selected
findings from other studies. In addition to our review, we interviewed
members of FLEC, including officials from the federal financial
regulatory agencies, FTC, the Department of the Treasury, and other
relevant federal agencies. We also met with industry organizations,
including furnishers of consumer credit data, the three nationwide
CRAs, and consumer advocacy groups that educate and inform consumers on
credit issues and the law. We discussed and collected data from these
officials on consumers' knowledge of credit reports, credit scores, and
the dispute resolution process and other credit-related issues. We used
this information for comparative and background purposes only and did
not verify its accuracy. Appendix I contains additional details on our
scope and methodology, and appendix II reproduces a copy of our survey
with the frequency of responses by question. In addition, an electronic
appendix (app. VI) including additional subgroup results can be found
on GAO's Web site at [Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-
05-411SP. We conducted our review from March 2004 through February 2005
in accordance with generally accepted government auditing standards.
Background:
CRAs are private sector companies that receive information from
businesses that offer credit and from other sources and compile this
information into credit reports that are sold for a fee to consumers
and other businesses.[Footnote 11] The U.S. credit reporting system is
voluntary, as federal law does not require lenders and other creditors
to report to CRAs. Creditors may report consumer information to one,
all, or none of the CRAs, and information on individual consumers may
differ among the three agencies. Lenders rely on credit reports when
deciding whether to offer credit to an individual, at what rate and on
what terms.[Footnote 12] In addition, a growing number of
decisionmakers, including potential employers, insurance underwriters,
and landlords are using credit reports to assess applicants'
creditworthiness.
The credit reports the CRAs provide generally contain certain standard
information, including:
* Credit histories that list consumers' experiences with credit and
credit behavior--for example, any loans and credit cards on record and
payment histories for each;
* Identifying information such as names, addresses, Social Security
numbers, and employment data (other than salary);
* Information related to monetary transactions that is a matter of
public record, such as tax liens (claims against property for unpaid
taxes), debt-related judgments or court orders, and bankruptcies; and:
* Inquiries made by others--for instance, mortgage lenders or banks--
about consumers' credit histories.
In addition to compiling credit reports, CRAs also calculate credit
scores, which attempt to predict the likelihood that a person will
repay a loan. Credit scores, which help creditors analyze information
in credit reports and make decisions on granting credit, are derived
using a mathematical model that takes into account the information
contained in the reports. In the 1950s, Fair Isaac Corporation
developed one of the first mathematical models for scoring credit
applications.[Footnote 13] Before credit scores became widely
available, lenders or credit examiners would evaluate information in
credit reports to decide whether or not to grant credit. Over time,
credit scores were viewed as a more efficient and consistent method for
evaluating information in credit reports, and by the early 1990s all
three nationwide CRAs had begun selling them to businesses along with
credit reports.
In 1970, Congress enacted FCRA--the primary federal legislation
governing the content and use of credit reports and the credit
reporting industry in general. The act contained several important
measures. For instance, FCRA:
* Provided a uniform basis for consumers' access to information in
their credit reports at a reasonable charge;[Footnote 14]
* Subject to certain exceptions, prohibited CRAs from reporting
specific types of negative information that was older than 7
years;[Footnote 15]
* Required CRAs to identify for consumers, on request, sources of
credit report information and the names of those receiving a credit
report within the time periods specified in the act;[Footnote 16]
* Required CRAs to provide free reports to any consumer who requested a
copy of a report following an adverse action, such as failure to obtain
credit, insurance, or employment, within a month of the action; and:
* Required CRAs to implement a dispute resolution process to
investigate and correct errors and to remove information found to be
erroneous.
The dispute resolution process established in FCRA required CRAs to
investigate within a reasonable period of time items that consumers
reported as inaccurate or incomplete. CRAs were required to delete
promptly any disputed data that they could not verify within that
period. For unresolved disputes, FCRA allowed consumers to add a brief
explanatory statement to their credit files. When information was not
deleted, CRAs were required to note the dispute in subsequent reports
containing the information in question and include the consumer's
statement or a clear summary of it. In addition, the act required CRAs,
upon the consumer's request, to notify those who received the reports
that information had been deleted or that the consumer had filed a
dispute statement.[Footnote 17] Finally, FCRA gave FTC responsibility
for enforcing CRA's compliance with the act to the extent that this
authority did not overlap the authority of other federal agencies
specified in the act.
Despite the consumer protections FCRA offered, starting in the late
1980s consumers began raising new concerns about credit reports and
CRAs. For example, consumers maintained that CRAs were not responding
to consumers' requests for assistance. In addition, lawsuits were
brought against the three CRAs on issues related to accuracy.[Footnote
18] In 1996, Congress amended FCRA.[Footnote 19] Among other things,
the 1996 amendments required CRAs to provide consumers, upon request,
with access to all information in their credit files (except credit
scores) at a cost not to exceed $8.00, improved the process for
investigating disputed information, and permitted use of credit reports
only for certain purposes, such as solicitations for credit and
insurance. The time period for investigations of a dispute was narrowed
from a "reasonable period of time" to, in general, 30 days, and CRAs
were required to conduct investigations at no cost to the consumer. The
1996 amendments also added the definition of an "adverse action" and
gave consumers access to a free credit report each year if they were
unemployed but intended to seek employment, were on public assistance,
or suspected that a credit report contained inaccurate information due
to fraud.
To extend the expiring FCRA preemption provisions and address
continuing concerns about inaccuracies in reports, identity theft, and
other matters, Congress passed the Fair and Accurate Credit
Transactions Act of 2003 (the FACT Act), which amended FCRA.[Footnote
20] Had the provisions expired, states generally could have enacted
laws covering matters that had been governed exclusively by FCRA.
Industry participants and others believed that national standards and
uniformity of credit information were key to maintaining efficiency in
credit markets and that the loss of federal preemption in those areas
that had been exclusively subject to FCRA would threaten this
efficiency. Among other things, the amendments:
* Expanded consumers' access to credit information and sought to
promote accuracy within credit reports by, among other things,
entitling consumers in all states to one free credit report each year
from each of the three CRAs.[Footnote 21] However, according to FTC,
seven states already allowed consumers at least one free copy annually
(Colorado, Georgia, Maine, Maryland, Massachusetts, New Jersey, and
Vermont);"[Footnote 22]
* Expanded creditors' responsibilities for furnishing accurate
data.[Footnote 23] The FACT Act mandated that the federal banking
agencies, National Credit Union Association (NCUA), and FTC issue joint
regulations identifying the circumstances under which furnishers of
information would be required to investigate the accuracy of
information in a credit report when asked to do so by a
consumer;[Footnote 24]
* Established a regulatory scheme under which creditors were required
to notify consumers when information from credit reports negatively
affected the terms of credit, such as interest rates on credit cards
and loans. The notification would also have to inform potential
borrowers that they could obtain a copy of their credit report from the
CRA, without charge; and:
* Strengthened victims' rights with respect to identity theft, made
credit history restoration easier, limited the sharing and use of
medical information within the financial system, required mortgage
lenders to disclose credit scores to applicants, and required CRAs to
offer credit scores to consumers for a reasonable fee.
Finally, as we have noted, the FACT Act created FLEC to improve
financial literacy and education and, among other things, required FLEC
to increase consumers' awareness and understanding of credit reports
and scores. Several public and private entities had already undertaken
and continue to operate initiatives to increase consumers' knowledge of
credit reporting issues. Within the federal government, a number of
agencies sponsor financial literacy initiatives that include a credit
literacy component, such as the Federal Deposit Insurance Corporation's
(FDIC) Money Smart program, the Department of Defense's Financial
Readiness Campaign, and the Federal Reserve's nationwide financial
education campaign, "There's a Lot to Learn about Money." In order to
fulfill its mandate to protect consumers from deceptive and unfair
business practices in the marketplace, FTC engages in several
activities, including maintaining a Web site that features educational
information on credit reporting and other financial topics and a toll-
free number for complaints and questions.[Footnote 25] Some private
sector initiatives include the Federal Home Loan Mortgage Corporation's
(Freddie Mac) CreditSmart® Program, which is devoted exclusively to
promoting credit literacy and features modules on interpreting credit
reports. Fair Isaac Corporation maintains a Web site to disseminate
information on credit education.[Footnote 26] In addition, all three
nationwide CRAs engage in educational efforts through their Web sites
and partnerships with national organizations or schools.
GAO has also addressed the issue of financial literacy among consumers.
The most recent effort was a forum hosted on July 28, 2004, on the role
of the federal government in improving financial literacy.[Footnote 27]
Forum participants included experts in financial literacy and education
from federal and state agencies, the financial industry, nonprofit
organizations, and academic institutions. During the course of the
forum, credit experts agreed that consumers would benefit from more
education on credit, including the cost of credit, how to use credit,
and how to manage credit responsibly. Forum participants also advocated
targeting low-and moderate-income individuals and families, immigrant
populations, and young people in these education efforts. Similarly, a
Federal Reserve study on financial education found that successful
financial education programs target specific groups, such as youth or
minority populations.[Footnote 28]
Results in Brief:
Our survey showed that most consumers understood the basic concepts of
credit reporting and had seen their credit reports but that many lacked
knowledge of some credit report contents, uses, and other impacts.
Based on our survey responses, we found that many consumers knew how to
obtain their credit reports, most of the information reports contained,
some impacts of their credit history--including its effect on their
ability to obtain loans and the interest rate on loans--and some of the
rules governing access to their reports. However, many consumers lacked
other important information, such as how long information stayed on
their reports and how their credit history could affect insurance
premiums and employment. For example, about two-thirds of consumers did
not know that their credit history could impact their insurance
premiums and employment, and about half (53 percent) did not know that
information could stay on their report for 7 or 10 years. We found that
many consumers (almost 60 percent) had viewed their credit reports,
most often because they were making a large purchase or refinancing a
loan, and most of these consumers said that they understood their
reports.
The results of our survey also showed that most consumers had a basic
understanding of credit scores and that around one-third had obtained
them. While many consumers (70 percent) correctly identified the
definition of a credit score and understood many of the factors that
could impact credit scores, only 28 percent could provide a number
within a range of possible credit scores. In addition, consumers were
more familiar with some of the factors that affected credit scores than
with others. For example, while most consumers knew that skipping loan
payments or making late credit card payments had a negative effect on
credit scores, about half did not know that using all the credit
available to them, such as reaching the maximum limit on a credit card
or home equity loan, had a negative effect. Also, when asked about
information that had no effect on credit scores (such as a low checking
account balance), about half of consumers answered the questions
incorrectly or said that they did not know.
Most consumers (around 90 percent) knew that they could dispute
inaccurate information on their credit reports, and approximately 18
percent said that they had initiated a dispute. However, we found that
over half of consumers did not fully understand their rights in the
dispute process or the responsibilities of the CRAs in responding to
disputes. For example, relatively few consumers--about one-third--knew
that CRAs investigate disputed information for free. In addition, the
majority of consumers (94 percent) did not name FTC as the federal
agency they would contact with a complaint about a CRA--for example,
about an unresolved dispute--although FTC is the federal agency
primarily responsible for enforcing consumers' rights with respect to
credit reporting by CRAs. The types of information consumers disputed
varied, but among the items most frequently mentioned were information
that did not belong to the individual and incorrect payment histories--
for example, late payments. We also found that of those who had
disputed information, most (69 percent) reported that they had
information removed from their report.
Many factors are associated with consumers' knowledge of credit
reporting issues. We found that some demographic factors and types of
experiences appeared to affect consumers' scores on our survey. For
example, less educated, low-income, and Hispanic consumers scored lower
on our survey than other demographic groups. Additionally, we found
that having experience with credit--for example, obtaining an
automobile loan or a mortgage--and the credit reporting system appeared
to raise consumers' scores. In contrast, we found that other factors
and experiences, such as having been a victim of identity theft,
gender, or living in one of the seven "free report states," had little
effect on consumers' credit reporting knowledge. The factors that
appeared to affect consumers' knowledge of credit issues had a
cumulative impact. For example, a consumer with a college degree,
credit experience, and a high income could score much higher than a
consumer with less than a high school education, relatively low income,
and no credit experience.
This report makes recommendations to the Secretary of Treasury and the
Chairman of FTC that are designed to promote efforts to improve the
credit reporting literacy of U.S. consumers. We provided a draft of
this report to the Department of the Treasury and FTC for comment and
they provided written comments that are reprinted in appendixes VII and
VIII, respectively. Both agencies generally agreed with our findings
and outlined efforts underway to improve credit reporting literacy. A
summary of their written comments and our response are presented at the
end of this report.
Consumers Generally Understood Credit Reports and Had Viewed Their
Reports, but Many Lacked Specific Knowledge:
On the basis of responses to our questionnaire, we found that consumers
generally understood credit reports and had viewed their reports but
that many lacked specific knowledge about what their credit reports
contained, how they were used, and other potential impacts of their
credit history. Correct responses to individual questions varied
widely, ranging from a high of 95 percent who knew that their credit
history could affect their ability to get a loan to a low of 7 percent
of consumers who knew that the credit reporting system was voluntary.
We found that basic questions generated the most correct answers. For
example, most consumers could correctly choose a description of a CRA
and were aware that credit reports contained information such as credit
card balances. But many consumers did not know other important
information, such as how long information remained on their report and
the impact their credit history could have on insurance and employment.
We also found that more than half of consumers--58 percent--said that
they had viewed their reports, generally before making a large purchase
or refinancing a mortgage.
Consumers' Average Survey Scores Were Slightly More Than 50 Percent:
We designed our survey to assess consumers' awareness of three distinct
aspects of credit reporting--credit reports, credit scores, and the
dispute resolution process--and collected responses from a random
probability sample of people throughout the United States. Our survey
included 58 questions. Twenty-three survey questions (worth a total of
56 points) were created to test consumers' knowledge of credit
reporting issues. In addition, we incorporated 22 questions that were
designed to obtain consumers' opinions about these issues and to
examine their experiences with the credit reporting process, along with
13 demographic questions.[Footnote 29] The mean score or average score
on the survey was 55 percent, measured as the percentage of correct
answers to the 23 questions designed to test knowledge (see app. I).
Many Consumers Knew How to Obtain Their Credit Reports:
Many consumers appeared to know how to obtain a copy of their credit
report. For example, 25 percent of consumers said that they would go to
a CRA, and an additional 6 percent named a specific CRA. Twenty-six
percent of consumers said that they would go online, and 3 percent said
that they would go online to a specific CRA. Some consumers who did not
identify a direct source of credit reports, such as a specific CRA,
said that they would ask a bank or financial institution, a credit card
company, or the institution where they had applied for a loan--all
places that could direct them to a CRA or another means of obtaining
their report. Seventy-one percent of consumers also said that they
understood that they could order a copy of their credit report at any
time for any reason. The Consumer Federation of America (CFA) found in
its July 2003 study that the majority of its respondents--97 percent--
knew that consumers had the right to see their credit report.[Footnote
30] Appendix III contains more results from relevant studies.
Before consumers answered any questions about credit reports, our
survey asked them to provide their own definition of what a CRA did.
Based on their responses, we determined whether consumers had no
knowledge, their knowledge was unclear, or they appeared knowledgeable.
We found that 19 percent of consumers were able to correctly articulate
what a CRA does but that 53 percent were unclear on this point. Another
28 percent of consumers' explanations were clearly incorrect, or the
consumers said that they did not know what a CRA did.[Footnote 31]
However, as figure 1 shows, the majority of consumers--82 percent--were
able to select the correct description of a CRA from among multiple
choices. Only 19 percent of consumers answered the multiple choice
question incorrectly or said that they did not know.
Figure 1: Selected Survey Responses to Some Key Questions about Credit
Reports:
[See PDF for image]
Note: Percentages in each row may not equal 100 due to rounding. All
survey questions (in the format in which they were asked) and their
responses can be found in appendix II.
[End of figure]
However, many consumers did not know specific information related to
obtaining their reports, such as the names of the CRAs and the cost of
ordering credit reports. Twenty percent of consumers correctly named
one CRA, 10 percent were able to name two, and 3 percent were able to
name three. Sixty-seven percent of consumers were unable to name any of
the CRAs. CFA's 2003 survey found that 75 percent of respondents were
unable to name any CRAs and that just 15 percent could name one
CRA.[Footnote 32] Further, fewer than half of our respondents--
approximately 45 percent--knew the cost of a credit report.[Footnote
33] Before the implementation of the FACT Act, consumers generally had
to pay for copies of their credit report in all but a few
circumstances.[Footnote 34] According to FTC, the laws of seven states
(Colorado, Georgia, Maine, Maryland, Massachusetts, New Jersey, and
Vermont) provided for free access to credit reports, upon request, at
least annually. We asked consumers throughout the United States whether
they lived in one of these "free report states" and found that 49
percent of those consumers who did were aware of it. Forty-six percent
of these consumers did not know that they lived in a free report state,
and 5 percent responded incorrectly.
In addition, in our survey (conducted after the passing of the FACT Act
but prior to its implementation) we also asked consumers whether a new
law affecting their rights regarding credit reports and scores would
entitle all consumers to request one free credit report a year. Forty-
seven percent correctly responded that the new law did give consumers
this right, 49 percent said that they did not know, and 4 percent
responded incorrectly.
The Majority of Consumers Knew Most of What Was in Their Credit
Reports:
In addition to generally knowing how to obtain their credit reports, we
found that a majority of consumers could also identify most of the
information contained in their reports. As shown in figure 2, most
consumers knew that their credit reports contained their Social
Security numbers, credit payment history, and any bankruptcies. In
addition, 92 percent of consumers answered correctly that credit card
companies could report unpaid bills. In July 2003, the Federal Reserve
reported data showing that 81 percent of their survey respondents knew
that credit reports included employment data, payment histories, any
inquiries made by creditors, and information from public records (see
app. III).[Footnote 35]
Figure 2: Consumers' Understanding of the Information Credit Reports
Contain:
[See PDF for image]
[End of figure]
With respect to our survey data, we found that fewer consumers knew
what their reports might not contain. Fifty-nine percent of consumers
knew that their reports did not contain information on race, but 15
percent of consumers incorrectly thought race was included, and 25
percent did not know. Fifty-eight percent of consumers knew that
balances in their checking accounts were not on their report, while 21
percent of consumers thought balances were on their reports, and 21
percent did not know. We also asked consumers whether they thought
income was on their credit reports, and 37 percent correctly responded
that it was not, while 20 percent did not know if it was included.
While some confusion exists as to whether income is included on credit
reports, we spoke with all three of the nationwide CRAs to determine if
they included income in credit reports. Representatives from all three
nationwide CRAs told us that their credit reports do not show
information on consumers' incomes. In addition, according to the three
CRAs, the format CRAs use in creating credit reports for individuals--
the Metro 2 format--does not include a field for income, and CRAs do
not collect this information. However, in commenting on a draft of this
report, FTC officials told us that income sometimes, although rarely,
does appear on credit reports. Based on our discussions, we considered
the responses of consumers who said that income was not included in
their credit reports as correct.
In addition, before asking consumers multiple choice questions about
credit report contents, we asked them to provide their own definition
of a credit report. We found that 19 percent of consumers appeared
knowledgeable, but the level of knowledge was unclear for an estimated
63 percent. We found that 14 percent of consumers gave clearly
incorrect responses or responses showing that they did not know.
Most consumers also knew some sources of the information on their
credit reports, although few knew that the credit reporting system was
voluntary. Eighty-nine percent of consumers knew that banks and credit
card companies provided information to CRAs, and 79 percent knew that
debt collectors did. Fewer consumers--46 percent--knew that information
on their credit report could come from courthouse and other public
records. Still fewer--7 percent--knew that lenders were not required to
report information to any CRA. We found that 57 percent of consumers
did not know how many CRAs lenders must report to, and 25 percent
incorrectly thought that lenders were required to report to all of the
CRAs. Because the credit reporting system is voluntary, consumers need
to understand that their credit reports may differ from one another and
may not contain all information related to their credit history. For
example, if a consumer is making timely loan payments to a lender that
does not report to any of the CRAs, this payment history will not be
included on the consumer's credit reports.
We also asked consumers if they knew how long information stayed on
their report. We wanted to examine whether consumers were aware of the
full impact their credit behaviors, such as making late payments or
filing for bankruptcy, could have on their credit report. We found that
47 percent of consumers knew that information could generally stay on
their report for 7 or 10 years.[Footnote 36] Twenty-seven percent of
consumers said that they did not know how long information remained on
their reports, and 26 percent answered this question incorrectly.
Most Consumers Knew Some of the Possible Impacts of Credit Reports and
Some Rules Governing Access:
We found that most consumers understood some of the ways their credit
history--as contained in their credit reports--could affect their
lives, but around one-third were not aware of several potential uses of
their reports. For instance, most consumers (95 percent and 72 percent,
respectively) understood that the information in their credit reports
could affect their ability to obtain loans and rent housing (fig. 1).
However, fewer consumers (36 percent and 33 percent, respectively)
understood that their credit histories could affect insurance coverage
and premiums or that potential employers could use credit reports when
making hiring decisions.
We also asked consumers about rules governing access to this
information. Specifically, we wanted to determine whether consumers
understood when their consent was needed to release their credit
reports. For example, while only about one-third of consumers
understood that potential employers could use credit reports when
making hiring decisions, 61 percent knew that employers could not
obtain credit reports on potential or current employees without
consent. We also found that 85 percent of consumers knew that anyone
who wanted to view their report (not in connection with a product or
service) could not do so without their consent. Seventy-six percent
knew that someone with a court order had the right to see their report.
Fewer consumers--55 percent--answered correctly that someone who needed
to view their credit history in connection with a product or service
that they had requested could access their credit report without
consent.
In another study on credit scores and consumer credit, CFA and
Providian® found that the majority of respondents--81 percent--knew
that mortgage lenders could use their credit history (in the form of a
credit score) to determine whether they would receive a mortgage loan
and the cost of the loan. Fewer respondents--47 percent--understood the
effect of their credit history on decisions made by a home insurer.
Forty-eight percent of CFA and Providian's respondents knew the effect
their credit history could have on a landlord's decision to rent to
them.[Footnote 37]
More Than Half of Consumers Had Seen Their Credit Reports and Found
Them Understandable:
Our survey data showed that 58 percent of consumers had seen their
credit reports at some point in time and that 45 percent of this group
had viewed them within the last year. Other studies have reported that
from 43 to 63 percent of their respondents had seen their credit
reports. For example, in a July 2003 study the Federal Reserve reported
that 58 percent of consumers had reviewed their reports.[Footnote 38]
We also asked the Consumer Data Industry Association (CDIA)--the trade
group representing the CRAs--how many credit reports Equifax, Experian,
and TransUnion had provided to consumers in 2003. CDIA told us that
57.4 million credit reports were issued to consumers in 2003.[Footnote
39]
In our survey, we also asked the 58 percent of consumers who said that
they had viewed their credit reports whether they had ordered their
reports themselves or if someone else had ordered their report for
them. Of the consumers who had seen their credit reports, 53 percent
said that they had ordered their report themselves, and 47 percent said
it had been ordered by someone else for them, including:
* a mortgage company (29 percent),
* a bank or financial institution (25 percent),
* a lender in general (16 percent),
* a car dealership (12 percent),
* a credit card company (4 percent), and:
* other source (14 percent).
Of the 58 percent of consumers who told us that they had viewed their
reports, the largest percentage said that they had seen their reports
because they were making a large purchase, such as a car or home, or
were refinancing (fig. 3).
Figure 3: Reasons Consumers Viewed their Reports:
[See PDF for image]
[End of figure]
The credit reporting industry has also collected information on the
reasons consumers order their reports.[Footnote 40] In addition, in
2003, Louis Harris and Associates conducted a study that was designed
to determine consumers' interest in free credit reports and found that:
[Footnote 41]
* 39 percent of respondents looked at their credit reports because they
were curious,
* 31 percent were ready to apply for a loan or credit card,
* 10 percent had been denied credit,
* 6 percent were concerned about identity theft,
* 12 percent had some other reason, and:
* 1 percent did not know why they had seen their reports (see app.
III).
We also asked consumers who had seen their reports whether they
understood the information. The majority of them--79 percent--felt that
the information on their reports was very or somewhat easy to
understand. Eighteen percent felt that their reports were somewhat or
very hard to understand, and 3 percent did not know.
Consumers Had a Basic Understanding of Credit Scores, and About One-
Third Had Obtained Their Scores:
We found that most consumers generally understood credit scores but
lacked specific knowledge about certain factors that affected these
scores. For example, 87 percent of consumers correctly responded that
late credit card payments affected scores negatively, and 70 percent
knew the general purpose of a credit score. However, fewer--28 percent--
could identify a number within an acceptable range of possible scores.
And while many consumers were aware of some factors that could affect
their scores, such as skipping loan payments, many were also unaware of
others, such as the possible negative effect of using all available
credit. We also found that although consumers had some knowledge of
credit scores, only about one-third had actually obtained them.
Consumers Had a Basic Knowledge of Credit Scores:
Before consumers answered any other credit score questions, our survey
asked them to define a credit score. About 15 percent of consumers were
able to give answers clearly indicating that they knew the definition
of a credit score, while about 53 percent of responses were not
specific enough to indicate the extent of consumers' knowledge. Thirty-
two percent of consumers gave answers that indicated they either had an
incorrect understanding of credit scores or simply did not know the
definition. However, when asked to choose the correct definition of a
credit score from a list of choices, about 70 percent correctly
responded that a credit score predicted the likelihood that a consumer
would repay a loan.[Footnote 42] These correct responses from our
survey appear to be higher than those found in CFA and Providian's 2004
study, which found that 34 percent of consumers answered "true" to the
statement that a credit score mainly indicates the risk of repaying a
loan.[Footnote 43] On our survey, we also asked consumers to provide a
number that could be a possible credit score. About 28 percent of
consumers gave a number within the correct range (between 300 and
900).[Footnote 44]
We also asked consumers questions about the effects of certain factors
on credit scores. The mathematical models used to compute credit scores
are considered proprietary information, and companies have no
obligation to release this information to the public. However, FCRA, as
amended by the FACT Act, requires that CRAs disclose to consumers that
request the information the key factors that have adversely affected
their credit scores. In addition, though not required to disclose its
credit scoring models, the Fair Isaac Corporation has published
information on five factors that affect credit scores generated using
their models (FICO® scores).[Footnote 45] These factors are:
* Payment history: Paying bills on time can generate a higher score.
* Amounts owed: Using all available credit can result in a lower score.
* Length of credit history: Having a credit history for a relatively
long period of time can result in a higher score.
* Type of credit: Having a diverse mix of installment (mortgages and
loans) and revolving credit (credit cards) can generate a higher credit
score.
* New credit: Frequently applying for new credit can contribute to a
lower score.[Footnote 46]
According to Fair Isaac Corporation, payment history and amounts owed
have the most influence on credit scores (fig. 4).
Figure 4: Factors Affecting Credit Scores:
[See PDF for image]
[End of figure]
We found that consumers were aware of some of the most basic factors
affecting credit scores, like the importance of payment histories.
Survey data showed that more than 80 percent of consumers knew that
skipping a loan payment or making a late credit card payment could
lower their credit scores (fig. 5). Moreover, a majority of consumers
(79 percent) correctly responded that having had a credit history for a
long period of time could result in a higher score. In addition, 60
percent of consumers knew that frequently applying for credit could
have a negative effect on their scores. CFA and Providian's 2004 study
also found that most consumers understood the positive effect of making
timely payments (87 percent) and the negative effect of making payments
more than 30 days late (60 percent).[Footnote 47]
Many Consumers Were Not Familiar with Some Factors That Could Affect
Credit Scores:
We spoke with industry experts who said that while consumers generally
understood that scores help determine creditworthiness, many consumers
were confused by or unaware of various factors that affected credit
scores. Our survey showed that while consumers were aware of some
factors affecting credit scores, many lacked some knowledge in this
area. For example, of the eight questions on our survey pertaining to
the factors affecting credit scores, 52 percent of consumers missed
three or more. A complete understanding of these factors can be
important for consumers who want to improve their credit scores in
order to receive approval for credit or obtain a more favorable
interest rate.
For example, 48 percent of consumers were unaware of the negative
effect of using all available credit--the second-largest factor
influencing credit scores. It is important for consumers who want to
improve their credit scores to understand the multiple factors that
affect their scores; for example, that not charging the maximum on
their credit cards is almost as important as making timely payments. In
addition, about the same percentage did not know that having had a
credit history for a short time could lower their score. Consumers who
have had credit for a short period of time lack a history of positive
credit behavior to balance out possible bad credit decisions.
Therefore, these consumers may need additional information on the
advantages of making positive credit decisions as they begin
establishing credit histories. Finally, around half of consumers did
not know that some factors had no effect on credit scores, such as
having a low checking account balance or requesting a copy of their
credit report. The Federal Reserve's 2003 study and CFA and Providian's
2003 study also examined consumers' knowledge of the factors that
affect credit scores. For example, 45 percent of respondents to the CFA
and Providian study knew that using all available credit could lower
consumers' credit scores.[Footnote 48] The Federal Reserve's July 2003
study reported that around 60 percent of respondents answered "false"
to the following statement: "Your credit rating is not affected by how
much you charge on your credit cards."[Footnote 49]
Figure 5: Consumers' Knowledge of the Factors That Affect Credit
Scores:
[See PDF for image]
[End of figure]
Few Consumers Had Obtained Their Credit Scores:
Our survey found that about one-third of consumers had obtained their
credit scores, or about half the number of consumers who had seen their
credit reports. Fair Isaac Corporation's 2003 consumer survey found
that 31 percent of consumers had obtained their credit scores, and CFA
and Providian's 2004 survey showed that 53 percent of consumers had
obtained them.[Footnote 50] We also asked CDIA for the number of credit
score disclosures that the three nationwide CRAs made to consumers in
2003. According to CDIA, approximately 9.8 million credit score
disclosures were made in that year.[Footnote 51]
Consumers' Knowledge of the Dispute Process Was Limited, and Few
Consumers Had Disputed Information:
Consumers knew that they could dispute inaccurate information on their
credit reports, but fewer than half were able to answer specific
questions about the dispute process and their rights. For example,
about one-third of consumers correctly responded that CRAs would
investigate disputed information for free. In addition, the majority of
consumers (94 percent) did not name FTC as the federal agency they
would contact with a complaint about a CRA--for example, about an
unresolved dispute--although FTC is the federal agency primarily
responsible for enforcing consumers' rights for CRA's credit reporting.
Overall, around 18 percent of consumers said that they had actually
disputed information on their credit reports at some point. Some of the
most frequent reasons these consumers mentioned for disputing were that
the information on the report was not theirs or that the report showed
either incorrect payment histories or incorrect late payments.
Consumers reported that incorrect information was removed from their
reports in about two-thirds of the disputes.
Consumers Knew That They Could Dispute Information but Did Not
Understand the Dispute Process or Their Rights:
Ninety percent of consumers correctly responded that they could dispute
information they believed was inaccurate and request that it be
corrected. They also appeared to understand the importance of checking
their reports for potential errors, though many did not do so. While
the majority of consumers (86 percent) said that they should check
their reports periodically, only 61 percent of this group reported
actually having seen their credit reports.[Footnote 52] Further,
consumers were unclear about some of their specific rights under the
dispute process. For instance, prior to the FACT Act, FCRA provided
consumers with certain rights when they disputed information through
CRAs but not when consumers disputed information with those that had
furnished it, such as credit card companies.[Footnote 53] Yet almost
two-thirds of consumers (64 percent) said that they would contact their
lender first rather than a CRA to dispute an incorrect late payment on
their credit report. Another 18 percent said that they would contact
the CRA first, while 13 percent said they did not know whom to contact.
The CFA's 2003 study also found that 64 percent of consumers believed
that they had to contact their lender if they found an inaccuracy in
their reports or scores.[Footnote 54] The fact that consumers are more
likely to have ongoing business relationships with their lenders and
not with the CRAs could help explain these responses.
In addition, many consumers were unaware of other details of the
dispute process. For example:
* 72 percent did not know that CRAs investigate incorrect information
for free;
* 60 percent did not know that resolving a dispute with one CRA did not
mean that the other CRAs would automatically correct the
information;[Footnote 55]
* 59 percent did not know that they had the right to add an explanatory
statement to their credit reports if they were unable to resolve a
dispute; and:
* 62 percent did not know which agency to contact if they were not
satisfied with a CRA's work--for instance, if a dispute was not
resolved--and few (6 percent) named FTC as the government agency they
would contact.
This lack of knowledge is consistent with what we heard from some in
the credit reporting industry--that consumers do not know about the
specifics of the dispute process unless they have undertaken it
themselves and that they learn the process once they contact a CRA. In
commenting on a draft of this report, FTC noted that consumers are
provided with the information they need to dispute information when
they obtain a credit report from a CRA, because the CRAs are required
by law to provide a "summary of rights" with each credit report
consumers request. This "summary of rights" explains consumers' rights
under FCRA, including information on how to conduct a dispute, and
provides contact information for the FTC and other enforcement
agencies. While this requirement provides consumers with information
when they request a report from a CRA directly, many consumers do not
request reports themselves. Nearly half of the 58 percent of consumers
in our survey who obtained their reports said that someone else had
done the ordering for them. As a result, these consumers may not have
received a copy of the summary of rights and may not have had the
information necessary to conduct a dispute or file a complaint about
the work of a CRA.
The Few Consumers That Tried to Dispute or Correct Information Did So
for a Variety of Reasons:
We found that approximately 18 percent of consumers said that they had
tried to dispute information on their credit reports at some point. To
supplement our survey data, we asked CDIA to provide data from each of
the three nationwide CRAs on the number of disputes. CDIA found that of
the approximately 57 million report requests in 2003, about 12.5
million involved disputes.[Footnote 56]
Our survey data showed that the 18 percent of consumers who disputed
did so for a variety of reasons. For example, about 17 percent found
items on their reports that belonged to someone else (fig. 6). About 14
percent of consumers disputed for other reasons, including an incorrect
payment history (14 percent), incorrect late payments (14 percent),
incorrect bill information (13 percent), or incorrect credit card
information (13 percent). We also found that about 10 percent of
consumers had disputed personal identifying information, such as a name
or address. During our survey, consumers gave the following examples of
their reasons for disputing information:
* "They had put the loan on twice";
* "They said I hadn't paid some bill that showed up before I was born";
* "It said I was deceased"; and:
* "Wrong address, repossession of a car in California where I have
never lived."
Figure 6: Reasons Consumers Disputed Information on Their Credit
Reports:
[See PDF for image]
Note: Other reasons included incorrect bankruptcy information (4
percent), incorrect information from a former spouse (3 percent),
identity theft (2 percent), and another reason (4 percent).
[End of figure]
In Congressional testimony, industry officials have noted that it is
important to address the issue of inaccuracies in credit reports. But
they added that many questions that begin as consumer disputes might
not be attributable to errors. For instance, they said that consumers
might have not understood what kinds of information credit reports
could contain and could withdraw a presumed dispute after learning more
about the reporting process.[Footnote 57] These officials also said
that a presumed dispute could be something very easy to correct, such
as a misspelled last name or wrong address.
Based on our survey, of the 18 percent of consumers that had disputed,
around two-thirds (69 percent) said that the disputed information had
been removed from their credit reports. Another 23 percent said that
the information had not been removed, and 7 percent said that they did
not know if it had been. In 2003, CDIA testified that it had collected
information on the results of consumer disputes and found that data had
been deleted in 27 percent of the disputed cases, verified and left on
the person's report in 46 percent of the cases, and modified following
the instructions of the entity furnishing it in 27 percent of the
cases. CDIA said, however, that these actions could have been updates
of information as well as disputes.[Footnote 58] Our survey results
also showed that of those consumers who had disputed (18 percent),
about one-third contacted their lender, one-third contacted the CRA
only, and one-third contacted both.
Of the consumers in our survey who had disputed and had information
removed from their credit reports (69 percent), we found that about 72
percent said that the information had not reappeared,[Footnote 59]
although 15 percent of this group were unsure if it had been
reinserted. The issue of information being reinserted has been raised
in testimony before Congress, especially in cases of identity theft. In
addition, FTC continues to receive consumer complaints through its Web
site and toll-free number regarding the reinsertion of deleted items in
credit files without notice to the consumer. We also found that about
one-third of the consumers who disputed information in credit reports
chose to add an explanatory statement to their reports, as allowed
under FCRA. The CRAs have noted that lenders may take such a statement
into consideration when making credit decisions.
Certain Demographics and Credit Experiences Are Associated with
Consumers' Understanding of Credit Reporting Issues:
We found that education, income, race/ethnicity, employment status, and
age, as well as having had an automobile or mortgage loan and
experience with the credit reporting process, all had a statistically
significant effect on consumers' knowledge of credit reporting. These
factors have a cumulative affect on consumers' knowledge. For example,
a consumer with a college degree, credit experience, and a high income
could score much higher than a consumer without these factors. We found
that less educated consumers, those of Hispanic origin, those with
lower incomes, and younger and older consumers scored lower on the
survey than others. At the same time, we found that consumers who had
had experience with credit or the credit reporting process, including
obtaining a credit report, scored higher. Finally, we found that other
experiences, such as having been a victim of identity theft, gender,
and residence in one of the seven "free report states" did not have a
statistically significant effect on consumers' knowledge.
Those with Less Education and Lower Incomes Scored Lower on Our Survey:
Our survey results showed that awareness of credit reporting issues
generally increased as consumers' educational level increased.[Footnote
60] As figure 7 shows, consumers with less than a high school education
had a lower mean or average score--measured as the percentage of
correct answers to questions designed to test knowledge--overall
compared with those with more education (36 percent compared with 63
percent for those with a bachelor's degree or more). In addition, as
educational level increased, consumers obtained reports and scores and
disputed more often than those with less education. For example,
consumers with less than a high school education had not viewed their
credit reports as frequently as their more educated counterparts,
especially those with a bachelor's degree or more (25 percent and 69
percent, respectively). In addition, we found that consumers with less
than a high school degree were less likely to know some of the "basics"
about credit reporting--for example, that skipping loan payments and
making late credit card payments had a negative effect on their credit
scores. Specifically, 55 percent of those with less than a high school
degree knew that skipping loan payments had a negative effect, compared
with 95 percent of those with a bachelor's degree or more.[Footnote 61]
In addition, although the Internet is one of many possible sources for
consumers to obtain a copy of their credit reports, those with less
than a high school education were less likely to name the Internet as a
source compared with those with a bachelor's degree or more (10 percent
and 34 percent, respectively). Further, government agencies, CRAs, and
others use the Internet as a tool to disseminate educational
information on credit reporting issues.
Figure 7: Average Total Scores and Demographic Groups:
[See PDF for image]
Note: Brackets on each bar represent the sampling error (or confidence
interval) for that estimate at the 95 percent level of confidence.
[End of figure]
Consumers with relatively low household incomes also had less knowledge
of credit reporting than consumers in higher-income households. For
example, those in the lowest household income group (less than $25,000
annually) correctly answered an average of 47 percent of survey
questions, while those in the highest household income group ($75,000
or more annually) correctly answered around 63 percent (fig. 7). In
addition, the percentage of consumers who had viewed their reports and
scores and disputed inaccurate information generally increased with
household income level. For example, consumers with household incomes
of $25,000 or less were less likely to have viewed their credit reports
than consumers in the highest household income group (38 percent
compared with 75 percent). We also found that consumers in households
making less than $25,000 were less likely than the highest income group
to know that their credit history could affect the interest rates
lenders offered them (73 percent compared with 89 percent) and that
frequently applying for credit affected credit scores negatively (44
percent versus 78 percent). In addition, those in the lowest income
group were less likely to name the Internet as a place to go for a
credit report than those with household incomes of $75,000 or more (17
percent and 36 percent, respectively).
Race/Ethnicity Was Associated with Differences in Survey Scores:
We also found that race/ethnicity had a statistically significant
effect on consumers' knowledge of credit reporting. For example,
African Americans and whites scored similarly, showing approximately
equal levels of knowledge, while Hispanics scored consistently lower on
our survey. As shown in figure 7, African Americans correctly answered
an average of 55 percent of survey questions, whites an average of 58
percent, and Hispanics an average of 43 percent.[Footnote 62] In
addition, we found that African Americans and whites were almost
equally as likely to have viewed their credit reports and obtained
their credit scores but that Hispanics were less likely to do either
(fig. 8). More African Americans than whites said that they had ordered
their credit reports themselves--65 percent and 50 percent,
respectively.[Footnote 63]
Figure 8: Credit Reporting Experience by Race/Ethnicity:
[See PDF for image]
[End of figure]
Table 1 highlights some specific survey responses for each
racial/ethnic group. We found that 84 percent of African Americans and
92 percent of whites knew that skipping loan payments could lower their
credit score. And around the same percentage of both groups (about 90
percent) knew that they could dispute information on their credit
report. Hispanics had lower scores on both of these questions--63
percent and 78 percent, respectively. However, when asked if race was
or was not on their credit reports, Hispanics scored virtually the same
as African Americans, though still lower than whites. Finally, we found
that more whites (35 percent) named the Internet as a source for
obtaining their credit reports than African Americans and Hispanics (18
and 16 percent, respectively).
Table 1: Selected Survey Responses by Race/Ethnicity:
Credit Report Section:
Survey Question: Knew they had the right to obtain their report at any
time;
Race/Ethnicity (in percentages): African American: 66%;
Race/Ethnicity (in percentages): Hispanic: 48%;
Race/Ethnicity (in percentages): White: 77%.
Survey Question: Knew race was not on their credit report;
Race/Ethnicity (in percentages): African American: 51%;
Race/Ethnicity (in percentages): Hispanic: 50%;
Race/Ethnicity (in percentages): White: 63%.
Credit Score Section:
Survey Question: Knew the correct definition of a credit score;
Race/Ethnicity (in percentages): African American: 59%;
Race/Ethnicity (in percentages): Hispanic: 39%;
Race/Ethnicity (in percentages): White: 78%.
Survey Question: Knew skipping loan payments could affect scores
negatively;
Race/Ethnicity (in percentages): African American: 84%;
Race/Ethnicity (in percentages): Hispanic: 63%;
Race/Ethnicity (in percentages): White: 92%.
Survey Question: Knew making late credit card payments could affect
scores negatively;
Race/Ethnicity (in percentages): African American: 86%;
Race/Ethnicity (in percentages): Hispanic: 69%;
Race/Ethnicity (in percentages): White: 91%.
Dispute Section:
Survey Question: Knew they had the right to dispute information on
their credit report;
Race/Ethnicity (in percentages): African American: 89%;
Race/Ethnicity (in percentages): Hispanic: 78%;
Race/Ethnicity (in percentages): White: 92%.
Source: GAO.
[End of table]
Our survey findings are consistent with information we collected during
interviews with federal agencies and other organizations, including the
National Council of La Raza (La Raza).[Footnote 64] Officials at the
FDIC, the Office of the Comptroller of the Currency, and FTC told us
that their financial literacy efforts are often target certain
populations, including Hispanics, that may more often be "unbanked" or
less financially literate.[Footnote 65] Government officials and
representatives of other organizations, including La Raza, discussed
the need for financial literacy programs among the low-income Latino
community.[Footnote 66] They also told us that some Latinos tend to
avoid debt, making them less likely to obtain credit and thus less
likely to have experience with credit reports and scores. In our
survey, we found that more white consumers had experience with credit-
related products than African Americans or Hispanics, with Hispanics
having the least amount of experience (table 2).
Table 2: Experience with Credit-Related Products and Race/Ethnicity:
Have had the following credit-related products at some point in time:
Credit card;
Race/Ethnicity: (in percentages): African American: 75%;
Race/Ethnicity: Hispanic: 62%;
Race/Ethnicity: White: 92%.
Have had the following credit-related products at some point in time:
Automobile loan;
Race/Ethnicity: (in percentages): African American: 62%;
Race/Ethnicity: Hispanic: 45%;
Race/Ethnicity: White: 80%.
Have had the following credit-related products at some point in time:
Home loan/mortgage;
Race/Ethnicity: (in percentages): African American: 42%;
Race/Ethnicity: Hispanic: 28%;
Race/Ethnicity: White: 73%.
Source: GAO.
[End of table]
Younger and Older Consumers Scored Lower on Our Survey:
We also found that consumers' knowledge of credit reporting differed
with age and that the youngest consumers (aged 18 to 24) and the oldest
consumers (aged 65 and older) scored lower compared with other age
groups. The average scores for the youngest and oldest groups were 50
and 49 percent, respectively, compared with an average score of almost
60 percent for all other age groups (fig.). We also found that
experience with the credit reporting process varied by age. For
example, 7 percent of the youngest consumers and 17 percent of the
oldest consumers had obtained their credit scores, compared with 33 to
48 percent of those in all other age groups.[Footnote 67] We also found
that only 18 percent of those aged 18 to 24 knew that their credit
history could affect employment decisions--an issue of particular
importance to persons in this age group, who are likely to be engaging
in their first job search--while 29 to 46 percent of those in all other
age groups knew this.[Footnote 68] Younger consumers were also less
aware that having a credit history for a short time could affect scores
negatively, especially when compared with those aged 25 to 34 (41
percent and 61 percent, respectively).[Footnote 69] Finally, those over
age 65 were less likely to say that they should check their credit
reports from time to time for possible errors than all other age groups
(71 percent versus 88 to 92 percent) and less likely to name the
Internet when compared to all other age groups (14 percent versus 29 to
37 percent).[Footnote 70]
Consumers with Credit Experience Scored Higher on Our Survey:
Our analysis showed that having had certain credit experiences
increased consumers' knowledge of credit reporting issues. This finding
is consistent with opinions from experts in the field of financial
literacy, who have noted that consumers who have experience with a
financial product or transaction--in other words, those who have been
able to "learn by doing"--are better informed about financial products
and transactions than those who lack such experience.
First, we found that consumers who had viewed their reports, obtained
their scores, or disputed information demonstrated more knowledge of
credit reporting. For example, we found that consumers who had viewed
their credit reports had a higher average survey score (62 percent)
than those who had not (47 percent) (fig. 9). Consumers who had seen
their credit scores were also better informed about credit issues than
those who had not (64 percent and 51 percent, respectively).
Figure 9: Average Total Scores and Credit Experiences:
[See PDF for image]
Note: Brackets on each bar represent the sampling error (or confidence
interval) for that estimate (95 percent level of confidence).
[End of figure]
We found that having had one of these three experiences could increase
responses to certain questions by 15 to almost 30 percent--for
instance, of those who had obtained their credit reports, 59 percent
knew how long information remained on a credit report, compared with 30
percent of those who had not obtained a report. In addition, 76 percent
of those who had obtained their credit scores but 52 percent of those
who had not knew that frequently applying for new credit could have a
negative effect on a scores.
Second, we found that consumers who had either obtained an automobile
loan or a home loan/mortgage at some point in time scored higher on the
overall survey than those who had not had these loans. Specifically,
consumers who had experience with these loans--about 74 percent of
consumers--scored an average of 58 percent on the survey, while those
who had not scored an average of 45 percent (fig. 9). In addition,
those who had obtained either an auto or mortgage loan were more likely
to view their credit reports and scores and to dispute information on
their reports--66 percent compared with 29 percent of those that had
not had these loans. Consumers who had had an auto loan or mortgage
also appeared to be more familiar with some of the factors that might
or might not affect credit scores, such as frequently applying for
credit and using most of the credit available to them, than consumers
who had not had such loans. For example, 65 percent of those who had
had these loans were aware that frequently applying for new credit
affected scores negatively, compared with 43 percent of those who had
not had these loans. In addition, consumers who had taken out these
loans were more likely to know how long information remained on credit
reports than consumers who had not (52 percent and 30 percent,
respectively).
Being a Victim of Identity Theft and Living in a Free Report State Had
Little Effect on Consumers' Survey Scores:
Government and industry officials we interviewed told us that certain
experiences--such as having been a victim of identity theft or having
requested a credit report because of an adverse action, such as denial
of credit--might increase consumers' knowledge of credit issues.
However, in our regression analysis we found that these experiences did
not have a statistically significant effect on consumers'
knowledge.[Footnote 71] Figure 9 shows that consumers who said they had
been victims of identity theft (approximately 10 percent of survey
respondents) and those who had obtained their reports because of
adverse action scored only slightly higher than other consumers.
Although we did not see large differences in overall knowledge of
credit reporting between victims of identity theft and other consumers,
we found that victims of identity theft were more likely to obtain
their credit reports (72 percent, compared with 56 percent) and
disputed information more frequently (36 percent and 16 percent,
respectively).[Footnote 72] In addition, like all other groups, victims
of identity theft did not name FTC as the agency to contact if they had
a problem with a CRA, although Congress designated FTC as the agency
responsible for identity theft claims. Identity theft complaints to FTC
have increased in recent years.[Footnote 73] We also did not see much
difference in overall knowledge between those who had obtained their
reports because of an adverse action and those who had not. However,
consumers who had experienced an adverse action and requested a report
were more likely to order their reports themselves (78 percent) than
other consumers (49 percent) and more likely to dispute information (52
percent compared with 15 percent).[Footnote 74]
Finally, we looked at consumers who lived in free report states--one of
the seven states where consumers could order a free copy of their
credit report annually before the FACT Act was passed (about 21 percent
of respondents). Based on our discussions with government and industry
officials, we decided to test whether living in a free report state
would increase consumers' knowledge about credit reporting issues
compared with those who did not live in these states. But as figure 7
shows, we found that consumers residing in free report states had only
slightly higher average survey scores than consumers elsewhere (59
percent compared with 55 percent). As reported by the Congressional
Research Service in 2003, TransUnion found in a 2003 study that around
4 percent of consumers in free report states had requested their
reports, compared with around 2 percent of consumers in other
states.[Footnote 75] Colorado was the exception, with more than 10
percent requesting their reports, possibly because some consumers were
notified of this right by mail.[Footnote 76] In our study, 46 percent
of consumers in free report states said that they did not know they had
this right.
We also assessed the effect of gender and employment on consumers'
knowledge. In our regression analysis we found that gender did not have
a statistically significant effect. Actively employed consumers scored
slightly higher on the survey than consumers in other employment
groups. More survey results for these and other factors we looked at
are reported in appendix IV.
Conclusions:
Given that credit reports and the credit scores derived from them
impact many important aspects of consumers' lives, it is increasingly
important that consumers understand what is in reports, how reports and
scores are used and the potential impacts of both. We found that most
consumers understood the basics of credit reporting but were less aware
of other important information, including:
* The impact of information contained in credit reports, how long
information remains on reports, some types of information reported, and
the possible impact of credit history on insurance coverage and
premiums and employment;
* How various behaviors impact credit scores--for example, that using
all available credit and frequently applying for new credit can lower a
score;
* The dispute process and federal protections under FCRA, as amended by
the FACT Act--for example, that CRAs investigate erroneous information
for free, that the FACT Act allows consumers a free credit report each
year, and FTC is the federal agency charged with enforcing consumer
protections for credit reporting by CRAs.
Although both industry and government agencies have numerous efforts
under way to increase financial literacy, our analysis highlights
specific areas for targeting educational efforts to increase consumer
knowledge of credit reporting. We believe our analysis could help FLEC
develop its national financial literacy strategy and Treasury develop
its multimedia campaign based on this strategy, as mandated in the FACT
Act. At our forum on financial literacy, credit experts agreed that
consumers would benefit from more education on credit, including the
cost of credit, how to use credit, and how to manage credit
responsibly. Without this understanding, it is difficult for consumers
to know whether they are receiving the most favorable decisions when
applying for credit or how to improve their credit history to achieve
higher credit scores.
Specifically, it would be beneficial for consumers to know what
information is included in a credit report and what information has the
greatest impact on credit decisions and therefore would be important to
review for accuracy and completeness. Our survey showed that 41 percent
of consumers did not review their reports, although periodic reviews
are an important step in ensuring the accuracy of credit histories and
possibly help mitigate against the effects of identity theft. Further,
we do not believe that consumers fully understand their rights under
the dispute process, since many did not know that CRAs investigate
incorrect information for free or that FTC is the federal agency
primarily responsible for enforcing consumers' credit reporting rights
by CRAs, such as in the case of an unresolved dispute. Given the
changes under the FACT Act, consumers would also benefit from knowing
that they are entitled to one free credit report each year.
Our analysis also provides some insight into possible subgroups of
consumers that would likely benefit from more targeted education. We
found that certain credit experiences--having had an auto or mortgage
loan--and some demographic factors--income, education, age, and
race/ethnicity--were associated with consumers' knowledge of credit
reporting issues. These results suggest that less educated, lower-
income, and Hispanic consumers may not have sufficient knowledge to
understand the negative repercussions of a poor credit history,
erroneous information on credit reports, or identity theft. FLEC and
Treasury will need to use creativity in their national strategy and
media campaign in order to reach these groups, going beyond
establishing an Internet link to federal resources, because these
consumers may not have Internet access. Further, since we found that
those with less than a high school education and younger consumers
scored lower on our survey, FLEC and others need to consider whether an
earlier introduction of credit reporting and other financial concepts,
such as in high school curriculums, may begin to address this apparent
knowledge gap.
Recommendations for Executive Action:
Consistent with FLEC's and Treasury's mandate under FCRA, as amended by
the FACT Act to improve consumers' understanding of credit reports,
credit scores, and the need to dispute inaccurate information, we
recommend that:
* The Secretary of the Treasury, in his capacity as Chairman of FLEC,
work with its members to improve consumers' understanding of their
federal rights through FLEC's national strategy and Treasury's
multimedia campaign, targeting those populations that scored the lowest
on our survey--for example, consumers with less than a high school
education and relatively low income levels, certain age groups (under
25 and 65 and older), and Hispanics.
* In targeting certain subpopulations, FLEC should expand efforts to
provide financial education on credit reporting issues, for example in
high school curriculums and in venues likely to reach Hispanics.
In addition, we recommend that the Chairman of FTC, through ongoing
educational initiatives, materials, and public announcements, and in
light of FTC's responsibility to protect consumers and enforce consumer
rights:
* Improve consumers understanding of how credit reports and scores are
used;
* Encourage consumers to obtain their credit reports and credit scores
and if necessary, dispute inaccurate information on their reports; and:
* Educate consumers about FTC's role in enforcing consumers' rights in
credit reporting.
Agency Comments and Our Evaluation:
We requested comments on a draft of this report from the heads of
Treasury and FTC. We received written comments from Treasury and FTC
that are summarized below and reprinted in appendixes VII and VIII,
respectively. Treasury and FTC also provided technical comments that we
have incorporated as appropriate.
While not specifically commenting on our recommendations, the Deputy
Assistant Secretary of the Office of Financial Education at Treasury
stated that the information the report presented would allow them to
better understand which parts of the credit reporting system are least
understood and channel their efforts to increasing knowledge of those
areas. Treasury outlined several ongoing efforts to improve financial
literacy and noted that FLEC is in the process of developing a national
strategy to promote basic financial literacy and education among all
Americans.
In commenting on the draft report, the Chairman of FTC stated that the
report's findings were useful to their ongoing outreach efforts and
described efforts they have underway to inform consumers of their
rights regarding credit reporting. FTC stated that it appreciated the
report's recommendation and, in response, stated that FTC has been
engaging in outreach efforts to reach certain groups, such as Hispanic
consumers and high school and college students. FTC also emphasized
that credit bureaus are to include FTC's "Summary of Rights" with every
credit report provided to consumers, as required in FCRA section
609(c). FTC stated in its comments that the summary of rights tells
consumers what they need to know, when they need to know it. Since the
summary of rights includes FTC's contact information, they believe this
is reaching the targeted audience--those consumers that requested and
reviewed their credit reports from CRAs. While we agree that this
information should assist those consumers who request their credit
reports, this information may not reach other consumers who have not
ordered or reviewed their credit report from CRAs. In this regard, when
we asked consumers where they would go to get a copy of their report,
only 34 percent said they would go to CRAs. Further, of the 58 percent
of consumers that stated they had seen their credit report, only about
half said that they had ordered their report themselves and therefore
should have received this summary of rights. As we recommended, FTC
should encourage consumers to obtain their credit report and credit
scores, and if necessary, dispute inaccurate information on their
reports. On a related point, FTC stated in their letter that they did
not believe our survey results shows that consumers who might wish to
complain about a credit bureau would be unaware of FTC's role and
described a variety of efforts for publicizing their enforcement
authority. Even with these efforts, our survey showed that only 6
percent of respondents indicated that the FTC was the agency they would
contact if they were to file a complaint against a CRA, such as in the
case of an unresolved dispute with the CRAs. This finding indicates to
us that few consumers are aware of the vital role FTC plays in
protecting consumers' rights in regards to credit reporting and the
CRAs. As we recommended, FTC should do more to educate consumers of its
role in enforcing these rights.
We are sending copies of this report to the Chairman of FTC and the
Secretary of the Treasury. We will make copies available to others upon
request. In addition, the report will be available at no charge on the
GAO Web site at [Hyperlink, http://www.gao.gov].
If you have any questions concerning this report, please contact me at
(202) 512-8678 or [Hyperlink, jonesy@gao.gov] or Debra R. Johnson at
(202) 512-9603 or [Hyperlink, johnsond@gao.gov]. Other contacts and
acknowledgements are listed in appendix IX.
Signed by:
Yvonne D. Jones:
Director, Financial Markets and Community Investment:
[End of section]
Appendixes:
Appendix I: Objectives, Scope, and Methodology:
The objectives of this report were to examine the extent to which
consumers (1) understand and review their credit reports; (2)
understand and review their credit scores; and (3) know how to dispute
information on their credit reports and actually do so. In addition,
this report discusses some of the factors that are associated with
consumers' understanding of these issues.
Scope and Methodology:
To determine the extent of consumers' knowledge of credit reporting
issues, we conducted a telephone survey of 1,578 randomly-sampled
noninstitutionalized U.S. adults 18 and over. This survey was designed
to gauge respondents' knowledge of credit reports, credit scores, and
the dispute resolution process, and to estimate the extent to which
U.S. consumers have obtained their credit reports and scores or have
experienced other credit-related events. The survey data collection was
conducted by a private research firm under contract to GAO from late
July to early October 2004, shortly before the provision of the Fair
and Accurate Credit Transactions Act (FACT Act) that gives all U.S.
consumers the right to receive one free credit report a year was
implemented.
Before and during the design of our survey instrument, we also
collected data and testimonial evidence on credit reporting literacy
and some general financial literacy issues from relevant federal
agencies, industry representatives, non-profit organizations and
consumer agencies. To develop survey questions and to provide
background and context for our survey findings, we collected and
analyzed data from relevant surveys and other studies conducted by
several of these agencies and organizations, including the Consumer
Data Industry Association (CDIA), the Federal Reserve Board (FRB), the
American Association of Retired Persons (AARP), the Consumer Federation
of America (CFA), Federal Home Loan Mortgage Corporation (Freddie Mac),
Fair Isaac Corporation, members of the Financial Services Roundtable,
Jump$tart Coalition, the Congressional Research Service (CRS), and the
three nationwide CRAs (Equifax, Experian, TransUnion). Please see
appendix III for listing of findings from some of the surveys and
studies that were reviewed.
Design of the Telephone Sample:
Our survey used random digit dialing methods to generate a probability
sample of non-insitutionalized U.S. adults aged 18 and older. Analytica
Research, in affiliation with Dove Consulting Group, Inc, obtained and
dialed 9,784 randomly generated phone numbers from Survey Sampling
International. These phone numbers, which include both listed and
nonlisted numbers, were generated from "banks" of numbers (the
telephone area code, exchange, and next two digits of the phone number)
assigned to different regions of the country.
To enable GAO to make relatively precise estimates of knowledge and
behavior within selected subgroups, we oversampled phone numbers likely
to include a relatively high proportion of these subgroup members. The
sample was designed to include a relatively higher proportion of
responses from (1) the seven states where consumers could receive at
least one free credit report per year prior to implementation of the
FACT Act, (2) African Americans, (3) Hispanics, and (4) nonminorities
with annual family incomes under $25,000.[Footnote 77] A minimum number
of interviews was specified for each of these subgroups. The quotas
were achieved in ways that distributed the oversampled interviews
naturally across the population. For example, the seven-state
oversample was drawn in proportion to the population sizes of each
state, and the oversamples for the minority subgroups were spread
across geographic areas that had predicted incidences of minorities of
as low as 48 percent, so as not to sample in only the most concentrated
minority areas. A minimum sample was also set for nonminorities with
annual family incomes under $25,000, but no additional oversample was
necessary to achieve that target.
The 1578 respondents included 300 Black/African American (non-Hispanic)
respondents, 300 respondents of Spanish, Hispanic, or Latino origin
(Hispanics), 302 non-Hispanic, non-Black/African American respondents
with current total annual family incomes of less than $25,000, and 676
nonminority respondents with incomes of $25,000 or greater. In
addition, 324 of the above respondents resided in the seven free credit
report states--Colorado, Georgia, Maine, Maryland, Massachusetts, New
Jersey, and Vermont.
Initially, the selection of respondents within each household reached
was also randomized. After determining the number of adults in the
household who were 18 years of age or older, the interviewer selected
the person with the most recent birthday to respond to the survey. This
random selection of household members was suspended, however, in some
calls to oversampled subgroups when the initially selected respondent
was not available for the interview and the person answering the phone
met the criteria for the quota. GAO decided that the need to obtain
sufficient interviews with subgroup members outweighed the advantages
of complete randomness in selecting consumers to interview.
Developing the Questionnaire:
To develop areas of inquiry for the survey, we reviewed previous GAO
work related to credit reports, agency materials, and previous surveys
that asked questions about financial knowledge and credit reporting
issues. We considered questionnaires from multiple organizations such
as, the AARP, the Federal Reserve Board (FRB), the Jump$tart Coalition,
and CFA and obtained suggestions on questions from these organizations
based on their experiences. We used these sources and our own analysis
to develop an initial set of questions that (1) measured knowledge at
varying degrees of difficulty, (2) had one correct answer among several
plausible ones, and (3) covered different aspects of credit reporting
knowledge (e.g. where credit reports and scores come from, what they
are used for, and what factors influence them). In addition, we
developed questions about consumers' experience with credit reports,
credit scores, and the dispute process and their perceptions of these
documents and processes. For example, we asked consumers if they had
obtained a credit report or score, if they had ever disputed
information on a credit report, whom they would contact first in the
case of a dispute, and whether they believed that their credit report
was understandable.
In addition to internal review by a GAO survey methodologist, we
obtained expert reviews of our draft questionnaire from the FTC, the
FRB, the Office of the Comptroller of the Currency (OCC), the Federal
Deposit Insurance Corporation (FDIC), Treasury, AARP, CFA, the Fair
Isaac Corporation, the three national credit reporting agencies (CRAs),
CDIA, and a survey research contractor. The survey also included a
number of questions on the respondent's familiarity with Social
Security benefit statements, collected for use in another GAO
engagement.
The questionnaire was translated into Spanish. To develop the Spanish
version, we obtained information on credit reporting issues that had
already been translated from various entities. For example, we obtained
the Spanish versions of FDIC's MoneySmart curriculum and Freddie Mac's
CreditSmart curriculum. After our general review, we contracted with
the U.S. Department of State to perform the actual translation. After
the State Department completed its work, GAO staff reviewed the survey
to identify any words that were overly formal or technical, and we made
some adjustments to the survey.
We further developed and refined the questions by conducting pretests
of successive drafts of the questionnaire with members of various
demographic groups. Overall, we conducted 17 pretests, including 5 in
Spanish. After the questionnaire was programmed for use as a computer-
assisted telephone interview script, AOS Inc., one of the two call
centers conducting the interviews, conducted a pilot test of over 20
calls to a random sample of U.S. households in late July 2004, and we
made final revisions to the questionnaire as a result. See Appendix II
for a reproduction of the final questionnaire administered.
Administering the Survey:
Interviewers were trained in the use of the questionnaire and
background material to answer respondents' questions about the survey.
In addition, bilingual interviewers were available during fieldwork if
an interview had to be conducted in Spanish. Data collection on the
survey began on July 22, 2004 and ended October 7, 2004.
The survey contractor implemented the sample by first conducting a
limited number of interviews in a nationwide sample to obtain some
minorities, low-income residents, and residents of the seven states
with free credit reports, and then began the oversamples of high-
minority and seven-state areas to fully meet the subgroup requirements.
The quota of 300 low income, nonminority respondents was met in calling
in the nationwide and other oversample groups, without an additional
oversample being drawn for this group.
Calls were made throughout the week at various times to maximize the
chance of finding someone at home. A minimum of 10 attempts, at
different times and days of the week, were made to call back working
numbers that were not answered.
In addition to repeated calling, other techniques were used to increase
response rates. Messages were left on answering machines for potential
respondents offering a toll free number that the respondent could call
during business hours. After hours, a recorded announcement requested
that the caller leave the date and time for a callback. GAO staff were
also available by telephone to answer respondents' questions and
address concerns about the survey. Nonrespondents were offered cash
incentives of $10 when recontacted to solicit their participation.
To increase participation in the survey, in mid-August a reverse
directory was used to identify mailing addresses for 1,417
nonresponding numbers. Postcards were subsequently sent to these
households with an appeal to participate when interviewers next called.
An additional 537 postcards were mailed in early September. A second
telephone survey call center began making interviews in mid-August to
complete the fieldwork at a faster rate.
Survey Response:
We received a total of 1,578 usable responses, for an overall response
rate of 48 percent. We calculated the response rate as the total number
of usable (complete and partially complete responses) surveys divided
by the total eligible sample called.[Footnote 78] In determining the
total eligible sample, we excluded numbers discovered to be nonworking,
disconnected, businesses, or otherwise not households with members
eligible to be interviewed in one of the sample categories with quotas
still to be met. Calls to people in groups for which quotas had already
been met were also deemed ineligible. We also assumed that a proportion
of those working numbers attempted at least 10 times without an answer
were also not eligible. The proportion of eligibility among those not
contacted was estimated using practices commonly followed in the survey
research industry. That is, the same proportion of eligible respondents
found among those phone numbers actually contacted and determined to be
eligible or ineligible was applied to numbers not contacted and of
unknown eligibility.
The cooperation rate for the survey was 59 percent. The cooperation
rate measures the success at completing interviews with contacted
people who are determined to be eligible. It is defined as the number
of complete interviews divided by the total number of complete and
partial interviews, refusals, and callbacks that were arranged but not
completed.[Footnote 79] The difference between the response rate of 48
percent and the cooperation rate of 59 percent suggests that the
difficulty in contacting anyone at the phone numbers generated was a
significant factor in the low response rate, because interviewers were
able to convince nearly 60 percent of those they contacted to complete
the survey.
In addition to nonresponse to the entire survey, respondents could
choose not to answer individual questions. This item-level nonresponse
was uniformly low in the survey, usually 1 percent or less. Eleven
percent of respondents declined to answer the income question, which
had the highest item nonresponse rate.
Table 3 indicates that call outcomes varied somewhat by sample
subgroup. Response rates were somewhat lower among the samples of high
minority areas, particularly those targeted to meet the Hispanic
subgroup quotas. The same pattern was observed for cooperation rates.
Table 3: Sample Dispositions and Response Rates:
A. Total calls attempted: Entire Sample.
A. Total calls attempted;
Entire Sample: 9784;
Sample Subgroups: Nationwide: 5423;
Sample Subgroups: Seven States with Free Reports: 877;
Sample Subgroups: African American areas: 1205;
Sample Subgroups: Hispanic areas: 2279.
B. Completed interviews;
Entire Sample: 1578;
Sample Subgroups: Nationwide: 936;
Sample Subgroups: Seven States with Free Reports: 184;
Sample Subgroups: African American areas: 191;
Sample Subgroups: Hispanic areas: 267.
C. Refused;
Entire Sample: 893;
Sample Subgroups: Nationwide: 516;
Sample Subgroups: Seven States with Free Reports: 6;
Sample Subgroups: African American areas: 117;
Sample Subgroups: Hispanic areas: 254.
D. No interview, known eligible[A];
Entire Sample: 223;
Sample Subgroups: Nationwide: 80;
Sample Subgroups: Seven States with Free Reports: 0;
Sample Subgroups: African American areas: 32;
Sample Subgroups: Hispanic areas: 111.
E. No interview, known ineligible[B];
Entire Sample: 5319;
Sample Subgroups: Nationwide: 2867;
Sample Subgroups: Seven States with Free Reports: 590;
Sample Subgroups: African American areas: 620;
Sample Subgroups: Hispanic areas: 1242.
F. No interview, unknown eligibility[C];
Entire Sample: 1771;
Sample Subgroups: Nationwide: 1024;
Sample Subgroups: Seven States with Free Reports: 97;
Sample Subgroups: African American areas: 245;
Sample Subgroups: Hispanic areas: 405.
Response rate[D];
Entire Sample: 48%;
Sample Subgroups: Nationwide: 50%;
Sample Subgroups: Seven States with Free Reports: 83%;
Sample Subgroups: African American areas: 45%;
Sample Subgroups: Hispanic areas: 35%.
Cooperation rate[E];
Entire Sample: 59%;
Sample Subgroups: Nationwide: 61%;
Sample Subgroups: Seven States with Free Reports: 97%;
Sample Subgroups: African American areas: 56%;
Sample Subgroups: Hispanic areas: 44%.
Source: GAO.
[A] Includes cases with partial responses, callbacks not completed, and
those in which language barriers prevented interviews.
[B] Includes cases with disconnected and nonworking phone numbers, fax
numbers, business numbers, duplicates, and those drawn in states with
quotas already met, as well as 50 percent of the interviews not
completed because family income was outside of the range of eligibility
after a quota had been met.
[C] Includes cases in which phones were not answered, only answering
machines were encountered, the line was busy, the caller was
immediately told that the number was on the "do not call list," or
"privacy manager" call blocking was in effect, as well as 50 percent of
the interviews not completed because family income was outside of the
range of eligibility after a quota had been met.
[D] The response rate is defined as B/(B+ C + D + (e * F)), where e is
the eligibility rate found among cases for which eligibility could be
determined. That rate, 33.6 percent, is assumed to apply to the no-
interview cases for which eligibility could not be determined. That is,
we assume that the ratio of eligible-to-ineligible cases among those we
are unsure about is the same among those for which a determination of
eligibility could be made. The effect of this assumption is to exclude
approximately one-third of cases of unknown eligibility that could not
be contacted from the base of eligible and possibly eligible phone
numbers.
[E] The cooperation rate is defined as B/(B+ C + D), without the
inclusion of no-interviews as a result of language barriers in D, the
no-interviews known to be eligible as defined by AAPOR. Language
barriers are not displayed separately from row D above.
[End of table]
Survey Error and Data Quality:
The practical difficulties of conducting any sample survey may
introduce errors into estimates made from them. These errors include
sampling, coverage, measurement, nonresponse, and processing errors. We
made efforts to minimize each of these.
Because we followed a probability procedure based on random selections,
our sample is only one of a large number of samples that we might have
drawn, and thus our results are subject to sampling error. Since each
sample could have provided different estimates, we express our
confidence in the precision of our particular sample's results as a 95
percent confidence interval around each estimate--for example, plus or
minus 10 percentage points. That is, we are 95 percent confident that
each of the confidence intervals for estimates in this report will
include the true value that we would have obtained had the entire
population been surveyed. All percentage estimates from this survey
have confidence intervals of plus or minus 6 percentage points or less,
unless otherwise noted.
Coverage errors in survey estimates can occur from the failure of the
sample to adequately cover the actual study population. While our
sample included households with unlisted as well as listed numbers, we
were able to reach only people in households with telephones, estimated
at approximately 95 percent of U.S. households. The randomly generated
phone numbers we called do not include cellular phone exchanges,
resulting in undercoverage of those households reachable only through
cellular phones. In addition, we sampled a higher proportion of some
subgroups. To adjust the interviewed sample so that it would be
representative of the country as a whole, weights were developed to
statistically adjust the contribution each interview made towards the
total. Interviews in the 7 states were weighted down so that the
proportion of interviews conducted in those states would match the
ratio of population in those 7 states to the population of the nation,
as determined by the U.S. Bureau of the Census. Similar adjustments
were made to the responses resulting from the other oversamples:
African Americans, Hispanics, and low-income non-minorities. Additional
weight adjustments were made for age and gender to conform to national
distributions (app. II and V show the proportion of socio-economic and
experience variables of respondents to our survey).
We attempted to minimize measurement errors arising from respondents
who accidentally or purposely misreported, or from interviewer mistakes
in administering the questions. In addition to the quality assurance
steps described above in developing the questions so they would
accurately measure the concepts of interest, the survey contractor
silently monitored some interviews on a daily basis, and provided
feedback and coaching to interviewers to improve data collection. In
questions that are meant specifically to test the knowledge of the
respondent, guessing can result in correct answers by chance and might
reflect an artificially high level of knowledge. To counteract this
tendency, we explicitly allowed respondents to specify that they did
not know answers in our interview and asked multiple knowledge
questions to lessen the effects of guessing on our overall analysis of
knowledge. We also asked some open-ended questions without categories
to choose from.
In addition, behaviors such as ordering a credit report or seeing a
credit score may be overreported because respondents wish to appear
diligent to interviewers or have mistaken one type of information for
another. However, others may tend to underreport these behaviors
because of a failure to recall distant events, or similarly because
they mistake one type of information for another. We have no additional
information on the extent or direction of this kind of error. However,
we compared our results with those of other surveys and found them to
be generally similar.
Nonresponse error can arise in the form of bias from not obtaining
interviews from nonrespondents who would have differed in their answers
from those who did respond. Our response rate of 48 percent, while
comparable to or higher than many other telephone surveys of this type,
means that we do not know how 52 percent of the population would have
responded. We do not know the impact this level of nonresponse has on
our results. While this level of nonresponse may affect survey results,
GAO believes the overall response rate is sufficient for the
conclusions drawn from the results, and can be generalized to the
population of U.S. adults age 18 and older.
A final source of error is data processing error, or mistakes made in
capturing or analyzing the data. In addition to the safeguards and
checks built into the Computer Assisted Telephone Interviewing (CATI)
software, after the survey, GAO performed checks on the survey data and
verified the analysis programming. The reliability of the coding of
open-ended answers into categories was assessed by having multiple
coders review the same open-ended responses and comparing their answers
for consistency.
Survey Data Analysis:
Our survey included a total of 58 questions. The survey questions were
created to test consumers' knowledge of credit reporting issues, obtain
their opinions about these issues, and examine their experiences with
the credit reporting process. Overall, there were 23 knowledge
questions (with 56 associated points) on the survey divided into 3
separate sections: credit report knowledge, credit score knowledge, and
dispute knowledge. We also incorporated 22 questions designed to obtain
consumers' opinions about these issues and to examine their experiences
with the credit reporting process, as well as 13 demographic questions.
In addition, there was one knowledge question that asked consumers
about the FACT Act.
Most of the questions had pre-coded answers, but some were open-ended
requiring unassisted responses. For example, there were 3 open-ended
knowledge questions (4, 7, and 43) that asked consumers to provide
their own explanation of what credit reporting agencies do, what credit
reports are, and what credit scores are. Some of the multiple-choice
questions had one part, while others had several parts. If a respondent
gave a correct answer to a multiple-choice question or sub-part of a
question, we awarded them one point for that question or subpart,
respectively. No points were given for an incorrect answer, for
refusing to answer, or for an answer of "don't know." For the open-
ended questions, we first developed and tested the following scoring
criteria: a respondent received no points if their response was clearly
incorrect or they said they did not know, one point if their response
was only partially correct or if their response was correct but vague,
and two points if their response was more detailed and fully correct.
Then, two team members separately reviewed and scored each open-ended
response. After all the responses were scored, a third member of the
team separately reviewed the assigned scores and adjudicated any
differences in the scores between the other two team members.
Based on the 23 knowledge questions, a respondent could earn a total of
56 points; 37 points based on 14 questions on the credit report
section, 12 points based on 4 questions on the credit score section,
and 4 points based on 4 questions on the dispute section. In addition,
consumers could earn 3 additional points by answering one question
about the FACT Act.[Footnote 80]
We generated an average or mean score for the survey as a whole. We
then analyzed responses for all of the survey questions and scores
based on groups of questions for the national sample and cross-
tabulated them across different demographic groups and across consumers
with different credit-related experiences, see appendix IV for a
discussion of these results. In addition, an electronic appendix (app.
VI) including additional subgroup results can be found on GAO's Web
site at [Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-05-411SP. We
also reviewed responses to survey questions about consumers'
experiences and opinions. Differences across demographic groups and
across consumers with different credit-related experiences were tested
for statistical significance at the 95-percent confidence level. In
addition to cross-tabulations, we used a regression analysis of
demographic and other factors that we thought would be associated with
consumers' knowledge of credit reporting issues. For a full discussion
of the regression analysis, see appendix V.
Interviews and Secondary Analysis:
To develop our survey and obtain additional data on consumers'
knowledge and awareness of credit reporting issues, we reviewed
existing sample surveys and other studies conducted by other
organizations and agencies on financial literacy, credit reports and
scores, and the dispute resolution process. We also obtained relevant
data from the 3 national CRAs. Surveys and other studies examining
financial literacy and credit reporting issues have been conducted by
several agencies and organizations, including AARP, FRB, CFA, FTC, U.S.
Public Interest Research Group (also known as the National Association
of State PIRGs), Fannie Mae Foundation, Freddie Mac, members of the
Financial Services Roundtable, Credit Research Center of Georgetown
University (CRC), Jump$tart Coalition, CDIA, the 3 national CRAs, and
Fair Isaac Corporation. The focus of these studies varies--some examine
credit-related issues, such as identity theft; others were conducted
for marketing purposes; and others deal with financial literacy in
general. For a listing of selected findings from some of these studies,
please see appendix III.
While we made an attempt to assess the methodology of these studies, we
were not able to collect enough information from study sponsors to
fully determine the quality of their results. Of the surveys we
identified, we included in Appendix III only those which appeared to
use a probability-based sample of a defined population. The surveys
that met these criteria were typically random-digit dialing (RDD)
telephone surveys of U.S. households. Because each survey or other
study we chose to report in Appendix III had limitations and risk of
error in estimates, we report only selected results for comparative and
background purposes. We did not publish detailed results from these
surveys because of the possibility of errors that we could not assess.
Results from these surveys are best considered in the context of other
evidence that might corroborate or disconfirm findings from individual
studies.
To obtain information on financial literacy efforts in general and on
credit reporting in particular, and to obtain organizational experience
with the credit reporting process and the credit reporting industry, we
spoke with most of the members of the Financial Literacy and Education
Commission (FLEC) and other organizations, including FTC, FRB, Chicago
Federal Reserve, OCC, FDIC, Office of Thrift Supervision (OTS),
Securities and Exchange Commission (SEC), Commodity Futures Trading
Commission (CFTC), Social Security Administration (SSA), National
Credit Union Association (NCUA), Federal National Mortgage Association
(Fannie Mae), Freddie Mac, and the Departments of the Treasury, Labor,
Education, Defense, and Agriculture. We also spoke with many industry
representatives, including those from academia, nonprofit organizations
and consumer agencies such as: The Partnership to Protect Consumer
Credit, National Council of La Raza, Financial Services Roundtable,
Fair Isaac Corporation, CDIA, AARP, CRC, CFA, Case Western Reserve
University, American Savings Education Council (ASEC), and the Employee
Benefits Research Institute (EBRI). We conducted our review in
Washington, D.C. from March 2004 through February 2005 in accordance
with generally accepted government auditing standards.
[End of section]
Appendix II: Survey Questionnaire:
[See PDF for image]
[End of figure]
[End of section]
Appendix III: Review of Selected Findings from Other Studies:
GAO conducted a literature search to identify some of the relevant non-
GAO sample surveys and other studies on consumers' knowledge of credit
reporting issues, the extent to which consumers review their credit
reports and scores and their reasons for doing so, and other related
matters. We attempted to evaluate the quality of the data and findings
of these reports, but the limited methodological documentation
available to us prevented a thorough review. We included only those
surveys for which we had some indication that a random sample of a
known population, typically a nationwide random-digit dial telephone
survey of U.S. adults, had been conducted. However, due to possible
shortcomings in important design aspects of these studies that are not
known, the possibility for error still exists in each of these studies.
Therefore, GAO does not consider any one estimate from the following
studies to be sufficiently accurate to base conclusions on. However, in
the aggregate, these other results may serve as a general comparison to
the GAO survey of U.S. consumers and provide a background description
of what other researchers have reported on these issues.
Table 4: Review of Selected Findings from Other Studies:
Credit Reports: The number of consumers who review their credit
reports:
Results: There were 57.4 million credit file disclosures in 2003. CDIA
estimates that there are a total of approximately 200 million credit
files maintained by each of the three nationwide CRAs;
Study/Source of Data: Consumer Data Industry Association (CDIA). Data
was collected from the three nationwide CRAs--Equifax, Experian, and
TransUnion--and provided to GAO in January 2005. The data on credit
files issued includes all direct-to-consumer disclosures--meaning
offers to consumers (for a fee) and online products, which commonly
include file disclosures, credit scores, credit score and analyses,
file monitoring, fraud alert systems and more. This data also includes
all consumer relations disclosures--meaning disclosures made to
consumers who contact a CRA to receive a file disclosure because of an
adverse action notice, concern about fraud, being unemployed and
seeking employment, or being on public assistance. These data are an
estimate of total requests by all consumers and could include multiple
requests from the same consumer.
Results:
* TransUnion estimates that in the 44 states that do not have a law
requiring CRAs to provide an annual free credit report upon request,
the consumer request rate ranges between 1.5% and 2%;
* In GA, MA, NJ, MD, Vermont (where credit reports are made available
for free) the request rate ranges between 3.5% and 4%;
* In Colorado (where consumers review one free report per year and, in
addition, are notified via mail of this right under two specific
circumstances), the request rate is between 10% and 11%;
Study/Source of Data: Congressional Research Service (CRS). "A
Consumer's Access to a Free Credit Report: A Legal and Economic
Analysis." Washington, D.C.: December 2003.
Results: An estimated 16 million are issued to consumers each year, 95%
of them for free;
Study/Source of Data: CDIA. Testimonies from CEO and President Stuart
Pratt: (1) before the U.S. Congress, Senate Committee on Banking,
Housing, and Urban Affairs, The Fair Credit Reporting Act and Issues
Presented by Reauthorization of the Expiring Preemption Provisions,
Hearings, 108[TH] Congress, 1st sass., July 10, 2003;
and (2) Fair and Accurate Credit Transactions Act of 2003, hearing H.R.
2622, 108th Congress, 1st sess., July 9, 2003.
Results: 34% of respondents 45 years and older said they had reviewed a
copy of their credit report;
Study/Source of Data: American Association of Retired People (AARP).
"Consumer Experience Survey: Insights on Consumer Credit Behavior,
Fraud and Financial Planning." Washington, D.C., October 2003.
Results: 43% of respondents said that they had reviewed their credit
report in the last 2 years;
Study/Source of Data: Consumer Federation of America (CFA). "Consumers
Lack Essential Knowledge and Strongly Support New Protections on Credit
Reporting and Scores," Washington, D.C. July 2003. Information
available at http://www.consumerfed.org.
Results: 58% of respondents said that they had reviewed their credit
report;
Study/Source of Data: Hilgert, Marianne A., Jeanne M. Hogarth, and
Sondra G. Beverly. "Household Financial Management: The Connection
between Knowledge and Behavior." Federal Reserve Bulletin, July 2003.
Results: 63% of respondents said that they had reviewed their credit
report;
Study/Source of Data: Fair Isaac Corporation. "myFICO Consumer Survey
Results," March 2003. Unpublished.
Results: Almost 20% of respondents had never seen their credit report;
Study/Source of Data: Household International. "Household Your Credit
Counts Challenge," December 2002. Unpublished.
Credit Reports: Why consumers obtained their credit reports:
Results: Of the 16 million reports requested by consumers:
* 84% were disclosed due to adverse action;
* 11.5% for fraud;[A];
* 5.25% for curiosity;
* 0.4% due to being unemployed or seeking employment; and;
* 0.1% due to being on public assistance;
Study/Source of Data: CDIA. Prepared testimony from CEO and President
Stuart Pratt, The Fair Credit Reporting Act and Issues Presented by
Reauthorization of the Expiring Preemption Provisions.
Results: Reasons why respondents looked at their credit report:
* 39% curiosity;
* 31% ready to apply for loan/credit;
* 10% denied credit, insurance, employment;
* 6% concerned you were a victim of identity theft;
* 12% some other reason; and;
* 1% did not know;
Study/Source of Data: Louis Harris and Associates. "Consumer Interest
in Free Credit Reports," 2003. Unpublished. Selected findings from this
study were provided to GAO by CDIA.
Credit Reports: How much consumers know about the data collection
process:
Results:
* 94% of respondents 45 years and older knew that a missed credit card
payment could be reported on their credit report;
* 88% of respondents 45 years and older were aware that a lender could
report a non-payment of a mortgage payment on the homeowner's credit
report;
Study/Source of Data: AARP. "Consumer Experience Survey."
Results: 81% of respondents answered "true" to the following statement:
Your credit report includes employment data, your payment history, any
inquiries made by creditors, and any public record information;
Study/Source of Data: Hilgert, Hogarth, and Beverly. "Household
Financial Management."
Credit Reports: What consumers know about obtaining their credit
reports:
Results:
* 97% answered "true"--consumers have the right to see their credit
report;
* 46% answered "true"--consumers in almost every state must pay a fee
to obtain their credit report;
* 81% answered "true"--consumers who fail to qualify for a loan have
the right to a free credit report;
* 3% were able to name all three CRAs, 7% named two correctly, 15%
named one correctly, and 75% named none of the CRAs correctly;
Study/Source of Data: CFA. "Consumers Lack Essential Knowledge."
Credit Reports: What consumers know about the possible impacts of their
credit histories:
Results: 88% of respondents 45 years and older answered "true" to the
following statement: "An individual's bill paying history on their
credit record affects their ability to get a home loan.";
Study/Source of Data: AARP. "Consumer Experience Survey."
Results: Who determines your interest rate (unaided)?
* 52%--my lender;
* 22%--I do;
* 20%--federal government (6%--Alan Greenspan); and;
* 5%--my credit rating/credit history;
Study/Source of Data: Fair Isaac Corporation. "myFICO Consumer Survey
Results."
Credit Reports: Other:
Results: 48% of respondents said their knowledge of credit reports was
excellent/good and 50% said it was fair/poor;
Study/Source of Data: CFA. "Consumers Lack Essential Knowledge."
Credit Scores: What consumers know about the factors that can impact
credit scores and their use:
Results: 9.8 million credit scores were provided to consumers by the
CRAs during calendar year 2003;
Study/Source of Data: CDIA collected data from the three nationwide
CRAs--Equifax, Experian, and TransUnion. This data does not include the
entire score disclosure market place, which, in addition to scores sold
by the three CRAs, includes scores sold by Fair Isaacs Corporation and
by lenders, such as banks.
Results:
* 34% of respondents answered "true"--a credit score mainly indicates
the risk of repaying a loan;
* These percentages of respondents answered "true" to the following
statements:
* 76%--credit scores change whenever a credit report changes;
* 52%--a married couple has a combined credit score;
* 43%--those with a credit score have only one score; and;
* 16%--only those at least 21 years of age have a credit score;
Study/Source of Data: CFA and Providian.[®] "Most Consumers Do Not
Understand Credit Scores According to a New Comprehensive Survey,"
Washington, D.C., July 2004. Information available at
http://www.consumerfed.org.
Results: If your credit score is below a certain level, you will
probably either be denied credit or have to pay a higher, subprime
rate. Would you say this level is in the …;
* 33% of respondents said the low 500s;
* 13%--high 500s;
* 12%--low 600s;
* 5%--high 600s;
* 7%--low 700s; and;
* 30%--did not know/none of the above;
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
Results: If your credit score is above a certain level, creditors will
usually charge you the lowest rates. Would you say this level is in
the…;
* 11% of respondents said the high 500s;
* 8%--low 600s;
* 10%--high 600s;
* 13%--low 700s;
* 29%--high 700s;
and;
* 29%--did not know/none of these;
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
Results: Percentage of respondents who indicated that the following
factors can influence one's credit score:
* 87%--whether loan payments have been paid on time;
* 79%--amount of debt related to income;
* 66%--whether one has maxed out the credit line on a credit card;
* 65%--income;
* 38%--age;
* 37%--marital status;
* 25%--education; and;
* 3%--did not know/none of these;
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
Results: Thinking about when a young adult gets and begins using their
first major credit card, how much lower or higher would their credit
score go if they did each of the following? Would you say much higher
(MH), somewhat higher (SH), little or no change (NC), somewhat lower
(SL), or much lower (ML) [or don't know (DK)]? The following data are
percentages of respondents;
[See PDF for image]
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
[End of subtable]
Results: Which of the following service providers often use credit
scores to decide whether you can buy a service or at what price? The
following percentages said "yes";
* 81%--mortgage lender;
* 77%--credit card lender;
* 50%--cell phone company;
* 48%--landlord;
* 47%--home insurer;
* 30%--electric utility; and;
* 8%--did not know/none of these;
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
Results: 90.1% of respondents 25 to 49 years old (and of certain
education and income levels) selected the response: "A credit score is
a rating of my credit history. It is calculated by credit bureaus and
most often reviewed by lenders in considering whether to grant
credit.";
Study/Source of Data: Providian. "Credit Cards and Mainstream America,"
February 2004. Unpublished.
Results: The following percentages of respondents said that second to
filing for bankruptcy, the following element had the greatest influence
on their credit score:
* 5.4%--the number of bankcard charge accounts that you have;
* 20.8%--the amount of available credit you have compared with the
outstanding balances on your credit cards;
* 5.9%--paying credit card bills online;
* 66%--late or missed payments reported on your credit accounts; and;
* 1.9%--the number of times you use an automatic teller machine (ATM)
in a month;
Study/Source of Data: Providian. "Credit Cards and Mainstream America."
Results: 60% of respondents answered "false" to the following
statement: Your credit rating is not affected by how much you charge on
your credit cards;
Study/Source of Data: Hilgert, Hogarth, and Beverly. "Household
Financial Management."
Results:
* 67% of respondents answered "false"--your credit score mainly
measures your knowledge of consumer credit. 27% of respondents answered
"true.";
* 38% answered "true"--your credit score may be lowered if you apply
for a credit card and 55% said "false.";
* 45% answered "true"--your credit score may be lowered if you use all
of the credit available on your credit card;
* Regarding who uses credit scores to decide whether they can use a
service and/or at what cost:
*92% answered that mortgage lenders may;
*87%--credit card companies;
*62%--cell phone companies;
*59%--landlord;
*59%--home insurers; and;
*40%--electric utility companies;
Study/Source of Data: CFA. "Consumers Lack Essential Knowledge."
Credit scores: How many consumers have obtained credit scores:
Results: Have you ever obtained your credit score?
* 53%--yes; and;
* 45%--no;
* 2%--did not know;
Of those who had obtained their score, they obtained it:
* 46% within the past year;
* 22%--1 to 2 years ago;
* 19%--2 to 5 years ago;
* 10%--more than 5 years ago; and;
* 2%--did not know;
Of those who had obtained their score, they said they had obtained it
from:
* 35% mortgage lender or broker;
* 28% credit bureau or reporting agency;
* 12% credit issuer or other consumer lender;
* 19% some other company or source; and;
* 5% did not know/none of these;
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
Results: 31.9% of respondents said that they knew what their current
personal credit score was;
Study/Source of Data: Providian. "Credit Cards and Mainstream America."
Results: 24% of respondents 45 years and older had obtained a copy of
their credit score;
Study/Source of Data: AARP. "Consumer Experience Survey."
Results: 25% of respondents said that they knew their credit score;
Study/Source of Data: CFA. "Consumers Lack Essential Knowledge."
Results: Have you ever seen or reviewed your own credit score?
* 30.7% yes; and;
* 69.3% no;
Those who had seen their score were asked for what purpose did they see
their credit score:
* 65.9% said due to a loan/major purchase; and;
* 27.5% said to keep informed/personal/curiosity;
Study/Source of Data: Fair Isaac Corporation. "myFICO Consumer Survey
Results."
Credit scores: Other:
* 10% of respondents could answer the question, "What is your credit
score?" with a correct three-digit response ranging between 300 and
850;
* One in eight knew what constituted a "good" credit score (650-850);
Study/Source of Data: TrueCredit, "TrueCredit Survey Reveals Startling
Illiteracy: Nearly 90 Percent of Americans Do Not Know Their Credit
Scores," September 15, 2004.
Results: 34% of respondents over the age of 45 said their knowledge of
credit scores was excellent/good and 61% said it was fair/poor;
Study/Source of Data: AARP. "Consumer Experience Survey."
Results: 85% of respondents knew what a credit score was;
Study/Source of Data: Household International. "Household Your Credit
Counts Challenge."
Results: 78% of consumers agree that it was important to know their
credit score, yet only 26% actually did;
Study/Source of Data: Household International. "Household Your Credit
Counts Challenge."
Results: 65% of respondents believed they should check their credit
score at least once per year, but only 53% of these people had done so
within the last year;
Study/Source of Data: Household International. "Household Your Credit
Counts Challenge."
Dispute Resolution: What consumers understand about the dispute
resolution process:
* 32% of respondents answered "false" to the following statement:
"Consumers who believe their credit reports or scores are inaccurate
must contact their lenders;"
* 64% of respondents answered "true;"
Study/Source of Data: CFA. "Consumers Lack Essential Knowledge."
Dispute Resolution: Disputes and inaccuracy:
Results: Of the 57.4 million credit files disclosed to consumers in
2003, 12.5 million were reinvestigated/disputed;
Study/Source of Data: CDIA collected data from the three nationwide
CRAs and provided it to GAO in January 2005. As noted on page 1 of this
appendix, the data on credit files issued from 2003 includes all direct-
to-consumer disclosures and all consumer relations disclosures.
Results:
* 14% of revolving credit accounts were reported without information
about credit limits;
* Individuals with scores below 600 tended to have the highest
frequency of data problems and were the most likely to experience a
score increase of 10 points or more in response to corrections of data
problems; and;
* Individuals with scores above 660 had the lowest incidence of data
problems;
Study/Source of Data: Avery, Robert B., Paul S. Calem, and Glenn B.
Canner. "Credit Report Accuracy and Access to Credit." Federal Reserve
Bulletin, Summer 2004. The Federal Reserve analyzed the credit records
(drawn as of June 30, 2003) of a nationally representative sample of
individuals and examined the possible effects of data limitations on
consumers by estimating the changes in consumers' credit history scores
that would result from correcting data problems in their credit
records.
Results:
* 25% of the credit reports had serious errors that could result in the
denial of credit;
* 54% had personal demographic information that was incorrect;
* 22% listed the same mortgage or loan twice;
* almost 8% were missing major credit, loan, mortgage, or other
consumer accounts that demonstrate the creditworthiness of the
consumer;
* 30% contained credit accounts that had been closed by the consumer
but remained listed as open; and;
* altogether 79% contained either serious errors or other mistakes;
Study/Source of Data: U.S. Public Interest Research Group (PIRG)--also
known as the National Association of State PIRGs. "Mistakes Do Happen:
A Look at Errors in Consumer Credit Reports," June 2004. Available at
http://uspirg.org/uspirg.asp?id2=13649&id3=USPIRG&. PIRG asked adults
in 30 states to order their credit reports and complete a survey on the
reports' accuracy. All of those surveyed were PIRG citizen members, in
addition to PIRG staff, coalition partners, and friends and family.
PIRG collected 197 surveys from 154 adults in 30 states.
Results:
* Over 8 million consumers (of the 16 million who received credit
reports) did not call a CRA back after receiving a copy of their credit
report;
* 50% of callbacks each month to CRAs are of an educational nature--
that is consumers simply have questions about their reports;
Study/Source of Data: CDIA. Prepared testimony from CEO and President
Stuart Pratt, The Fair Credit Reporting Act and Issues Presented by
Reauthorization of the Expiring Preemption Provisions.
Results:
* 46% of credit file data was verified as the information reported;
* 27% of data was modified per data furnisher's instruction but this
could be due to an update of information rather than a dispute;
* 10.5% of data was deleted per data furnisher's direction; and;
* 16% of data was deleted due to expiration of the 30-day period and no
response received from the data furnisher (it cannot be determined
whether the data was accurate or not);
Study/Source of Data: CDIA. Prepared testimony from CEO and President
Stuart Pratt, The Fair Credit Reporting Act and Issues Presented by
Reauthorization of the Expiring Preemption Provisions. These results
are based on an industry-wide data set of disclosures to consumers and
dispute responses to disputes submitted. CDIA collected the data from
their nationwide consumer reporting members.
Results:
* Average number of files issued per year (for the data set)--180,000;
* Average number of contacts/disputes relative to this population--
9,000; and;
* Based on these numbers, approximately 5% of consumers requesting
their files disputed;
Study/Source of Data: CDIA. Prepared testimony from CEO and President
Stuart Pratt, The Fair Credit Reporting Act and Issues Presented by
Reauthorization of the Expiring Preemption Provisions. Based on a data
set collected by CDIA on the number of consumers who reviewed their
credit files after receiving notices about (1) a certain level of
inquiry activity or (2) additional adverse information being reported.
Results: Only four tenths of one percent of all reports sold in this
country ended up with the consumer contacting a CRA;
Study/Source of Data: CDIA. Prepared testimony from CEO and President
Stuart Pratt, The Fair Credit Reporting Act and Issues Presented by
Reauthorization of the Expiring Preemption Provisions. These results
were based on data collected by CDIA on 2 billion credit reports sold
annually.
Results:
* 61 (32%)--number of the 189 reports that were accurate but needed to
be updated due to a new account or other new additional information;
and;
* 2 (1%)--number of reports with an identified inaccuracy based on
direct contact with the original provider of the information in the
report;
Study/Source of Data: CDIA. Prepared testimony from CEO and President
Stuart Pratt, The Fair Credit Reporting Act and Issues Presented by
Reauthorization of the Expiring Preemption Provisions. These results
were based on data collected by CDIA based on a sample of 189 three-
file merged reports that were reviewed by a mortgage reporting service
and some were updated via direct contact with data furnishers.
Results:
* About 70% of individuals in the sample had a missing credit limit on
one or more of their revolving accounts;
* About 40% of individuals with public records had more than one such
record, and about 40% of those with accounts reported by collection
agencies had more than one collection item. For many of these
individuals, the multiple record items appeared to be pertaining to the
same episode;
Study/Source of Data: Avery, Robert B., Paul S. Calem, Glenn B. Canner,
and Raphael W. Bostic. "An Overview of Consumer Data and Credit
Reporting." Federal Reserve Bulletin, February 2003. The Federal
Reserve analyzed 248,000 credit records (excluding any identifying
information) that were randomly selected from a nationally
representative sample of credit records (as of June 1999) provided by
one of the nationwide CRAs.
Results:
* 78% of credit files omitted a revolving account in good standing;
* 33% were missing a mortgage account that had never been late;
* 67% omitted other types of installment accounts that had never been
late;
* 82% of credit files had inconsistencies regarding the balance of
revolving accounts or collections; and;
* 96% had inconsistencies regarding an account's credit limit;
Study/Source of Data: CFA and National Credit Reporting Association,
Credit Score Accuracy and Implications for Consumers, December 2002.
CFA and NCRA initially reviewed 1,704 credit files representing
consumers from 22 states and then reexamined a sample of 51 three-
agency merged files. These results are based on the reexamination.
Results: Including all types of ID theft, a total of 4.6 percent of
respondents indicated that they had discovered they were victims of
identity theft in the past year;
Study/Source of Data: Federal Trade Commission (FTC). "Federal Trade
Commission--Identity Theft Survey Report," September 2003. Available at
www.ftc.gov.
Identity Theft:
Results:
* Are you aware of ID theft?
*83.6%--yes; and;
*16.4% no;
* Have you or someone you know been the victim of ID theft?
*28.2%--yes;
* How concerned are you about being the victim of ID theft?
*20.4%--extremely concerned;
*29.1%--very concerned;
*34.4%--somewhat;
*10.7%--not very; and;
*5.1%--not at all;
Study/Source of Data: Fair Isaac Corporation. "myFICO Consumer Survey
Results."
Legislation:
Results: Later this year, a new federal law giving consumers more
credit score rights will go into effect. Do you think this new law
gives consumers the right to obtain their credit score… (the following
percentages of respondents said "yes" to the statements below):
* 74%--for free if they have been denied a mortgage loan;
* 72%--for free once a year;
* 71%--for free if they have been denied a credit card; and;
* 69%--at any time for a small fee;
Study/Source of Data: CFA and Providian. "Most Consumers Do Not
Understand Credit Scores."
Source: GAO.
[A] In the body of our report, we rounded this and other percentages to
the nearest whole number.
[End of table]
[End of section]
Appendix IV: Certain Demographics and Credit Experiences Are Associated
with Consumers' Understanding of Credit Reporting Issues:
Using two methodologies--a regression analysis and cross-tabulation
analysis--that generally provided consistent conclusions, we found that
certain factors were associated with consumers' knowledge of credit
reporting issues.[Footnote 81] First, we found that having experience
with credit--for example, having an automobile loan or a mortgage--and
the credit reporting process had a statistically significant effect on
consumers' knowledge of credit issues. In addition, we found that
certain demographic factors, such as educational level, income level,
race/ethnicity, and age had a significant effect on consumers'
awareness of these issues. Finally, we found that other factors,
including gender and residence in one of the seven "free report
states," did not have a significant effect on consumers' credit
reporting knowledge.[Footnote 82]
Those with Less Education and Lower Incomes Scored Lower on Our Survey:
Our survey results show that awareness of credit reporting issues
generally increased as a consumers' educational level increased. This
trend was most evident for those with less than a high school
education. As figure 10 shows, consumers with less than a high school
education had a lower mean score overall (36 percent) than those with a
high school education or more (at least 52 percent). Our regression
analysis also confirmed that increases in educational level had a
significant effect on consumers' knowledge. For example, compared with
having less than a high school degree, having earned at least a
bachelor's degree could increase scores by more than 13 percentage
points.
Figure 10: Average Total Scores and Demographic Groups:
[See PDF for image]
Note: Brackets on each bar represent the sampling error (or confidence
interval) for that estimate at the 95 percent level of confidence.
[End of figure]
In addition, as educational level increased, consumers obtained reports
and scores and disputed information more often than those with less
education. Figure 11 shows at least 50 percent of those with a high
school education or more had viewed their credit reports, compared with
25 percent of consumers with less than a high school education.
Similarly, at least 23 percent of consumers with a high school
education or more had obtained their scores, compared with 9 percent of
consumers with less than a high school education. Further, those with
some postsecondary education disputed more frequently than those with
less education (22 percent versus 7-13 percent, respectively).
Figure 11: Credit Reporting Experience by Educational Level:
[See PDF for image]
[End of figure]
We also found that consumers with less than a high school degree were
less aware of credit reporting issues in several areas, including the
"basic" questions, than those with more education (table 5). For
example, we found that those with less than a high school degree were
less able to identify the correct definition of a credit reporting
agency (CRA) (44 percent) than those with higher levels of education
(74 percent and more). These consumers were also less aware that they
had the right to check their credit reports at any time for any reason
and less likely to know that skipping loan payments and making late
credit card payments had a negative effect on their credit scores. We
found that 10 percent of consumers with less than a high school
education named the Internet as a source for obtaining their credit
report, compared with 24 percent or more of those with a high school
education or more.
Table 5: Selected Survey Responses and Educational Level:
Credit Report Section:
Survey Question: Knew the correct definition of a CRA;
Less than high school: 44%;
Education (in percentages): High school: 74%;
Education (in percentages): Some post-secondary: 89%;
Education (in percentages): Bachelor's degree or above: 97%.
Survey Question: Knew they had the right to obtain their report at any
time;
Less than high school: 37%;
Education (in percentages): High school: 63%;
Education (in percentages): Some post-secondary: 76%;
Education (in percentages): Bachelor's degree or above: 85%.
Survey Question: Knew race was not on their credit report;
Less than high school: 37%;
Education (in percentages): High school: 54%;
Education (in percentages): Some post-secondary: 65%;
Education (in percentages): Bachelor's degree or above: 66%.
Survey Question: Knew credit history could impact insurance coverage
and premiums;
Less than high school: 22%;
Education (in percentages): High school: 34%;
Education (in percentages): Some post-secondary: 38%;
Education (in percentages): Bachelor's degree or above: 41%.
Survey Question: Credit Score Section:
Survey Question: Knew the correct definition of a credit score;
Less than high school: 31%;
Education (in percentages): High school: 57%;
Education (in percentages): Some post-secondary: 79%;
Education (in percentages): Bachelor's degree or above: 88%.
Survey Question: Knew skipping loan payments could affect scores
negatively;
Less than high school: 55%;
Education (in percentages): High school: 84%;
Education (in percentages): Some post-secondary: 93%;
Education (in percentages): Bachelor's degree or above: 95%.
Survey Question: Knew that making late credit card payments could
affect scores negatively;
Less than high school: 56%;
Education (in percentages): High school: 83%;
Education (in percentages): Some post-secondary: 93%;
Education (in percentages): Bachelor's degree or above: 96%.
Dispute Section:
Survey Question: Knew they had the right to dispute information on
their report;
Less than high school: 72%;
Education (in percentages): High school: 87%;
Education (in percentages): Some post-secondary: 93%;
Education (in percentages): Bachelor's degree or above: 96%.
Source: GAO.
Note: The sampling errors for those with less than high school
education are +/-7 percentage points or less.
[End of table]
Consumers with relatively low household incomes were also less aware of
issues related to credit reports, credit scores, and the dispute
process than consumers in higher-income households. For example, those
in the lowest household income group (less than $25,000 annually)
correctly answered an average of 47 percent of survey questions, while
those in the highest household income group ($75,000 and above
annually) answered around 63 percent (fig. 10). Our regression model
confirmed these results and found that income had a significant effect
on knowledge. For example, a consumer with a household income of less
than $25,000 per year could score almost 3 percentage points lower on
the overall survey than a consumer with a higher household income.
The percentage of consumers who had viewed their reports and scores and
disputed inaccurate information generally increased with household
income level (fig. 12). For example, of those consumers with household
incomes of $25,000 or less, 38 percent had viewed their credit reports,
19 percent had obtained their scores, and 10 percent had undertaken the
dispute process (fig. 12). In contrast, of those consumers with
household incomes of $75,000 or more, 75 percent had viewed their
credit report, 49 percent had obtained their credit score, and 24
percent had undertaken the dispute process.
Figure 12: Credit Reporting Experience and Household Income Level:
[See PDF for image]
Note: The sampling errors for those with incomes of $50,000 -$74,999
and $75,000 or more are +/-7 percentage points or less.
[End of figure]
Consumers with relatively low household incomes--in particular, those
in households making less than $25,000--generally were less aware of
issues pertaining to credit reports, scores, and the dispute process
compared with those in higher income groups (table 6). For example,
consumers in households making less than $25,000 were less likely than
those in households making $50,000 or more to know that their credit
history could affect the interest rates lenders offered them. In
addition, consumers in the lowest household income group were less
likely to name the Internet as a place to go for one's credit report
compared with other household income groups (about 17 percent compared
with 32-36 percent).[Footnote 83] Finally, the differences in knowledge
between lower-and higher-income consumers were most pronounced for
credit scores. For example, 44 percent of consumers in the lowest
household income group knew that frequently applying for credit
affected scores negatively, compared with 78 percent of consumers in
the highest income group.
Table 6: Selected Survey Responses of Different Household Income
Levels:
Credit Report Section:
Survey Question: Knew credit history could affect interest rates on
loans;
Income (in percentages): Less than $25K: 73%;
Income (in percentages): $25K -$49K: 80%;
Income (in percentages): $50K -$74K: 88%;
Income (in percentages): $75K and above: 89%.
Survey Question: Knew credit history could affect employment decisions;
Income (in percentages): Less than $25K: 27%;
Income (in percentages): $25K -$49K: 33%;
Income (in percentages): $50K -$74K: 34%;
Income (in percentages): $75K and above: 39%.
Credit Score Section:
Survey Question: Knew the correct definition of a credit score;
Income (in percentages): Less than $25K: 49%;
Income (in percentages): $25K -$49K: 71%;
Income (in percentages): $50K
-$74K: 81%;
Income (in percentages): $75K and above: 88%.
Survey Question: Knew having credit history for a short time could
affect scores negatively;
Income (in percentages): Less than $25K: 36%;
Income (in percentages): $25K -$49K: 47%;
Income (in percentages): $50K -$74K: 58%;
Income (in percentages): $75K and above: 60%.
Survey Question: Knew frequently applying for credit could affect
scores negatively;
Income (in percentages): Less than $25K: 44%;
Income (in percentages): $25K -$49K: 60%;
Income (in percentages): $50K -$74K: 68%;
Income (in percentages): $75K and above: 78%.
Survey Question: Knew using most of your available credit could affect
scores negatively;
Income (in percentages): Less than $25K: 40%;
Income (in percentages): $25K -$49K: 52%;
Income (in percentages): $50K -$74K: 59;
Income (in percentages): $75K and above: 62%.
Survey Question: Knew requesting a copy of your own report had no
effect on credit scores;
Income (in percentages): Less than $25K: 35%;
Income (in percentages): $25K -$49K: 50%;
Income (in percentages): $50K -$74K: 53%;
Income (in percentages): $75K and above: 64%.
Source: GAO.
Note: The sampling errors for those with incomes of $50,000 -$74,999
and $75,000 or more are +/-7 percentage points or less.
[End of table]
Race/Ethnicity Was Associated with Differences in Survey Scores:
We also found that race/ethnicity was associated with consumers'
knowledge of credit reporting issues. We found that African Americans
and whites scored similarly, showing approximately equal levels of
knowledge, while Hispanics scored consistently lower on our survey. As
shown in figure 10, African Americans earned an average score on the
survey of 55 percent, whites had a score of 58 percent, and Hispanics
earned 43 percent. The regression confirmed these findings. It showed
that being African American rather than white had no significant effect
on consumer's scores but that being Hispanic could decrease survey
scores by approximately 6 percentage points compared with whites
(holding other variables, such as education and income
constant).[Footnote 84]
As shown in Figure 13, we found that African Americans and whites were
almost equally as likely to have viewed their credit reports and
obtained their credit scores and that Hispanics were less likely to
have done either. More African Americans than whites said that they had
ordered their credit reports themselves--65 percent compared with 50
percent of whites.[Footnote 85] In addition, more African Americans (15
percent) than whites (6 percent) said that they had ordered their
reports because of adverse action.
Figure 13: Credit Reporting Experience by Race/Ethnicity:
[See PDF for image]
[End of figure]
Table 7 highlights some specific survey responses for each
racial/ethnic group, including responses that varied somewhat from
overall survey scores. Whites and African-Americans scored similarly on
the survey and consistently higher than Hispanics. For example, 84
percent of African Americans and 92 percent of whites knew that
skipping loan payments could lower their credit score. And around the
same percentage of both groups (about 90 percent) knew that they could
dispute information on their credit reports. As shown in table 7,
Hispanics scored lower on these questions. However, when asked whether
race appeared on credit reports, Hispanics scored virtually the same as
African Americans, though still lower than whites. Finally, we found
that more whites (35 percent) named the Internet as a source for
obtaining their credit reports than African Americans (18 percent) and
Hispanics (16 percent).
Table 7: Selected Survey Responses by Race/Ethnicity:
Credit Report Section:
Survey question: Knew they had the right to obtain their report at any
time;
Race/Ethnicity (in percentages): African American: 66%;
Race/Ethnicity (in percentages): Hispanic: 48%;
Race/Ethnicity (in percentages): White: 77%.
Survey question: Knew race was not on their credit report;
Race/Ethnicity (in percentages): African American: 51%;
Race/Ethnicity (in percentages): Hispanic: 50%;
Race/Ethnicity (in percentages): White: 63%.
Credit Score Section:
Survey question: Knew the correct definition of a credit score;
Race/Ethnicity (in percentages): African American: 59%;
Race/Ethnicity (in percentages): Hispanic: 39%;
Race/Ethnicity (in percentages): White: 78%.
Survey question: Knew skipping loan payments could affect scores
negatively;
Race/Ethnicity (in percentages): African American: 84%;
Race/Ethnicity (in percentages): Hispanic: 63%;
Race/Ethnicity (in percentages): White: 92%.
Survey question: Knew making late credit card payments could affect
scores negatively;
Race/Ethnicity (in percentages): African American: 86%;
Race/Ethnicity (in percentages): Hispanic: 69%;
Race/Ethnicity (in percentages): White: 91%.
Survey question: Dispute Section:
Knew they had the right to dispute information on their credit report;
Race/Ethnicity (in percentages): African American: 89%;
Race/Ethnicity (in percentages): Hispanic: 78%;
Race/Ethnicity (in percentages): White: 92%.
Source: GAO.
[End of table]
Our survey findings are consistent with information we collected during
interviews with federal agencies and other organizations, including the
National Council of La Raza (La Raza). The Federal Deposit Insurance
Corporation (FDIC), the Office of the Comptroller of the Currency
(OCC), and FTC told us that their financial literacy efforts are often
targeted at certain populations, including Hispanics, who may more
often be "unbanked" or less financially literate.[Footnote 86]
Government officials and representatives of other organizations,
including La Raza, discussed the need for financial literacy programs
among the low-income Latino community.[Footnote 87] They also told us
that some Latinos tend to avoid debt, making them less likely to obtain
credit and thus less likely to have experience with credit reports and
scores. In our survey, we found that more white consumers had
experience with credit-related products than African Americans or
Hispanics, with Hispanics having the least amount of experience (table
8).
Table 8: Experience with Credit-Related Products and Race/Ethnicity:
Have had the following credit-related products at some point in time:
Credit card;
Race/Ethnicity (in percentages): African American: 75%;
Race/Ethnicity (in percentages): Hispanic: 62%;
Race/Ethnicity (in percentages): White: 92%.
Have had the following credit-related products at some point in time:
Automobile loan;
Race/Ethnicity (in percentages): African American: 62%;
Race/Ethnicity (in percentages): Hispanic: 45%;
Race/Ethnicity (in percentages): White: 80%.
Have had the following credit-related products at some point in time:
Home loan/mortgage;
Race/Ethnicity (in percentages): African American: 42%;
Race/Ethnicity (in percentages): Hispanic: 28%;
Race/Ethnicity (in percentages): White: 73%.
Source: GAO.
[End of table]
Younger and Older Consumers Scored Lower on Our Survey:
Our cross-tabulation analysis showed that consumers' knowledge of
credit reporting issues also differed with age and that the youngest
consumers (aged 18 to 24) and the oldest (aged 65 and older) were the
least likely to have obtained a credit report or score or to have
undertaken the dispute process. In the regression, each 1-year increase
in the respondent's age was associated with about a tenth of a
percentage point decrease in the respondent's total knowledge
score.[Footnote 88] The mean scores for credit knowledge for consumers
in the six age groups ranged from 49 to 60 percent, with the youngest
and oldest groups having the lowest scores (fig. 10).
We found that experience with the credit reporting process also varied
somewhat by age (fig. 14). For example, 24 percent of the youngest
consumers and 42 percent of the oldest had ordered their credit
reports, compared with 61 to 74 percent of all other age
groups.[Footnote 89] In addition, 7 percent of the youngest consumers
and 17 percent of the oldest consumers had obtained their credit
scores, compared with 33 to 48 percent of those in all other age
groups.[Footnote 90] The differences were similar for dispute, with 7
percent of the youngest group and 10 percent of the oldest disputing
information in their credit reports; in comparison, 16 to 27 percent in
all other age groups reported having disputed information.[Footnote 91]
Figure 14: Credit Reporting Experience by Age Group:
[See PDF for image]
Note: The sampling errors are +/-7 percentage points or less, except
for the 24 percent of those aged 18 to 24 who had seen their reports
(+/-9 percentage points).
[End of figure]
According to our survey, consumers in the youngest and oldest age
groups also knew fewer details about credit reports and scores than
those in other age groups (table 9). For instance, 18 percent of those
aged 18 to 24 knew that their credit history could affect employment
decisions--an issue of particular importance to persons in this age
group, who are likely to be engaging in their first job search--while
29 to 46 percent of those in all other age groups knew this
fact.[Footnote 92] In addition, fewer consumers aged 18 to 24--55
percent--and those aged 65 and over--60 percent--could identify the
correct definition of a credit score compared with consumers in other
age groups (table 9).[Footnote 93] Younger consumers were also less
aware that having had a credit history for a short time could affect
scores negatively, especially when compared with those aged 25 to 34
(41 percent compared with 61 percent).[Footnote 94] In addition, 71
percent of consumers aged 65 and over said they should check their
credit reports from time to time, while 88 to 92 percent of consumers
in all other age groups said they should.[Footnote 95] Finally, 14
percent of consumers in the oldest age group named the Internet as a
source for obtaining their credit report, while 29 to 37 percent of
those in all other age groups cited the Internet.[Footnote 96]
Table 9: Selected Survey Responses by Age Group:
Credit Report Section:
Survey Question: Knew credit history could affect employment decisions;
Age (in percentages): 18 to 24: 18%;
Age (in percentages): 25 to 34: 29%;
Age (in percentages): 35 to 44: 35%;
Age (in percentages): 45 to 54: 31%;
Age (in percentages): 55 to 64: 46%;
Age (in percentages): 65 +: 36%.
Survey Question: Knew credit history could affect ability to rent an
apartment;
Age (in percentages): 18 to 24: 80%;
Age (in percentages): 25 to 34: 74%;
Age (in percentages): 35 to 44: 75%;
Age (in percentages): 45 to 54: 73%;
Age (in percentages): 55 to 64: 79%;
Age (in percentages): 65 +: 61.
Credit Score Section:
Survey Question: Knew the correct definition of a credit score;
Age (in percentages): 18 to 24: 55%;
Age (in percentages): 25 to 34: 74%;
Age (in percentages): 35 to 44: 80%;
Age (in percentages): 45 to 54: 77%;
Age (in percentages): 55 to 64: 74%;
Age (in percentages): 65 +: 60.
Survey Question: Knew having credit for a short time could affect
scores negatively;
Age (in percentages): 18 to 24: 41%;
Age (in percentages): 25 to 34: 61%;
Age (in percentages): 35 to 44: 53%;
Age (in percentages): 45 to 54: 53%;
Age (in percentages): 55 to 64: 52%;
Age (in percentages): 65 +: 30.
Survey Question: Knew frequently applying for new credit could affect
scores negatively;
Age (in percentages): 18 to 24: 60%;
Age (in percentages): 25 to 34: 74%;
Age (in percentages): 35 to 44: 62%;
Age (in percentages): 45 to 54: 59%;
Age (in percentages): 55 to 64: 65%;
Age (in percentages): 65 +: 44.
Source: GAO.
Note: The sampling errors are +/-7 percentage points or less, except
for the three estimates under the credit score section for those aged
18 to 24. These errors are +/-10 percentage points; for the 18 percent
of those aged 18 to 24 who knew their credit history could affect
employment decisions, the error is +/-8 percentage points.
[End of table]
Actively Employed Consumers Scored Slightly Higher on Our Survey:
We also analyzed survey results for consumers in different employment
groups and found some differences. Specifically, we found that actively
employed consumers working either full or part time showed slightly
more knowledge of credit reporting issues than any other group
answering 59 percent of the survey questions correctly. Students and
unemployed consumers (including homemakers, retirees, and those lacking
any job) had somewhat lower scores--53 percent for students and 49
percent for consumers who were unemployed. However, our regression
analysis showed that students made up the only employment group that
had a significant effect and that their employment status could
potentially lower their scores by about 3 percentage points.[Footnote
97]
As figure 15 shows, we also found that actively employed consumers were
more likely to obtain their credit reports and scores than consumers in
other employment categories. However, we found little difference among
the percentages of those in each employment group that reported having
disputed information on their credit reports.
Figure 15: Credit Reporting Experience by Employment Group:
[See PDF for image]
Note: The sampling error for students is +/-11 percentage points or
less, except for the 41 percent of student who had obtained their
report (+/-17 percentage points).
[End of figure]
Table 10 highlights some other responses to specific survey questions
among the employment groups. Also, more students and unemployed
consumers were unable to name any CRAs (84 percent and 79 percent,
respectively), while 59 percent of actively employed consumers were
unable to name any.[Footnote 98]
Table 10: Selected Survey Responses by Employment Group:
Survey Question.
Credit Report Section:
Knew credit history could affect employment decisions;
Employment group (in percentages): Actively employed: 32%;
Employment group (in percentages): Student: 29%;
Employment group (in percentages): Not employed: 34%.
Credit Score Section:
Knew the correct definition of a credit score;
Employment group (in percentages): Actively employed: 77%;
Employment group (in percentages): Student: 70%;
Employment group (in percentages): Not employed: 58%.
Dispute Section:
Knew they could add a statement to their credit report; Employment
group (in percentages): Actively employed: 43%;
Employment group (in percentages): Student: 21%;
Employment group (in percentages): Not employed: 39%.
Source: GAO.
Note: The sampling error for students who knew that their credit
history could affect employment decisions is +/-16 percentage points;
all other sampling errors for students are +/-14 percentage points or
less.
[End of table]
Consumers with Credit Experiences Scored Higher on Our Survey:
Our analysis also showed that consumers who had viewed their reports,
obtained their scores, or disputed information demonstrated more
knowledge of credit reporting issues. Our findings are consistent with
expert opinion in the field of financial literacy. Several experts have
noted that consumers who have experience with a financial product or
transaction--in other words, those who have been able to "learn by
doing"--are better informed about financial products and transactions
than those who lack experience. Specifically, as shown in figure 16, we
found that consumers who had viewed their credit reports had a higher
average survey score--62 percent--than those who had not--47 percent.
Consumers who had seen their credit scores were better informed about
credit issues, with a mean survey score of 64 percent, while those who
had not obtained their scores had a mean survey score of 51 percent.
Those consumers who had undertaken the dispute process also
demonstrated increased knowledge. Those who had disputed had a mean
score of 65 percent, while those who had not had a mean score of 53
percent. Our regression analysis also confirmed that having viewed a
report or score or having disputed were associated with an increase in
the consumer's total knowledge score. Each experience was associated
with at least a 4 percentage point increase in the total knowledge
score.
Figure 16: Average Total Scores and Credit Experience:
[See PDF for image]
Note: Brackets on each bar represent the sampling error (or confidence
interval) for that estimate at the 95-percent level of confidence.
[End of figure]
Table 11 shows some of the differences in selected survey responses for
those consumers who had obtained their credit reports and scores and
had engaged in the dispute process and those who had not had these
experiences. We found that having one of these three experiences could
increase responses to certain questions by almost 20 percent--for
instance, knowing how long information remained on a credit report or
that frequently applying for new credit could have a negative effect on
a credit score.
Table 11: Selected Survey Responses and Consumers' Credit Reporting
Experience:
Credit Report Section:
Survey Question: Knew how long information remained on report;
Credit report (in percentages): Viewed credit report: 59%;
Credit report (in percentages): Did not view credit report: 30%;
Credit score (in percentages): Obtained credit score: 64%;
Credit score (in percentages): Did not obtain credit score: 38%;
Dispute (in percentages): Disputed: 69%;
Dispute (in percentages): Did not dispute: 42%.
Survey Question: Knew credit history could affect the rate of interest
charged on a loan;
Credit report (in percentages): Viewed credit report: 88%;
Credit report (in percentages): Did not view credit report: 71%;
Credit score (in percentages): Obtained credit score: 93;
Credit score (in percentages): Did not obtain credit score: 75%;
Dispute (in percentages): Disputed: 91%;
Dispute (in percentages): Did not dispute: 79%.
Credit Score Section:
Survey Question: Knew having a credit history for a short time could
affect scores negatively;
Credit report (in percentages): Viewed credit report: 59%;
Credit report (in percentages): Did not view credit report: 36%;
Credit score (in percentages): Obtained credit score: 64%;
Credit score (in percentages): Did not obtain credit score: 41%;
Dispute (in percentages): Disputed: 61%;
Dispute (in percentages): Did not dispute: 46%.
Survey Question: Knew frequently applying for credit could affect
scores negatively;
Credit report (in percentages): Viewed credit report: 70%;
Credit report (in percentages): Did not view credit report: 47%;
Credit score (in percentages): Obtained credit score: 76%;
Credit score (in percentages): Did not obtain credit score: 52%;
Dispute (in percentages): Disputed: 75%;
Dispute (in percentages): Did not dispute: 57.
Survey Question: Knew using most of your available credit could affect
scores negatively;
Credit report (in percentages): Viewed credit report: 59%;
Credit report (in percentages): Did not view credit report: 41%;
Credit score (in percentages): Obtained credit score: 61%;
Credit score (in percentages): Did not obtain credit score: 47%;
Dispute (in percentages): Disputed: 68%;
Dispute (in percentages): Did not dispute: 48%.
Source: GAO.
[End of table]
Second, we found that consumers with certain types of credit
experiences, such as having had either an automobile loan or a home
loan/mortgage, scored higher on the overall survey and that more of
these consumers had viewed their credit reports and scores and disputed
information on their reports (fig. 17). As we have seen, consumers who
had obtained either an automobile loan or a mortgage--about 74 percent
of consumers--had higher overall scores than those who had not (fig.
16). The regression showed that having an automobile loan or mortgage
had a significant effect and could increase a consumer's score by
around 4 percentage points.
Figure 17: Credit Reporting Experience and Experience with Automobile
Loans or Home Loans/Mortgages:
[See PDF for image]
[End of figure]
Table 12 highlights some of the larger differences in survey responses
between consumers who had had automobile loans or mortgages and those
who had not. For example, consumers who had had automobile loans or
mortgages were most familiar with some of the factors that might or
might not affect credit scores--frequently applying for credit, having
low checking account balances, and using most of the credit available
to them--than consumers who had not had such loans.
Table 12: Selected Survey Responses and Experience with Automobile
Loans or Home Loans/Mortgages:
Credit Report Section:
Survey Question: Knew they had the right to obtain their report at any
time;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 75%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
56%.
Survey Question: Knew how long information remained on report;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 52%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
30%.
Credit Score Section:
Survey Question: Knew the correct definition of a credit score;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 76%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
49%.
Survey Question: Knew having a credit history for a long time could
affect scores positively;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 85%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
59%.
Survey Question: Knew frequently applying for credit could affect
scores negatively;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 65%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
43%.
Survey Question: Knew low checking account balances had no effect on
scores;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 48%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
27%.
Survey Question: Knew using most of the credit available to you could
affect scores negatively;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 57%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
34%.
Dispute Section:
Survey Question: Knew they had the right to add a statement to their
credit report;
Auto loan/mortgage (in percentages): Had an auto/mortgage loan: 46%;
Auto loan/mortgage (in percentages): Did not have auto/mortgage loan:
24%.
Source: GAO.
[End of table]
Being a Victim of Identity Theft and Being from a Free Report State Had
Little Effect on Consumers' Survey Scores:
Government and industry officials told us that certain experiences,
including having been a victim of identity theft or having requested a
credit report because of an adverse action such as denial of credit,
might increase consumers' knowledge of credit issues. However, we found
that these experiences did not affect consumers' overall scores. As
figure 16 shows, consumers who said they had been victims of identity
theft (approximately 10 percent of our survey respondents) scored only
slightly higher than other consumers. The regression analysis showed
that being a victim of identity theft did not have a significant effect
on consumers' knowledge. Similarly, consumers who had obtained their
credit report because of an adverse action scored only slightly higher
than those who had not.[Footnote 99]
Although we did not see large differences in the overall knowledge of
credit issues between those who had been victims of identity theft and
those who had not been, we found that identity theft victims were more
likely to obtain their credit reports (72 percent versus 56
percent).[Footnote 100] In addition, we found that victims of identity
theft disputed information more frequently than those who had not had
this experience. Of those who had disputed, identity theft victims had
information removed from their reports less frequently than other
consumers who had disputed (48 percent versus 74 percent) and were more
likely to add an explanatory statement.[Footnote 101] Further, we found
that consumers who had been victims of identity theft had more
knowledge of certain details about credit reporting than consumers who
had not been victims. More--53 percent--knew that correcting
information with one CRA did not mean that the others would also make
the correction, compared with 39 percent of other consumers.[Footnote
102] In addition, like all other groups, victims of identity theft did
not name the Federal Trade Commission (FTC) as the agency to contact if
they had a problem with a CRA, although Congress designated FTC as the
agency that handles the identity theft claims of consumers. The number
of identity theft complaints filed with FTC doubled every year between
November 1999 and 2002, and in 2003 FTC received about 215,000 identity
theft complaints. Similarly, we did not see large differences in
overall knowledge between those who had obtained their report because
of an adverse action and those who had not. However, we found that
consumers who had had an adverse action against them were more likely
to dispute information. These consumers were also far more likely to
order their reports themselves than those who had not had this
experience (78 percent versus 49 percent).[Footnote 103]
We also looked at consumers who live in free report states--one of the
seven states where residents could order a free copy of their credit
report annually before the FACT Act was passed (about 21 percent of
respondents). On the basis of our discussions with government and
industry officials, we decided to test whether these consumers would
have more knowledge about credit reporting issues than consumers in
states where free reports were not available. As figure 10 shows,
consumers residing in free report states had only slightly higher
survey scores than consumers who did not (59 percent compared with 55
percent). The regression analysis showed no significant effect for
those living in a free report state compared with those that did not.
TransUnion, one of the three nationwide CRAs, found in a 2003 study
that around 4 percent of consumers in free report states had requested
their reports, compared with around 2 percent of consumers in other
states. Colorado was the exception, with more than 10 percent
requesting their reports, possibly because consumers were notified of
this right by mail under two specific circumstances.[Footnote 104] Our
study showed that 46 percent of consumers in free report states said
that they did not know they had the right to order a free credit
report.
Finally, we assessed the effect of gender (male versus female) on
consumers' knowledge and found that gender did not impact consumers'
knowledge of credit reporting, with males and females having similar
overall survey scores--57 percent and 54 percent, respectively (see
fig. 10). The regression model also showed that gender did not have a
significant independent effect on knowledge of credit issues. In
addition, approximately the same percentages of males and females had
obtained their credit reports and scores and had disputed information.
Further, we found few specific areas in which knowledge differed
between genders. Among those we did find, females were somewhat more
likely to know how long negative information remained on a credit
report (51 percent compared with 42 percent of males) and to say that
they had added a written statement to their report after a dispute (38
percent, compared with 21 percent of males).[Footnote 105] In contrast,
males were slightly more likely to know that their credit history could
affect the interest rate charged on a loan (85 percent compared with 77
percent of females) and that having had a credit history for a short
time had a negative effect on their credit score (54 percent, compared
with 44 percent of females).
[End of section]
Appendix V: A Regression Model to Explain Differences in Total
Knowledge Scores:
This appendix provides more detail on the regression model that GAO
developed to explain how different variables such as education are
associated with the survey scores, or the "total credit reporting
knowledge scores," discussed in the body of the report. Such a model
can be used to illustrate the influences of different independent
variables on the total knowledge score.[Footnote 106] In the first
section, we describe the variables used in the model. In the next two
sections, we describe the structure of the model and the statistical
results. In the last section, we discuss our interpretation of the
regression model.
Variables in the Model:
The model includes variables taken from the survey that represent
consumers' experience with the credit and financial system, and
consumers' socio-economic characteristics that might influence
consumers' knowledge of credit reporting issues. Table 13 shows the
proportion of socio-economic and experience variables of respondents to
our survey. There are many factors that might potentially affect
knowledge and this model only considered those that we asked
respondents about in our survey. Specifically, the socio-economic
variables are: race/ethnicity, age, income, employment status, gender,
and education. The race/ethnicity variable tests whether certain racial
or ethnic groups systematically score differently on the survey. Age is
included because of concerns that younger and older consumers may be
less aware of the importance of credit reports and scores. The income
variable tests the influence of different levels of a household's
annual income on total scores and the employment status variable tests
for the influences of having full-or part-time work, or none at all, as
well as being a student. The gender variable is included to determine
whether males and females have a different understanding of credit
reporting issues. Education levels are included to determine whether
consumers' level of formal education influence total knowledge scores.
Several variables reflect factors that we believe could directly affect
consumers' experience with and knowledge of credit reporting. For
example, living in a state where consumers receive copies of their
credit reports free each year before the passage of the FACT Act is
included because we hypothesized that having this right might have
encouraged more individuals to look at their credit reports, which
could increase their credit reporting knowledge. Similarly, we thought
that having had a mortgage or an automobile loan could mean that
consumers were more likely to know more about credit reporting issues.
We included having seen a credit report or credit score, disputing
information on a credit report, and experiencing identity theft to
assess the possible influence of these experiences on a consumer's
understanding of credit reporting issues.
Table 13: Variables in the Statistical Model:
Variable: Dependent Variable = Total Knowledge Score (percent correct).
Variable: Independent Variables: Ethnic Classifications;
Variable:
* White;
Percent of Sample: 68.8%.
Variable:
* Black;
Percent of Sample: 12.4%.
Variable:
* Hispanic;
Percent of Sample: 13.9%.
Variable:
* Other;
Percent of Sample: 4.9%.
Variable:
* Age*;
Percent of Sample: Not applicable since it is a continuous variable.
Variable: Independent Variables: Household Income (per year).
Variable:
* Below $25,000;
Percent of Sample: 28.2%.
Variable:
* $25,000 to $49,999;
Percent of Sample: 27.4%.
Variable:
* $50,000 to $74,999;
Percent of Sample: 20.3%.
Variable:
* $75,000 and above;
Percent of Sample: 24.1%.
Variable: Independent Variables: Employment Status.
Variable:
* Actively employed (full-or part-time employment);
Percent of Sample: 60.4%.
Variable:
* Student;
Percent of Sample: 3.7%.
Variable:
* Not employed (homemakers, retired or unemployed);
Percent of Sample: 31.1%.
Variable:
* Other;
Percent of Sample: 4.8%.
Variable: Independent Variables: Gender.
Variable:
* Male;
Percent of Sample: 48.2%.
Variable:
* Female;
Percent of Sample: 51.8%.
Variable: Independent Variables: Education Level.
Variable:
* Less than high school;
Percent of Sample: 10.4%.
Variable:
* High school completed;
Percent of Sample: 30.4%.
Variable:
* Some postsecondary;
Percent of Sample: 28.5%.
Variable:
* At least a college degree (Bachelor's degree or more);
Percent of Sample: 30.7%.
Variable: Independent Variables: Financial Experience.
Variable:
* Living in a free credit report state;
Percent of Sample: 12.8%.
Variable:
* Having had a mortgage or an automobile loan;
Percent of Sample: 77.0%.
Variable:
* Having seen one's credit report;
Percent of Sample: 57.8%.
Variable:
* Having seen one's credit score;
Percent of Sample: 32.8%.
Variable:
* Having disputed the contents of a credit report;
Percent of Sample: 17.9%.
Variable:
* Having experienced identity theft;
Percent of Sample: 9.8%.
Variable:
*The average age was 45.2 with a 17.3 standard deviation.
Source: GAO.
[End of table]
The Structure of the Model:
This multiple regression model is designed to estimate how various
independent variables influence the dependent variable -total knowledge
score (results are presented in table 14). Evaluating a multiple
regression requires considering both the influence of each independent
variable on the dependent variable (the total knowledge score, or the
percent correct answers) and the statistical significance of the
estimates.[Footnote 107] Total scores are cumulative and represent the
sum of the influences of all the independent variables in the model.
Much of the data puts consumers into categories (i.e. whether or not
one has completed high school or different income categories) and is
called categorical data. When categorical data is used for variables
there is always a base state and the variable information shows whether
the consumer is in or not in that base state. For example, in the case
of gender, being female is the base state. Thus being a male increases
the total knowledge score on average by about six-tenths of a
percentage point as compared to being female. Other binary variables
are constructed in a similar manner. Several categorical variables have
multiple states. For example, there are four states of education with
having less than a high school degree being the base state. In general,
a consumer who has completed high school will have about a 9 percentage
point increase in his total knowledge score as compared to a consumer
who never finished high school, while having completed college
increases the total knowledge score on average by about 14 percentage
points over the base case. Other multiple state variables are
constructed in a similar manner.
Table 14: A Regression Model to Explain Differences in Total Knowledge
Scores:
Dependent Variable = Total Knowledge Score (percent correct answers):
Independent Variable: Intercept;
Coefficient: 40.47%*.
Ethnic Classifications:
Independent Variable:
* White; Coefficient: Base state.
* Black; Coefficient: 15.
* Hispanic; Coefficient: -6.31*.
* Other; Coefficient: 1.31.
Independent Variable: Age of the individual (years):
Coefficient: -.09*.
Independent Variable: Household Income (per year);
* Below $25,000; Coefficient: Base state.
* $25,000 to $49,999; Coefficient: 3.11*.
* $50,000 to $74,999; Coefficient: 2.94*.
* $75,000 or above; Coefficient: 3.28*.
Independent Variable: Employment Status;
* Actively employed (full-or part-time employment); Coefficient: Base
state.
* Student; Coefficient: -3.33*.
* Not employed (homemakers, retired or unemployed); Coefficient: -1.56.
* Other; Coefficient: -3.13.
Independent Variable: Gender;
* Female; Coefficient: Base state.
* Male; Coefficient: .59.
Education Level:
Independent Variable:
* Less than high school;
Coefficient: Base state.
Independent Variable:
* High school completed;
Coefficient: 8.51*.
Independent Variable:
* Some postsecondary;
Coefficient: 11.49*.
Independent Variable:
* At least a college degree (Bachelor's degree or more);
Coefficient: 13.79*.
Financial Experience:
Independent Variable:
* Living in a free credit report state;
Coefficient: 1.38.
Independent Variable:
* Having had a mortgage or an automobile loan;
Coefficient: 3.73*.
Independent Variable:
* Having seen one's credit report;
Coefficient: 5.15*.
Independent Variable:
* Having seen one's credit score;
Coefficient: 4.24*.
Independent Variable:
* Having disputed the contents of a credit report;
Coefficient: 5.13*.
Independent Variable:
* Having experienced identity theft;
Coefficient: 1.14.
Source: GAO.
Note: R2 = .455. Observations in the analysis = 1380.
Items with a * are statistically significant at the alpha = .05 level.
[End of table]
The Statistical Results:
The model shows that the independent variables explain about 45 percent
of the variation in the total knowledge score as shown in the R2
statistic. Most of the independent variables in the model are
statistically significant, and their influences are
reasonable.[Footnote 108] For example, having a relatively high
household income or educational level increases the total knowledge
score as shown in table 14. For several variables, the influences are
statistically significant. If an effect is statistically significant we
can reject the null hypothesis that the variable in question has no
influence on the consumer's total knowledge score, i.e. the variable
does influence the total knowledge score. In our analysis, a
coefficient is considered statistically significant if the p-value for
the t-test is less than 0.05.
In the following discussion of the regression model, we usually address
the statistically significant coefficients. Statistically insignificant
coefficients are discussed only when they have important implications.
When discussing each variable we assume that all other variables are
being held constant. As a result, the coefficient of each independent
variable shows, on average, its independent influence on the total
knowledge score.
The regression shows that socio-economic factors influence scores. The
consumer's race/ethnicity can influence the total knowledge score. For
example, in the model Hispanics will score about 6 percentage points
lower than whites on average, holding all other variables constant. Age
also influences the score with each additional year lowering the score
by one-tenth of 1 percentage point.[Footnote 109] In addition, those
with incomes above $25,000 have higher total knowledge scores than
those earning less than $25,000 per year. For example, consumers
earning between $25,000 and $49,999 scored about 3 percentage points
higher than those earning less than $25,000. But only one employment
state variable is significant--being a student rather than working full
or part time. According to our model, students score about 3 percentage
points less on average than those employed full or part time. Finally,
knowledge increases with educational levels, so that those who finish
high school score about 9 percentage points higher on average than
those who have not.
We tested how living in a free credit report state, having experienced
identity theft, and experience with credit and the financial system can
influence total knowledge scores. Living in a free report state and
having experienced identity theft had no statistically significant
effect on the total knowledge scores. Having a mortgage or an
automobile loan increases a consumer's score by about 4 percentage
points on average. Similarly, having seen a credit report raises the
total knowledge score by about 5 percentage points. Having seen a
credit score increased the consumer's total knowledge score, on
average, by about 4 percentage points, while having disputed a credit
report increased the consumer's total knowledge score by about 5
percentage points. These increases in scores are consistent with a
belief that increased experience with credit and credit reporting
improves the consumer's knowledge of these issues.
Using the Regression to Illustrate the Simultaneous Influences of
Several Variables:
A multiple regression shows the simultaneous and cumulative influences
of all the independent variables on the total knowledge score. In
contrast, cross-tabulations show differences in average total knowledge
scores for different populations and do not allow the independent
influences of various factors to be separated. The average scores from
the cross-tabulations reflect the cumulative influences of all the
independent variables. For example, consumers from high-income
households are likely to also be highly educated and to have a car loan
or mortgage. Thus their average score is generally higher than scores
for those with lower household incomes because of the positive
influences of several variables.
To illustrate this tendency, we constructed two "straw" consumers and
used the statistical model to predict their total knowledge scores as
reported in table 15. In constructing our two "straw" consumers and
showing how their total knowledge scores vary based on differences in
the values of the independent variables, we are assuming that our model
is reasonable and representative of consumer behaviors.[Footnote 110]
Consumer 1 is a 25-year-old male Hispanic with a household income of
less than $25,000 a year, who does not have a high school degree, does
not have an automobile loan or mortgage, and has no experience with
credit reports or scores. Consumer 2 is also a 25-year-old male
Hispanic, but with at least a college degree, an automobile loan or
mortgage, and some experience with credit reports and scores. The model
allows us to determine the influences of the independent variables on
the total knowledge scores for each of these consumers. According to
the model, consumer 1 would score about 33 percent, while consume 2
would score about 63 percent. Changing the race/ethnicity of both
consumers to white raises the scores to about 39 percent and 69
percent, respectively. The differences in the scores reflect the
effects of a number of individual variables.
Table 15: Comparing Two Hypothetical Hispanics Using the Regression
Model:
Consumer 1: Intercept; 40.47%;
Consumer 2: Intercept; 40.47%.
Consumer 1: Hispanic; -6.31; Consumer 2: Hispanic; -6.31%.
Consumer 1: Age 25 (age times -.09); -2.25; Consumer 2: Age 25 (age
times -.09); -2.25%.
Consumer 1: Household income; less than $25,000; 0; Consumer 2:
Household income over $75,000; 3.28%.
Consumer 1: Employed; 0; Consumer 2: Employed; 0%.
Consumer 1: Male; .59; Consumer 2: Male; .59%.
Consumer 1: Less than high school education; 0; Consumer 2: At least a
Bachelor's degree; 13.79%.
Consumer 1: No auto loan or mortgage; 0; Consumer 2: Has an auto loan
or mortgage; 3.73%.
Consumer 1: Has not seen a credit report; 0; Consumer 2: Has seen a
credit report; 5.15%.
Consumer 1: Has not seen a credit score; 0; Consumer 2: Has seen a
credit score; 4.24%.
Consumer 1: Total Knowledge Score; 32.5; Consumer 2: Total Knowledge
Score; 62.69%.
Source: GAO.
[End of table]
[End of section]
Appendix VI: Web Appendix on Credit Reporting Literacy Survey Data:
This document presents survey data for all respondents, as well as
cross-tabulation results for demographic and other subpopulations of
respondents, and can be found on GAO's Web site at h [Hyperlink,
http://www.gao.gov/cgi-bin/getrpt?GAO-05-411SP.
[End of section]
Appendix VII: Comment from the Department of the Treasury:
DEPARTMENT OF THE TREASURY:
WASHINGTON, D.C. 20220:
March 14, 2005:
Ms. Debra Johnson:
Assistant Director, Financial Markets and Community Investment:
United States Government Accountability Office:
Washington, DC 20548:
Dear Ms. Johnson:
Thank you for the opportunity to review and provide comments for the
Government Accountability Office's proposed report entitled Credit
Reporting Literacy: Consumers Understood the Basics but Could Benefit
from Targeted Educational Efforts, GAO-05-223. The Department of the
Treasury is a leader in promoting financial education through its own
Office of Financial Education and through the Secretary of the
Treasury's leadership of the Financial Literacy and Education
Commission, and we greatly appreciate the opportunity to be involved in
this valuable study of consumer knowledge and behavior regarding credit
reporting.
The Office of Financial Education employs a multi-faceted approach in
its efforts to promote access to the financial education tools that can
help all Americans make wiser choices in all areas of personal
financial management, with a special emphasis on saving, credit
management, home ownership and retirement planning. First, through our
public outreach efforts, we are raising awareness of the importance of
financial education. Second, we have established standards for
effective financial education programs, and through a variety of means
we recognize programs that meet those standards. Third, through our
Technical Assistance Center we offer advice to those trying to
establish or improve financial education programs. Fourth, the Treasury
uses its contacts within the government and financial education
community to bring together those who need financial education with
those who provide it. And finally, the Treasury coordinates the federal
financial education effort by leading the Financial Literacy and
Education Commission.
Through the Office's efforts, we have become very familiar with the
issues facing consumers in today's increasingly complex financial
environment. As the GAO study has indicated, many consumers understand
basic issues of credit and have some sense of how their past behavior
may influence their ability to meet future financial needs. However, we
agree that most consumers would benefit from increased exposure to more
detailed information about credit reports in general and how their
actions can influence the contents of their own reports.
The information in this study will be invaluable to the Department of
the Treasury and the Financial Literacy and Education Commission in two
ways. First, the study will enable the Commission and other groups
following its lead to identify and penetrate particular demographic
groups which would benefit the most from financial education efforts.
The information will give us the ability to prioritize particular
segments of the population in our efforts. Second, the report will
allow us to understand which parts of the credit reporting system are
least understood. We can then channel our efforts to increasing
knowledge of those areas.
Since the survey portion of the GAO study was concluded in early
October 2004, the Commission has taken some important steps in its
effort to improve the financial literacy and education of persons in
the United States. That same month, the Commission successfully
launched the MyMoney.gov website and the 1-888-MyMoney toll free
hotline, both of which are available in English and Spanish. The
content available through the MyMoney resources provides information on
a wealth of financial topics including how to build a better credit
record, understanding consumer rights regarding credit and the credit
reporting process, credit scoring, and choosing a credit card. In the
first five months of operation, MyMoney.gov has recorded more than
100,000 hits and distributed more than 275,000 publications in English
and Spanish.
The Commission is also in the process of developing a national strategy
to promote basic financial literacy and education among all Americans.
The Department of Treasury, through a formal public comment process in
2004, has received input from more than one hundred fifty organizations
and individuals representing government, non-profit, private, and
academic sectors. Additionally, the Commission is holding six
information-gathering public meetings which will provide further input
to the national strategy. The GAO report will be a valuable tool in
identifying particular areas of emphasis for the national strategy and
will help guide future Commission efforts.
Again, we appreciate the opportunity to take part in this process and
welcome the valuable input provided by the study. In addition to the
comments contained herein, we are providing several technical
observations in an addendum to this correspondence. If I may be of
additional assistance or if you have any questions regarding this
matter, please contact me.
Sincerely,
Signed by:
Dan Iannicola, Jr.
Deputy Assistant Secretary:
Office of Financial Education:
[End of section]
Appendix VIII: Comments from the Federal Trade Commission:
UNITED STATES OF AMERICA:
FEDERAL TRADE COMMISSION:
WASHINGTON, D.C. 20580:
OFFICE OF THE CHAIRMAN:
March 4, 2005:
Ms. Yvonne D. Jones:
Acting Director:
Financial Markets and Community Investment:
U.S. Government Accountability Office:
441 G Street, N.W.:
Washington, DC 20548:
Dear Ms. Jones:
Thank you for the opportunity to comment on a draft of the Government
Accountability Office's survey report entitled Credit Reporting
Literacy: Consumers Understood the Basics but Could Benefit from
Targeted Educational Efforts (GAO-05-223). Consumer knowledge of a
topic as complex as credit reporting is difficult to measure, and we
commend the Government Accountability Office ("GAO") for having
undertaken a nationwide consumer survey.
We understand that the staff of the Federal Trade Commission ("FTC" or
"the Commission") provided technical comments on a draft of the report
to the GAO staff on February 28, 2005, and on March 1, GAO staff
responded informally to the FTC staff's suggestions. GAO has requested
the Commission's continent letter by March 4, 2005. Although we have
not seen a copy of the final report, we understand that it will reflect
the FTC staff's most significant comments. We would like to take this
opportunity, however, to reiterate some of those suggestions, respond
to the GAO's recommendations, and highlight some of the efforts that
the FTC already is undertaking to improve consumers' understanding of
the credit reporting process.
The draft report reviewed by our staff contains many useful findings.
Most notably the draft report found that a large percentage of
consumers surveyed have a good understanding of the basics of credit
reporting: they appear to know how credit reports are used, what
information is contained in a credit report, and that they have a right
to dispute inaccurate information. Moreover, those respondents who had
seen their own credit report had a better understanding of the credit
reporting process than those respondents who had never seen their
report.
Nonetheless, the draft report also concluded that there is significant
room for improvement in consumers' understanding of credit reporting.
For example, many respondents did not appear to understand that credit
reports are used to price and underwrite insurance policies. The draft
report also found that Hispanic respondents and respondents under the
age of 25 were less knowledgeable about the credit reporting process.
The draft recommended targeted educational initiatives. Specifically,
it recommended that the Commission educate consumers about its role in
enforcing the Fair Credit Reporting Act ("FCRA") and encourage
consumers "to understand how credit reports and scores are used," and
to obtain their credit report and credit score and dispute any
inaccurate information.
The Commission appreciates GAO's recommendations, and over the last
several years has been engaging in outreach efforts such as those
recommended by GAO. Moreover, the FTC has been targeting those efforts
to reach certain groups, such as Hispanic consumers and high school and
college students. The FTC has developed a portfolio of practical, plain-
language information in English and Spanish that is available for free
by mail, toll-free telephone, and the Web. These materials address
most, if not all, of the areas where GAO found that consumers could
benefit from more education. [NOTE 1] FTC outreach efforts include the
following:
* The FTC has more than 50 credit-related publications for consumers;
topics range from avoiding abusive lending practices to using secured
credit cards to understanding the protections granted by the Fair Debt
Collection Practices Act. Publications specific to credit reporting,
credit scores, and dispute rights include: Building a Better Credit
Record; Credit Scoring; How to Dispute Credit Report Errors; Credit and
Your Consumer Rights; and Ready, Set ... Credit.
* In fiscal year 2004, the FTC distributed more than four million print
copies of credit publications in English and Spanish; web accesses for
the same materials exceeded 5.7 million.
* The FTC distributes these materials through its own outreach efforts
and those of its partners, such as AARP, the Jump$tart Coalition for
Personal Financial Literacy, and the American Savings Education Council.
* The FTC has produced specialized publications to reach college-age
students who have limited experience with credit, such as Getting
Credit: What You Need to Know About Your Credit, and conducts outreach
on college campuses, at local school districts' college fairs, and in
high schools nationwide.
* The FTC distributes public service announcements and press kits, and
FTC officials regularly conduct interviews on credit reporting issues
with local and national radio, television, and print media, in both
English and Spanish.
The Commission also notes that FCRA section 609(c) requires that credit
bureaus include the Commission's "Summary of Rights" with every credit
report they provide to consumers. [NOTE 2] According to the draft
report, credit bureaus provided 57.4 million credit reports to
consumers in 2003. Each of these reports was required by law to be
accompanied by the summary of rights. The summary of rights explains
consumers' rights under the FCRA and provides contact information for
the FTC and other enforcement agencies. Although the draft report found
that many consumers do not know of certain specific dispute rights
(such as the right to add a statement to one's file if a dispute is not
satisfactorily resolved), the Commission believes that targeted
education like the statutorily required summary of rights is the best
means to address this issue. Consumers who obtain their credit reports
and find possible errors will have at hand the summary of rights and
the necessary information to conduct a dispute. The summary of rights
tells consumers what they need to know, when they need to know it.
In addition to highlighting the Commission's substantial consumer
education efforts and the statutorily required summary of rights, we
would like to reiterate three additional comments made by FTC staff to
GAO staff:
* The draft report reviewed by our staff found that few consumers
identified the FTC as the agency responsible for enforcing the FCRA and
recommended that the FTC publicize its enforcement role.[NOTE 3] As
noted above, we have been publicizing our enforcement authority through
the summary of rights and our outreach efforts. These targeted efforts
inform consumers of the FTC's enforcement role when it is most useful,
for example, when consumers request and review their credit reports.
[NOTE 4]
* The draft report found that 69 percent of respondents who had
disputed information to credit bureaus reported that the information
was removed from their report. See Survey Question 38. The purpose of
GAO's survey is to measure consumers' understanding of the credit
reporting process, and it is important to avoid confusing that purpose
with any conclusion about the actual accuracy of the underlying credit
reports. In 2003 testimony, GAO recognized the complexity of studying
credit report accuracy. [NOTE 5] Nevertheless, potential inaccuracies
remain an important reason for consumers to understand and review their
credit reports. Section 319 of the Fair and Accurate Credit
Transactions Act of 2003 directs the FTC to study credit report
accuracy for the next 10 years. We provided Congress with our first
interim report late last year. [NOTE 6]
* We understand that the final report will reflect most of FTC staff's
technical suggestions to GAO, including the FTC staff's comment with
respect to the draft report's finding that respondents did not
understand that resolving a dispute with one credit bureau would not
result in an automatic correction of the inaccuracy in every credit
bureau's file. See Survey Question 32. The Commission believes that
this finding should be qualified by noting that the FCRA requires that
information corrected by a consumer in the file of one of the three
national credit bureaus be automatically corrected with all three
bureaus. See FCRA § 623(b)(1)(D) (furnishers must correct inaccurate
information with each of the three national credit bureaus to which
they furnished the information); FCRA § 611(a)(5)(D) (the three
national credit bureaus must maintain an automated system that shares
the results of reinvestigations among all three bureaus).
The Commission appreciates the opportunity to review and comment on the
draft GAO report, and looks forward to working further with GAO as both
agencies work to fulfill their many obligations under the Fair and
Accurate Credit Transactions Act of 2003.
By direction of the Commission.
Signed by:
Deborah Platt Majoras:
Chairman:
NOTES:
[1] For example, GAO found that many consumers did not know that (1)
their credit score is affected by how much of their available credit
they use and (2) that credit reports are used to price insurance or
evaluate applicants for employment. Various FTC publications address
these topics. See, e.g., Credit Scoring ("If the amount you owe is
close to your credit limit, that is likely to have a negative effect on
your score"); Your Access to Free Credit Reports ("Inaccurate
information ... could affect your ability to get credit, insurance, or
even a job.")
[2] FCRA section 609(d), added by the Fair and Accurate Credit
Transactions Act of 2003, requires credit bureaus to send a Commission-
promulgated "Summary of Identity Theft Rights" to consumers who
complain of identity theft. This provision just recently became
effective.
[3] In response to an open-ended question, six percent of respondents
identified the FTC as the government agency to which they would
complain if they were not satisfied with a credit bureau's work, 24
percent identified the Better Business Bureau ("BBB"), eight percent
identified another agency, and the remaining respondents answered "do
not know." See Survey Question 33. Given that this is an open-ended
question asked of a general population of respondents, we do not
believe that this result shows that consumers who might wish to
complain about a credit bureau would be unaware of the FTC's role. It
is also worth noting that many other agencies, including BBBs and state
and local consumer protection agencies, accept and act upon individual
consumer complaints, while the FTC does not mediate individual
disputes. But see FCRA § 611 (e) (requiring the FTC to refer certain
consumer complaints to credit bureaus).
[4] The effectiveness of these efforts is demonstrated to some degree
by the fact that the number of complaints the Commission receives about
credit bureaus has increased over the past few years, both in absolute
terms and as a percentage of overall complaints. In 2002, we received
13,278 complaints about credit bureaus; in 2003, we received 19,223;
and in 2004, we received 23,831. Based on complaints received directly
by the FTC (excluding identity theft and do not call registry-related
complaints), credit bureaus accounted for 5.59% of overall complaints
in 2002, 6.98% in 2003, and 6.89% in 2004.
[5] See General Accounting Office, Consumer Credit: Limited Information
Exists on the Extent of Credit Report Errors and their Implications for
Consumers (GAO-03-1036T), at 17 (July 31, 2003).
[6] Federal Trade Commission, Report to Congress Under Sections 318 and
319 of the Fair and Accurate Credit Transactions Act of 2003, at 22-27
(December 2004).
The following are GAO's comments to FTC's letter dated March 4, 2005.
GAO Comments:
Referring to our finding that respondents did not understand that
resolving a dispute with one credit bureau would not result in
automatic correction of the inaccuracy in every credit bureau's file,
FTC suggested a clarification. FTC stated that FCRA "requires that
information corrected by a consumer in the file of one of the three
national credit bureaus be automatically corrected with all three
bureaus" (citing FCRA sections 623(b)(1)(D), 611(a)(5)(D). We believe
that FTC overstates the effect of these provisions.
First, except for information obtained fraudulently or through identity
theft, FCRA does not require a CRA to notify other CRAs that
information in its files is inaccurate or incomplete. Instead, the
process described by FTC depends upon furnishers to notify CRAs of
corrections to disputed information. Specifically, under section
623(b)(1)(D), if a furnisher receives notice from a CRA that
information the furnisher provided has been disputed by a consumer, the
furnisher must conduct an investigation and, if the furnisher finds the
disputed information to be incomplete or inaccurate, the furnisher must
report the finding to all nationwide CRAs to whom the furnisher
provided the information. Because of these provisions, a dispute with
one CRA generally should result in the correction of the disputed
information maintained by the other CRAs. However, this would occur
only if all of the CRAs obtained the disputed information from the same
furnisher. Section 623(b)(1)(D) does not address situations where CRAs
obtained the disputed information from different sources. Moreover, if
a furnisher has gone out of business, the process under section
623(b)(1)(D) would not ensure that erroneous information the furnisher
gave to multiple CRAs would be corrected where a consumer initiated a
dispute with only one CRA. Assuming that the CRA could not verify the
information, it would have to be deleted. The CRA is not required to
report the deletion to other CRAs. With respect to section
611(a)(5)(D), we note that the section requires CRAs to establish an
automated system that furnishers may use to report findings of
incomplete or inaccurate information to other CRAs, but it appears that
a furnishers' use of this system would be optional.
[End of section]
Appendix IX: GAO Contacts and Staff Acknowledgements:
GAO Contacts:
Yvonne D. Jones, (202) 512-8678;
Debra R. Johnson, (202) 512-9603:
Staff Acknowledgements:
In addition to those named above, Allison M. Abrams, Davi D'Agostino,
Jason A. Bromberg, Emily R. Chalmers, Marc W. Molino, Mona E. Nichols,
Jennifer R. Popovic, Mitchell B. Rachlis, Carl M. Ramirez, Terry L.
Richardson, Sidney H. Schwartz, Paul G. Thompson, and Wendy B.
Wierzbicki also made key contributions to this report.
We also acknowledge the GAO staff that provided invaluable assistance
translating our survey into Spanish, Roberto R. Piñero, Benjamin A.
Fenderlein, Josie H. Sigl, and Dan Garcia-Diaz.
Special thanks to Heather T. Dignan, Jack E. Edwards, Marion A.
Gatling, Richard J. Hillman, Harry Medina, George A. Scott, and Richard
M. Stana.
(250186):
FOOTNOTES
[1] Estimate developed by the Fair Isaac Corporation and printed on its
Web site on January 19, 2005, at www.myFico.com. These estimates are
based on an interest rate of 5.69 percent for someone with a credit
score of 720-850 and 9.29 percent for a consumer with a score of 500-
559. Credit scores are derived using a mathematical model that takes
into account the information contained in credit reports and attempt to
predict the likelihood that a person will repay a loan.
[2] GAO, Limited Information Exists on the Extent of Credit Report
Errors and Their Implications for Consumers, GAO-03-1036T (Washington,
D.C.: July 31, 2003).
[3] Companies that assemble consumer credit information and sell this
information are referred to as "consumer reporting agencies" by the
legislation governing credit reports. See Fair Credit Reporting Act, 12
U.S.C. §§ 1681-1681x, as amended (2004). These companies can also be
referred to as "credit bureaus," "credit reporting companies," or--as
they are in this report--"credit reporting agencies." Equifax,
Experian, and TransUnion Corporation are the nation's largest credit
reporting agencies.
[4] These figures were reported by the Consumer Data Industry
Association (CDIA), the trade association that represents the three
national CRAs. Available at www.cdiaonline.org/about.cfm.
[5] Pub. L. No. 91-508, codified as amended at 15 U.S.C. §§ 1681-1681x.
[6] Federal agencies and the states have authority to bring injunctive
actions under FCRA, as well as actions for damages for violations of
certain provisions. Under FCRA, as amended by the FACT Act, there are
also private rights of action. See 15 U.S.C. §§ 1681n, 1681o, 1681s-
2(c). In addition, federally supervised banks that use consumer reports
or furnish consumer report information are subject to enforcement by
their respective federal banking regulators. See 15 U.S.C. §1681s.
[7] Pub. L. No. 108-159 (Dec. 4, 2003).
[8] Members of FLEC include the Secretary of the Treasury, the head of
each federal banking agency and the National Credit Union Association
(NCUA); the Securities and Exchange Commission; the departments of
Education, Agriculture, Defense, Health and Human Services, Housing and
Urban Development, Labor, and Veterans Affairs; Federal Trade
Commission; the General Services Administration; the Small Business
Administration; the Social Security Administration; the Commodity
Futures Trading Commission; and the Office of Personnel Management. In
addition, the President may appoint five additional members.
[9] Pub. L. No. 108-159 § 517. This provision also mandated that GAO
provide recommendations for improving general financial literacy among
consumers, a provision addressed in "Highlights of a GAO Forum, The
Federal Government's Role in Improving Financial Literacy," GAO-05-93SP
(Washington, D.C.: Nov. 15, 2004) available at
http://www.gao.gov/docsearch/abstract.php?rptno=GAO-05-93SP. In
addition, the FACT Act directed GAO to conduct two other studies that
are currently ongoing: (1) an assessment of the effectiveness of the
provisions relating to information available to identity theft victims
that is due in June 2005 (Section 151), and (2) an assessment of the
effectiveness of FLEC in promoting financial literacy and education
that is due in December 2006 (Section 517 (a)).
[10] Except where noted, all percentage estimates have sampling errors
of +/-6 percentage points or less at the 95 percent confidence level.
In addition, except where noted, all numerical estimates other than
percentages have margins of error of +/-5 percent or less of the value
of those numerical estimates. In reporting responses to individual
questions, percentages may not equal 100 because of rounding. See
appendix II for more information on individual survey questions.
[11] Equifax is publicly traded and listed (NYSE: EFX). Experian is not
listed on a U.S. stock exchange, but its parent company is traded on
the London Stock Exchange and listed (LSE: GUS). TransUnion is
privately owned.
[12] For example, lenders use credit reports and scores to determine
what interest rates to charge consumers based on a perceived level of
risk in what is commonly called "risk-based pricing." See Pub. L. No.
108-159 § 311. Scores can also be used to determine requirements for
collateral, down payments, or cosigners.
[13] The Fair Isaac Corporation produces software used by many consumer
reporting agencies, including the three national CRAs, to produce FICO®
scores, which industry sources told us are commonly used in the United
States.
[14] The act specifically excluded medical information from this
provision. See Pub L. No. 91-508 § 609 (a)(i).
[15] Exceptions included credit transactions, life insurance policies,
job salaries that exceeded a specified dollar amount, and bankruptcy
data. Originally bankruptcy data older than 14 years was to be deleted.
In 1978, this period was shortened to 10 years. See Pub. L. No. 95-598.
[16] If used for employment purposes, CRAs were to identify the sources
and names of those receiving reports within the last 2 years of the
consumer's request; for other purposes, the time period was within the
last 6 months. See Pub. L. No. 91-508 § 609(a)(3).
[17] For employment purposes, this requirement for notification applied
to those receiving reports within the 2 years before the deletion or
notation of the dispute; for other purposes, the time period was 6
months before the deletion or notation. See Pub. L. No. 91-508 § 611
(d).
[18] GAO-03-1036T. Prior to 1996, FTC brought formal enforcement
actions involving procedures to ensure the accuracy of credit reports
against TransUnion in 1983, TRW (which would later become Experian) in
1991, and Equifax in 1995.
[19] Consumer Credit Reporting Reform Act of 1996, Pub. L. No. 104-208,
Div. A, Title II, Subtitle D.
[20] A number of studies reported persistent inaccuracies in credit
reports. For instance, GAO reported on the limited amount of
information available on the frequency, type, and cause of credit
report errors and their possible impact on consumers. See GAO-03-1036T.
[21] Under FTC regulations implementing section 211 of the FACT Act,
the requirement takes effect in a different geographic part of the
country each quarter, beginning on the west coast in December 2004 and
ending on the east coast in September 2005. See 69 Fed. Reg. 35468
(June 24, 2004).
[22] See 69 Fed. Reg. 35468, 35488 n. 66 (June 24, 2004). Throughout
this report, we refer to these states as "free report states."
[23] Before the FACT Act, FCRA gave CRAs most of the responsibility for
investigating disputes about inaccurate information and provided
consumers with certain protections, such as time limits on
investigations. However, under FCRA furnishers were required to follow
certain requirements. Among other things, if a furnisher determined
that information given to a CRA was inaccurate, the furnisher was
required to notify and provide corrections to the CRA. If a consumer
disputed information furnished to a CRA, the furnisher could not
provide the information to any CRA without sending notice of the
dispute.
[24] Pub. L. No. 108-159 § 312 (c). The federal banking agencies
include the Comptroller of the Currency, the Director of the Office of
Thrift Supervision, the Board of Governors of the Federal Reserve
System, and the Federal Deposit Insurance Corporation (FDIC).
[25] Commenting on a draft of this report, FTC said that in FY 2004, it
distributed more than 4 million print copies of credit publications in
English and Spanish. Consumers also accessed more than 5.7 million
copies using the Internet. In addition, the consumer information page
dealing with credit is consistently one of the most viewed pages on
FTC's Web site.
[26] Fair Isaac Corporation's Web site is at www.myFICO.com.
[27] GAO-05-93SP.
[28] Fox, Lynn, Joy Hoffmann, and Carolyn Welch. "Federal Reserve
Personal Financial Education Initiatives." Federal Reserve Bulletin,
Autumn 2002.
[29] Appendix II contains the entire survey instrument, including all
the survey questions and the scoring associated with each knowledge
question.
[30] Consumer Federation of America (CFA), "Consumers Lack Essential
Knowledge and Strongly Support New Protections on Credit Reporting and
Scores" (Washington, D.C.: July 2003). Unpublished. Information
available from www.consumerfed.org. We did not assess the quality of
the data that we obtained from this and other sources. We used these
data for comparative and background purposes only. Appendix III
presents additional information about this and other studies.
[31] Some consumers may have guessed at the correct multiple choice
option. In addition, when asked to provide their own explanation of a
CRA, some consumers may have found it difficult to form an articulate
response despite actually knowing what a CRA does. (See app. I for
further discussion of this and related issues.)
[32] CFA, "Consumers Lack Essential Knowledge."
[33] We counted the responses of consumers who said between $8 and $40
as correct. This range reflects a possible cost of a credit report
based on (1) prices the three nationwide CRAs were charging for credit
reports and credit report-related products at the time the survey was
scored, and (2) information provided to consumers about credit report
costs from other relevant organizations. Although the survey was
conducted before the FACT Act was implemented, we counted responses of
"free" as correct because of the new provision in the FACT Act that
allows consumers to receive one free credit report a year, upon
request. "Free in certain instances," was also counted as correct
because of provisions under FCRA. In addition, if a resident of
Minnesota said $3 or a resident of Connecticut said $5, these responses
were counted as correct under these states' laws.
[34] Under FCRA, before enactment of the FACT Act, consumers could
request and receive a free credit report, upon request, under certain
circumstances. These included (1) receiving notification of an adverse
action, such as a denial of credit that was based on their credit
report (within 60 days of notification), (2) suspecting that
information in a CRA's file was inaccurate due to fraud, such as
identity theft, (3) being unemployed and intending to apply for
employment within the next 60 days, (4) receiving public welfare
assistance, (5) receiving an adverse decision related to their
employment based at least in part on a credit report, and (6)
requesting an investigation that resulted in revisions to a report.
[35] Hilgert, Marianne A., Jeanne M. Hogarth, and Sondra G. Beverly,
"Household Financial Management: The Connection between Knowledge and
Behavior," Federal Reserve Bulletin, July 2003.
[36] Since bankruptcies can remain on reports for 10 years and other
negative information can remain for 7 years, we credited respondents
with a correct answer for providing either number of years.
[37] CFA and Providian®, "Most Consumers Do Not Understand Credit
Scores According to a New Comprehensive Survey," Washington, D.C., July
2004. Information at www.consumerfed.org. CFA had similar findings in
its July 2003 study.
[38] Hilgert, Hogarth, and Beverly, "Household Financial Management."
[39] CDIA collected this data from the three nationwide CRAs and
aggregated it before providing it to GAO in January 2005. The data on
credit files issued included all direct-to-consumer disclosures
(meaning offers to consumers, for a fee, online products that commonly
include file disclosures, credit scores, credit score analyses, file
monitoring, fraud alert systems, and more) and all consumer relations
disclosures (disclosures made to consumers who contacted a CRA to
receive a file disclosure due to an adverse action, concern about
fraud, being unemployed and seeking employment, or being on public
assistance). CDIA told us that the report request figure was an
estimate of total requests by all consumers and could include multiple
requests from the same consumer.
[40] CDIA presented data collected from the three nationwide CRAs on
the reasons why consumers ordered their credit reports before the
Senate Committee on Banking, Housing, and Urban Affairs on July 10,
2003. CDIA stated that of the 16 million reports that were being issued
by their members to consumers each year, 84 percent were provided due
to adverse action, 12 percent due to fraud, 5 percent out of curiosity,
and less than 1 percent because the consumer was unemployed, seeking
employment, or on public assistance. According to one CRA, these data
included only consumer relations disclosures and not direct-to-consumer
disclosures. The 57.4 million reports issued in 2003 included both
consumer relations disclosures and direct-to-consumer disclosures.
[41] Louis Harris and Associates, Consumer Interest in Free Credit
Reports, 2003. Unpublished. Selected findings from this study were
provided to GAO by CDIA.
[42] Of the remainder, about 10 percent did not identify the correct
definition of a credit score, and about 20 percent said they did not
know what one was.
[43] CFA and Providian, "Most Consumers Do Not Understand Credit
Scores."
[44] This range generally reflects the range used by firms that
generate credit scores. FICO credit scores are the standard credit
score used by the industry, and all three major CRAs own FICO models
and generate FICO credit scores. In addition, some of the CRAs have
developed their own credit scoring models and generate their own
scores.
[45] As reported on www.myFico.com. The importance of these factors may
differ with other non-FICO credit scores.
[46] According to Fair Isaac Corporation, its scoring model does not
penalize consumers for "rate shopping." For instance, a consumer who is
looking for a mortgage may have multiple lenders request his or her
credit score. In order to not penalize consumers who comparison shop,
the score counts multiple inquiries within a 14-day period as just one
inquiry.
[47] CFA and Providian, "Most Consumers Do Not Understand Credit
Scores."
[48] CFA and Providian, "Consumers' Lack Essential Knowledge."
[49] Hilgert, Hogarth, and Beverly, "Household Financial Management."
[50] Fair Isaac Corporation, myFICO Consumer Survey Results, March 2003
(unpublished); CFA and Providian, "Most Consumers Do Not Understand
Credit Scores."
[51] The 9.8 million disclosures represent only those scores sold by
the three CRAs. CDIA described this estimate as low and not indicative
of the total number of consumers who had reviewed their credit scores,
as it did not reflect the "entire score marketplace," which includes
those scores sold by Fair Isaac Corporation and by lenders such as
banks.
[52] Most of the 18 percent of consumers from our survey who had
disputed had seen their credit report, but it is not necessary to see a
report to dispute inaccurate information. Of the 18 percent that
disputed, 94 percent of those said that they had seen their report, and
6 percent said that they had not.
[53] As mentioned previously, FCRA, now amended by the FACT Act, has
mandated that the federal banking agencies, NCUA, and FTC issue
regulations identifying the circumstances under which furnishers are
required to investigate the accuracy of information in credit reports
upon the request of the consumer. Pub. L. No. 108-159 § 312 (c).
[54] CFA, "Consumers Lack Essential Knowledge."
[55] As amended by the FACT Act, FCRA requires FTC and other identified
federal agencies to promulgate regulations identifying when furnishers
are required to investigate disputed information. If the investigation
reveals an inaccuracy, the provision requires the furnisher to notify
each CRA to whom the furnisher provided the information. 15 U.S.C. §
1681s-2(a)(8), as amended. However, resolving inaccurate information
could still be problematic if a furnisher has gone out of business or
there is an error in personal identifying information, such as a name
or address.
[56] As mentioned previously, CDIA collected these data from the three
nationwide CRAs and aggregated them before providing the information to
GAO in January 2005. CDIA told GAO that the report request figure was
an estimate of total requests by all consumers and could include
multiple requests from the same consumer.
[57] For example, a 30-or 60-day delinquency remains on a credit report
even after it has been paid.
[58] Prepared testimony from the Consumer Data Industry Association
(CDIA) CEO and President Stuart Pratt; in U.S. Congress, Senate, The
Fair Credit Reporting Act and Issues Presented by Reauthorization of
the Expiring Preemption Provisions, hearings before the Committee on
Banking, Housing, and Urban Affairs, 108TH Cong., 1ST sess., July 10,
2003. Of the 27 percent of cases in which information was deleted, 11
percent was deleted on the furnisher's instruction and 16 percent
because the 30-day period for responding to a consumer complaint
expired with no response from the furnisher. Under section 611(a)(5) of
FCRA, as amended, information that cannot be verified must be deleted
from the credit file. 15 U.S.C. § 1681i.
[59] The sampling error for the 72 percent of consumers who said
information had not reappeared was +/-7 percentage points or less at
the 95 percent confidence level.
[60] We used two methodologies--a regression analysis and cross-
tabulation analysis--that generally provided consistent conclusions.
See appendix IV for cross-tabulations results and appendix V for the
regression analysis.
[61] The sampling error for the 55 percent figure was +/-7 percentage
points.
[62] Such effects are cumulative. For example, a Hispanic consumer with
a college degree, credit experience, and a high income could score
around 30 percent higher than a Hispanic without a high school
education, low income (less than $25,000) and with no credit
experience.
[63] The sampling error for the 65 percent of African Americans who had
ordered their reports themselves is +/-7 percentage points.
[64] The National Council of La Raza is a private, nonprofit
organization established to reduce poverty and discrimination and
improve life opportunities for Hispanic Americans.
[65] "Unbanked" means not maintaining deposit accounts with mainstream
financial service providers.
[66] La Raza discusses the need for financial literacy programs in the
Latino community as well as the results of some of these efforts in
their 2004 report, Financial Education in Latino Communities: An
Analysis of Programs, Products, and Results/Effects, available at
www.nclr.org.
[67] The sampling error for the 33 and 48 percent of those who obtained
their scores is +/-7 percentage points.
[68] The sampling error for the 18 percent of those aged 18 to 24 who
knew that their credit history could affect employment decisions is +/
-8 percentage points. The sampling error for those in all other age
groups who answered correctly is +/-7 percentage points or less.
[69] The sampling error for the 41 percent of those aged 18 to 24 who
knew that having a credit history for a short time affected scores
negatively is +/-10 percentage points; for the 61 percent of those aged
25 to 34 who answered correctly, it is +/-7 percentage points.
[70] The sampling errors for the 88 to 92 percent of consumers in other
age groups who said that they should check a report from time to time
are +/-7 percentage points or less. The sampling errors for the 29 to
37 percent for all other age groups that named the Internet are +/-7
percentage points or less, except for those aged 18 to 24 (+/-10
percentage points).
[71] We did not use adverse action in the regression analysis because
it is a subset of those who requested their credit reports. The
regression was designed to predict the knowledge score of the general
population, not the score of those who have obtained their credit
report.
[72] The sampling error for the 72 percent of victims of identity theft
who obtained reports and the 36 percent of those who had disputed is +/
-8 percentage points.
[73] FTC data indicates that complaints of identity theft doubled every
year from November 1999 through 2002. In 2003, the agency received
about 215,000 identity theft complaints; in 2004, about 247,000. The
total number of complaints in FTC's database as of December 31, 2004
was 742,300.
[74] The sampling error for the 78 percent of those who obtained
reports because of adverse action, and who ordered reports themselves
is +/-8 percentage points. The sampling error for the 52 percent of
those who obtained reports because of adverse action and disputed is +/
-10 percentage points.
[75] Congressional Research Service (CRS). "A Consumer's Access to a
Free Credit Report: A Legal and Economic Analysis," Washington, D.C.,
December 2003.
[76] In Colorado, a CRA is required to notify consumers of their right
to a free credit report if a CRA has received eight credit inquiries
pertaining to the consumer or if the CRA has received a report that
would add negative information to the consumer's file. C.R.S. § 12-14.3-
104(2)(a).
[77] By the term "nonminorities," we mean non-African Americans and non-
Hispanics.
[78] We used the definition of the American Association for Public
Opinion Research (AAPOR) response rate #3. See
http://www.aapor.org/pdfs/standarddefs2004.pdf for the formula used.
[79] We used AAPOR cooperation rate #1. See
http://www.aapor.org/pdfs/standarddefs2004.pdf for the formula used.
[80] Appendix II reproduces the entire survey instrument with the
frequency of responses and the scoring associated with each knowledge
question.
[81] The regression analysis shows the independent effects of certain
factors on an individual's total knowledge score. It can also show the
simultaneous and cumulative effects of certain factors if we add the
effects of each statistically significant factor. The cross-tabulation
analysis shows the differences in average total knowledge scores and
specific survey question responses for different populations. The
statistical significance of a result is the probability that the
observed relationship between variables did not occur by pure chance.
Please see appendix V for additional discussion of the regression
analysis.
[82] Prior to the Fair and Accurate Credit Transactions Act (FACT Act),
seven states--Colorado, Georgia, Maine, Maryland, Massachusetts, New
Jersey, and Vermont--already allowed residents to request at least one
free copy of their credit report annually. See 69 Fed. Reg. 35468,
35488 n. 66 (June 24, 2004). In our report, we refer to these states as
"free report states."
[83] The sampling errors for the 32-36 percent of other household
income groups are +/-7 percentage points or less.
[84] Such effects are cumulative. For example, a Hispanic consumer with
a college degree, credit experience, and a high income could score
around 30 percentage points higher than a low-income ($25,000 or less)
Hispanic without a high school education or credit experience.
[85] The sampling error for the 65 percent of African Americans that
ordered their reports themselves is +/-7 percentage points.
[86] "Unbanked" populations do not maintain deposit accounts with
mainstream financial service providers.
[87] La Raza discusses financial literacy programs in the Latino
community as well as the results of some of these efforts in its 2004
report, Financial Education in Latino Communities: An Analysis of
Programs, Products, and Results/Effects, available at www.nclr.org.
[88] The small independent impact found for age with the regression
analysis and the larger differences among age groups found with the
cross-tabulation analysis suggest that the cross-tabulation analysis
captured not only age but also factors such as education and experience
with the credit system. See appendix V.
[89] The sampling error for the 24 percent of the youngest consumers
(aged 18 to 24) who had seen their reports is +/-9 percentage points.
The sampling errors for the oldest group (65 or over) and all other age
groups are +/-7 percentage points or less.
[90] The sampling errors for the 33 and 48 percent of those who had
obtained their scores are +/-7 percentage points.
[91] The sampling error for the 27 percent of those in the 55 to 64 age
group who said they had disputed is +/-7 percentage points.
[92] The sampling error for the 18 percent of those aged 18 to 24 who
knew that their credit history could affect employment decisions is +/
-8 percentage points. The sampling errors for all other age groups that
knew this fact are +/-7 percentage points or less.
[93] The sampling error for the 55 percent of those aged 18 to 24 who
could identify the correct definition of a credit score is +/-10
percentage points.
[94] The sampling error for the 41 percent of those aged 18 to 24 who
knew that having had a credit history for a short time affected scores
negatively is +/-10 percentage points. For the 61 percent of those aged
25 to 34 who knew this fact, the error is +/-7 percentage points.
[95] The sampling errors for the 88 to 92 percent of those in other age
groups who said they should check their report from time to time are +/
-7 percentage points or less.
[96] The sampling errors for the 29 to 37 percent of those in all other
age groups who named the Internet are +/-7 percentage points or less,
except for the 37 percent of those aged 18 to 24 (+/-10 percentage
points).
[97] Once again, the differences in the results of the cross-tabulation
and regression analyses reflect the cumulative effect of independent
factors. For example, age as well as employment status may affect the
results for students and retirees. See appendix V.
[98] The sampling error for the 84 percent of students who were unable
to name any of the CRAs is +/-12 percentage points.
[99] We did not use adverse action in the regression analysis because
it is a subset of those who requested their credit reports. The
regression was designed to predict the knowledge score of the general
population, not the score of those who had obtained their credit
report.
[100] The sampling error for the 72 percent of victims of identity
theft who had obtained reports is +/-8 percentage points.
[101] The sampling error for the 48 percent of identity theft victims
who had had information removed after a dispute is +/-14 percentage
points. The sampling error for the 74 percent of those who had not been
victims and had had information removed is +/-7 percentage points.
[102] The sampling error for the 53 percent of identity theft victims
who knew this fact was +/-9 percentage points.
[103] The sampling error for the 78 percent of those who obtained
reports because of adverse action and who ordered their reports
themselves is +/-8 percentage points.
[104] In Colorado, a CRA is required to notify consumers of their right
to a free credit report if a CRA has received 8 credit inquiries
pertaining to the consumer or the agency has received a report that
would add negative information to the consumer's file. C.R.S. § 12-14.3-
104(2)(a).
[105] The sampling error for the 38 percent of females and 21 percent
of males who said that they added a written statement to their report
after a dispute is +/-8 percentage points.
[106] The influence or statistical impact measures how much the value
of the dependent variable will change for a change in the value of a
specific independent variable. It is a statistical relationship and not
causal. The coefficient of the independent variable in the regression
model is the measure of influence of that variable.
[107] An effect is statistically significant if the value of a test
statistic used to test it provides strong evidence that the observed
effect did not occur by pure chance. Effects can be statistically
significant at different levels, for example, a result can be
statistically significant at a 5 percent level. The smaller the level
at which a result is statistically significant, the stronger the
evidence is that the observed effect did not occur by pure chance.
[108] Other factors not collected, and thus not studied, could have
affected the total knowledge score if they had been available. For
simplicity, our equation was linear. Using a more complex non-linear
functional form could affect both the explanatory power of the equation
and the coefficients on the independent variables. In our judgment,
variations in the functional form would increase the complexity of our
analysis without improving the ability of the equation to explain
variations in the total knowledge scores.
[109] This finding is consistent with the idea that older consumers are
less knowledgeable or familiar with modern credit reports and credit
scoring approaches to loan applications. We also tested to see if there
was a non-linear relationship between age and total knowledge scores to
see if both the younger and older consumers were less knowledgeable.
This tested non-linear relationship did not improve the model so we
stayed with the simpler linear model.
[110] In practice, our model is one of many models that could be used
to reasonably represent the relationship between total knowledge scores
and the independent variables. If we had chosen additional independent
variables or a different functional form, we could have produced other
reasonable models. However, given the data available to us and our
desire to keep a simple functional form with reasonable statistical
properties we chose to report this model and use it to evaluate how the
total knowledge score of our "straw" consumers varied.
GAO's Mission:
The Government Accountability Office, the investigative arm of
Congress, exists to support Congress in meeting its constitutional
responsibilities and to help improve the performance and accountability
of the federal government for the American people. GAO examines the use
of public funds; evaluates federal programs and policies; and provides
analyses, recommendations, and other assistance to help Congress make
informed oversight, policy, and funding decisions. GAO's commitment to
good government is reflected in its core values of accountability,
integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains
abstracts and full-text files of current reports and testimony and an
expanding archive of older products. The Web site features a search
engine to help you locate documents using key words and phrases. You
can print these documents in their entirety, including charts and other
graphics.
Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as "Today's Reports," on its
Web site daily. The list contains links to the full-text document
files. To have GAO e-mail this list to you every afternoon, go to
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order
GAO Products" heading.
Order by Mail or Phone:
The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or
more copies mailed to a single address are discounted 25 percent.
Orders should be sent to:
U.S. Government Accountability Office
441 G Street NW, Room LM
Washington, D.C. 20548:
To order by Phone:
Voice: (202) 512-6000:
TDD: (202) 512-2537:
Fax: (202) 512-6061:
To Report Fraud, Waste, and Abuse in Federal Programs:
Contact:
Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Automated answering system: (800) 424-5454 or (202) 512-7470:
Public Affairs:
Jeff Nelligan, managing director,
NelliganJ@gao.gov
(202) 512-4800
U.S. Government Accountability Office,
441 G Street NW, Room 7149
Washington, D.C. 20548: