This is the accessible text file for GAO report number GAO-07-708 
entitled 'Federal Housing Administration: Modernization Proposals Would 
Have Program and Budge Implications and Require Continued Improvements 
in Risk Management' which was released on July 2, 2007.

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 Requesters:

United States Government Accountability Office:

GAO:

June 2007:

Federal Housing Administration:

Modernization Proposals Would Have Program and Budget Implications and 
Require Continued Improvements in Risk Management:

GAO-07-708:

GAO Highlights:

Highlights of GAO-07-708, a report to congressional requesters

Why GAO Did This Study:

In recent years, the Federal Housing Administration (FHA) has 
experienced a sharp decline in market share. Also, the agency has 
estimated that, absent program changes, its Mutual Mortgage Insurance 
Fund (Fund) would require appropriations in 2008. To adapt to market 
changes, FHA has implemented new procedures and proposed the following 
major legislative changes: raising FHA’s loan limits, allowing risk-
based pricing, and lowering down-payment requirements. GAO was asked to 
report on (1) the likely program and budget impacts of FHA’s 
modernization efforts; (2) the tools, resources, and risk-management 
practices important to FHA’s implementation of the legislative 
proposals, if passed; and (3) other options that FHA and Congress could 
consider to help FHA adapt to market changes. To address these 
objectives, GAO analyzed FHA and Home Mortgage Disclosure Act (HMDA) 
data and interviewed officials from FHA and other mortgage institutions.

What GAO Found:

FHA’s recent changes to insurance approval and appraisal requirements 
have streamlined its insurance process, and FHA’s major legislative 
proposals could affect the demand for FHA’s loans, the cost and 
availability of insurance to borrowers, and the insurance program’s 
budgetary costs. Based on GAO’s analysis of HMDA data, the number of 
FHA-insured loans could have been from 9 to 10 percent greater in 2005 
had the higher, proposed mortgage limits been in effect. GAO’s analysis 
of data on 2005 FHA home purchase borrowers shows that 43 percent would 
have paid the same or less under the risk-based pricing proposal than 
they actually paid, 37 percent would have paid more, and 20 percent 
(those with the highest expected claim rates) would not have qualified 
for FHA insurance. While to be viewed with caution, FHA has made 
estimates indicating that the loans it expects to insure in 2008 would 
result in negative subsidies (i.e., net cash inflows) of $342 million 
if the major legislative changes were enacted, rather than requiring an 
appropriation of $143 million absent any program changes. 

FHA has taken or planned steps to enhance tools and resources and adopt 
risk-management practices important to implementing the legislative 
proposals, but does not intend to use a common industry practice, 
piloting, to mitigate the risks of any zero-down-payment product it is 
authorized to offer. In response to prior GAO recommendations, FHA has 
taken steps to improve the loan performance and scoring models it would 
use in risk-based pricing. It also has identified minor changes to its 
information systems and staff increases needed to implement the 
proposals but faces long-term challenges in these areas. Additionally, 
the legislative proposals would introduce new risks. The proposal to 
lower down-payment requirements is of particular concern given the 
higher default rates on these loans and the difficulty of setting 
prices for new products whose risks may not be well known. GAO has 
previously indicated that Congress may want to consider requiring FHA 
to limit the initial availability of any new products and also 
recommended that FHA itself consider piloting. However, FHA has 
indicated that it does not plan to pilot any no-down-payment product it 
might offer.

Mortgage industry participants and researchers have suggested more 
options that Congress and FHA could consider to help FHA adapt to 
changes in the mortgage market, but some changes could have budget 
impacts and complicate oversight efforts. Some administrative 
changes—such as implementing a more limited form of risk-based 
pricing—are within FHA’s existing authority. Congress also could grant 
FHA additional authority that would allow it to invest the Fund’s 
current resources in information technology and human capital, but this 
would increase the federal government’s budget deficit. Finally, 
Congress could contemplate other approaches to the provision of federal 
mortgage insurance, such as creating a government corporation. However, 
any fundamental changes to how the federal government provides mortgage 
insurance could require new oversight mechanisms and would require 
careful deliberation.

What GAO Recommends:

While making no new recommendations, GAO re-emphasizes the need for 
continued management attention to prior GAO recommendations that could 
help address risks and challenges associated with the legislative 
proposals. HUD commented that the report’s concerns about FHA’s risk 
management and emphasis on the need for piloting new products were 
unwarranted.

[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-07-708].

To view the full product, including the scope and methodology, click on 
the link above. For more information, contact William B. Shear at (202) 
512-8678 or shearw@gao.gov

[End of section]

Contents:

Letter:

Results in Brief:

Background:

Modernization Efforts Have Streamlined FHA Processes and Likely Would 
Affect Program Participation and Costs:

FHA Has Enhanced Tools and Resources Important to Implementing 
Proposals but Does Not Intend to Mitigate Risks by Piloting New 
Products:

Congress and FHA Could Consider Other Administrative and Legislative 
Changes to Help FHA Adapt to Changes in the Mortgage Market:

Observations:

Agency Comments and Our Evaluation:

Appendix I: Objectives, Scope, and Methodology:

Appendix II: Comments from the Department of Housing and Urban 
Development:

Appendix III: GAO Contact and Staff Acknowledgments:

Figures:

Figure 1: Proposed Changes to FHA's Loan Limits:

Figure 2: Impact of Borrower Credit Scores and LTV Ratios on Insurance 
Premiums under FHA's Risk-Based Pricing Proposal:

Figure 3: Impact of FHA's Risk-Based Pricing Proposal on Borrowers' 
Premiums, Including First-Time and Low-Income Homebuyers:

Figure 4: Impact of FHA's Risk-Based Pricing Proposal on Premiums Paid 
by Different Racial Groups:

Figure 5: Impact of FHA's Risk-Based Pricing Proposal on Racial 
Distribution of FHA Borrowers:

Abbreviations:

CBSA: core based statistical area:

FHA: Federal Housing Administration:

FTE: full-time equivalent:

HECM: Home Equity Conversion Mortgage:

HUD: Department of Housing and Urban Development:

HMDA: Home Mortgage Disclosure Act:

IRS: Internal Revenue Service:

LTV: loan-to-value:

MHC: Millennial Housing Commission:

United States Government Accountability Office:

Washington, DC 20548:

June 29, 2007:

The Honorable Richard C. Shelby: 
Ranking Member: 
Committee on Banking, Housing, and Urban Affairs: 
United States Senate:

The Honorable Wayne Allard: 
United States Senate:

The Department of Housing and Urban Development (HUD), through its 
Federal Housing Administration (FHA), has helped millions of families 
purchase homes by providing insurance for single-family home mortgages 
made by private lenders. However, FHA's single-family insurance program 
has faced several challenges in recent years, including rising 
delinquency rates and a sharp decline in the number of participating 
borrowers, due partly to increased competition from conventional 
mortgage providers.[Footnote 1] As conventional providers have improved 
their ability to evaluate risk, FHA has begun to experience adverse 
selection--that is, conventional providers have identified and approved 
relatively lower-risk borrowers in FHA's traditional market segment, 
leaving relatively higher-risk borrowers for FHA. Furthermore, the 
agency has estimated that, absent any program changes, the program 
would for the first time operate with a positive subsidy in fiscal year 
2008--meaning that the present value of estimated cash outflows (such 
as insurance claims) to FHA's Mutual Mortgage Insurance Fund (Fund) 
would exceed the present value of estimated cash inflows (such as 
borrower premiums). To avoid a positive subsidy in fiscal year 2008, 
FHA estimates that it would have to increase slightly the insurance 
premiums charged to borrowers.

To adapt to market changes, FHA has implemented new administrative 
procedures and proposed legislation designed to modernize its insurance 
processes and products. FHA's recent administrative changes include 
allowing higher-performing lenders to endorse, or approve, loans for 
FHA insurance without prior review by FHA and adopting conventional 
market appraisal requirements. The legislative proposals would, among 
other things, raise FHA's mortgage limits, give the agency flexibility 
to set insurance premiums based on the credit risk of borrowers, and 
reduce down-payment requirements from the current 3 percent to 
potentially zero. However, as we testified in June 2006, weaknesses in 
FHA's risk management raise questions about the agency's ability to 
successfully implement the proposed legislation.[Footnote 2] Given 
these concerns, you asked us to evaluate FHA's modernization efforts. 
Specifically, this report discusses (1) the likely program and 
budgetary impacts of FHA's modernization efforts; (2) the tools, 
resources, and risk-management practices important to FHA's 
implementation of the legislative proposals, if passed; and (3) other 
options that FHA and Congress could consider to help FHA adapt to 
changes in the mortgage market and the pros and cons of these options.

To determine the likely program and budgetary impacts of FHA's 
modernization efforts, we analyzed data collected under the Home 
Mortgage Disclosure Act (HMDA) and from FHA's Single Family Data 
Warehouse (SFDW).[Footnote 3] Specifically, we used 2005 HMDA data (the 
most current available) to examine the effect of raising loan limits on 
demand for FHA-insured loans. We determined the number of additional 
loans in different geographic areas that would have been eligible for 
FHA insurance under the revised loan limits and, based on FHA's current 
market share, estimated the percentage of those loans that FHA might 
have insured. We estimated the effects of risk-based pricing on 
borrowers' eligibility for FHA insurance and the premiums they would 
pay by analyzing SFDW data on FHA's 2005 home purchase borrowers to 
determine the characteristics of borrowers that could fall into FHA's 
proposed pricing categories. We reviewed recent administrative changes 
made by FHA and interviewed FHA officials, several FHA lenders, and 
mortgage and real estate industry groups about their effects on loan 
and insurance processing times and costs. We also examined the 
potential budgetary impacts of the legislative proposals by reviewing 
the President's fiscal year 2008 budget and FHA's cost estimates. To 
evaluate the tools, resources, and risk-management practices important 
to FHA's implementation of the proposals, we relied on our prior work, 
reviewed information provided by FHA, and interviewed officials from 
FHA, private mortgage insurers, and the government-sponsored 
enterprises (GSE) Fannie Mae and Freddie Mac.[Footnote 4] To determine 
other options that FHA and Congress could consider, we reviewed 
relevant literature and interviewed FHA officials, academic experts, 
FHA lenders, and private mortgage insurers. Appendix I contains 
additional information on our scope and methodology. We conducted this 
work in Washington, D.C., from September 2006 to June 2007 in 
accordance with generally accepted government auditing standards.

Results in Brief:

FHA's modernization efforts, which include completed administrative and 
proposed legislative changes, have streamlined the agency's insurance 
process and could affect the demand for FHA-insured loans, the cost and 
availability of insurance to borrowers, and the budgetary costs of the 
insurance program. In 2006, FHA made several administrative changes, 
such as allowing higher-performing lenders to approve FHA insurance 
without prior review by FHA and simplifying its appraisal process. FHA 
and mortgage industry officials with whom we spoke said that these 
changes have increased the efficiency of loan and insurance processing, 
making FHA products more attractive and, therefore, more likely to be 
used. For example, one FHA lender reported a 35 percent decrease in 
loan processing times and a 25 percent reduction in operating costs for 
its FHA business. In addition to these administrative changes, FHA has 
proposed legislation that would grant the agency new flexibilities 
intended to help address challenges, such as adverse selection, 
resulting from innovations and increased competition in the mortgage 
market. If passed, the legislative changes likely would affect borrower 
participation in the program and the program's budgetary costs. Based 
on our analysis of 2005 HMDA data, we estimate that the number of FHA- 
insured loans in 2005 could have been from 9 to 10 percent greater had 
the higher, proposed mortgage limits been in effect. Although the 
effect of introducing risk-based premiums on the demand for FHA-insured 
loans is especially difficult to estimate, risk-based pricing would 
affect the cost and availability of FHA insurance. Specifically, risk- 
based pricing would decrease premiums for lower-risk borrowers and 
increase them for higher-risk borrowers. Our analysis of data for FHA 
home purchase borrowers in 2005 shows that about 43 percent of those 
borrowers would have paid the same or less than they actually paid, 37 
percent would have paid more, and 20 percent would not have qualified 
for FHA insurance based on FHA's plans as of May 2007. (The same 
percentages hold true when risk-based pricing is compared with the 
higher across-the-board premiums that FHA estimates it would have to 
charge to avoid a positive subsidy in fiscal year 2008 absent any 
program changes.) The 20 percent who would not have qualified were 
borrowers with expected lifetime claim rates more than 2.5 times 
greater than the average claim rate. The legislative proposals also 
would have a budgetary impact, mostly reflected in subsidy costs. While 
to be viewed with caution, FHA has made estimates indicating that the 
loans it expects to insure in 2008 would result in negative subsidies 
of $342 million if the major legislative changes were enacted, rather 
than requiring an appropriation of $143 million absent any program 
changes.

FHA has taken or planned steps to enhance the tools and resources 
important to implementing the legislative proposals--and help address 
risks and challenges associated with the proposals--but does not intend 
to use a common industry practice, piloting, to mitigate the risks of 
any zero-down-payment product it is authorized to offer. To implement 
its risk-based pricing proposal, FHA would rely on statistical models 
that estimate the performance of loans and its mortgage scorecard, an 
automated tool that evaluates the default risk of borrowers. In 
response to our prior recommendations, FHA has improved the forecasting 
ability of its loan performance models by incorporating additional 
variables found to influence credit risk and is in the process of 
addressing a number of limitations in its mortgage scorecard that could 
reduce its effectiveness for risk-based pricing. For instance, as we 
reported in April 2006, the scorecard does not include a number of 
important variables included in other mortgage institutions' 
scorecards, such as the source of the down payment.[Footnote 5] FHA 
also has identified changes in information systems needed to implement 
the legislative proposals and has obligated or requested a total of $11 
million for this purpose. To address human capital needs, the 
President's fiscal year 2008 budget requests 21 additional staff for 
FHA to help analyze industry trends, align the agency's business 
processes with current mortgage industry practices, and promote new FHA 
products. Although FHA has taken actions to enhance key tools and 
resources, it operates in a highly competitive environment in which 
other market participants have greater flexibility to hire and 
compensate staff and invest in information technology, which enhances 
their ability to adapt to market changes. Additionally, the legislative 
proposals would introduce new risks and challenges. The proposal to 
lower down-payment requirements is of particular concern given the 
greater default risk of low-down-payment loans, housing market 
conditions that could put borrowers with such loans in a negative 
equity position, and the difficulty of setting prices for new products 
whose risks may not be well understood. FHA plans to take some steps, 
such as instituting stricter underwriting standards, to mitigate these 
risks and challenges. However, while other mortgage institutions use 
pilot programs to manage the risks associated with changing or 
expanding their product lines, FHA has indicated that it does not plan 
to pilot any zero-down-payment product it is authorized to offer and 
lacks the resources to do so. We have previously reported that Congress 
may want to consider requiring FHA to limit the initial availability of 
any new products and also recommended that FHA itself consider piloting.

Mortgage industry participants and researchers have suggested a number 
of additional administrative and legislative options that Congress and 
FHA could consider to help FHA adapt to changes in the mortgage market, 
but some changes could have budget impacts and complicate oversight 
efforts. Some administrative changes--such as adjusting premiums or 
even implementing a more limited form of risk-based pricing--are within 
FHA's existing authority. Congress also could consider granting FHA 
additional authorities that would increase the agency's operational 
flexibility. For example, Congress could allow FHA to invest the Fund's 
current resources--that is, negative subsidies that accrue in the 
Fund's reserves--in information technology and human capital. However, 
using the Fund's current resources would diminish its ability to 
withstand severe economic conditions and would also increase the 
federal government's budget deficit, all other things being equal. 
Additionally, Congress could expressly authorize FHA to offer and pilot 
new insurance products without prior congressional approval. Finally, 
Congress could consider various alternative approaches to the provision 
of federal mortgage insurance. For example, the federal government 
could continue to provide mortgage insurance but through a more 
independent government corporation, implement risk-sharing arrangements 
with private partners, or let market forces determine the future need 
for federal mortgage insurance by making no program changes and 
allowing FHA's role in the mortgage market to increase or decrease 
according to market conditions. However, any fundamental changes to how 
the federal government provides mortgage insurance also could require 
new oversight mechanisms and would require careful deliberation.

While our report does not make any new recommendations, we make 
observations about the need for careful implementation of the 
legislative proposals, if passed. While FHA has performed considerable 
analysis to support its legislative proposals and has made or planned 
enhancements to many of the specific tools and resources that would be 
important to its implementation, the proposals present risks and 
challenges and should be viewed with caution. Continued management 
attention to our prior recommendations, including piloting new products 
and steps to improve its mortgage scorecard, could help FHA address 
these risks.

We provided HUD with a draft of this report. HUD commented that the 
draft report provided a balanced assessment but also that the report's 
concerns about FHA's risk management and emphasis on the need for 
piloting zero-or lower-down-payment products were unwarranted. HUD 
indicated that it had a firm basis for anticipating the performance of 
these products as a result of its experience with loans with down- 
payment assistance from nonprofit organizations funded by home sellers. 
While we acknowledge that this experience could inform assessment of 
how a zero-down-payment product would perform, the product could be 
utilized by a different population of borrowers than seller-funded down-
payment assistance loans and may not perform similarly to these loans. 
Also, if authorized to offer a zero-down-payment product in the near 
future, FHA would be introducing it at a time when stagnating or 
declining home prices in some parts of the country could increase the 
risk of default. Because of these risks and uncertainties, we continue 
to believe that a prudent way to introduce a zero-down-payment product 
would be to limit its initial availability such as through a pilot 
program. We discuss HUD's comments in the agency comments section, and 
the agency's written comments are reproduced in appendix II.

Background:

Congress established FHA in 1934 under the National Housing Act (P.L. 
73-479) to broaden homeownership, protect and sustain lending 
institutions, and stimulate employment in the building industry. FHA 
insures a variety of mortgages for initial home purchases, construction 
and rehabilitation, and refinancing. In fiscal year 2006, FHA insured 
almost 426,000 mortgages representing $55 billion in mortgage 
insurance. FHA's single-family programs insure private lenders against 
losses from borrower defaults on mortgages that meet FHA criteria for 
properties with one to four housing units. FHA has played a 
particularly large role among minority, lower-income, and first-time 
homebuyers and generally is thought to promote stability in the market 
by ensuring the availability of mortgage credit in areas that may be 
underserved by the private sector or are experiencing economic 
downturns. In fiscal year 2006, 79 percent of FHA-insured home purchase 
loans went to first-time homebuyers, 31 percent of whom were minorities.

FHA is a government mortgage insurer in a market that also includes 
private insurers. Generally, borrowers are required to purchase 
mortgage insurance when the loan-to-value (LTV) ratio--the ratio of the 
amount of the mortgage loan to the value of the home--exceeds 80 
percent. Private mortgage insurance policies provide lenders coverage 
on a portion (generally 20 to 30 percent) of the mortgage balance. 
However, borrowers who have difficulty meeting down-payment and credit 
requirements for conventional loans may find it easier to qualify for a 
loan with FHA insurance, which covers 100 percent of the value of the 
loan. Because the credit risk is mitigated by the federal guaranty, FHA 
borrowers are allowed to make very low down payments and generally pay 
interest rates that are competitive with prime mortgages.

FHA Insurance Requirements:

Legislation sets certain standards for FHA-insured loans. FHA-insured 
borrowers are required to make a cash investment of a minimum of 3 
percent. This investment may come from the borrowers' own funds or from 
certain third-party sources. However, borrowers are permitted to 
finance their mortgage insurance premiums and some closing costs, which 
can create an effective LTV ratio of close to 100 percent for some FHA- 
insured loans. Congress also has set limits on the size of the loans 
that may be insured by FHA. These limits vary by county. The limit for 
an FHA-insured mortgage is 95 percent of the local median home price, 
not to exceed 87 percent or fall below 48 percent of the Freddie Mac 
conforming loan limit, which was $417,000 in 2006. Therefore, in 2006, 
FHA loan limits fell between a floor in low-cost areas of $200,160 and 
a ceiling in high-cost areas of $362,790. Eighty-two percent of 
counties nationwide had loan limits set at the low-cost floor, while 3 
percent had limits set at the high-cost ceiling. The remaining 15 
percent of counties had limits set between the floor and ceiling, at 95 
percent of their local median home prices.

FHA's Mutual Mortgage Insurance Fund:

FHA insures most of its single-family mortgages under its Mutual 
Mortgage Insurance Fund, which is supported by borrowers' insurance 
premiums. FHA has the authority to establish and collect a single up- 
front premium in an amount not to exceed 2.25 percent of the amount of 
the original insured principal obligation of the mortgage, and annual 
premiums of up to 0.5 percent of the remaining insured principal 
balance, or 0.55 percent for borrowers with down payments of less than 
5 percent. Currently, FHA uses a flat premium structure whereby all 
borrowers pay the same 1.5 percent up-front fee and a 0.5 percent 
annual fee.

The Omnibus Budget Reconciliation Act of 1990 requires an annual 
independent actuarial review of the economic net worth and soundness of 
the Fund. The actuarial review estimates the economic value of the Fund 
as well as the capital ratio to see if the Fund has met the capital 
standards in the act.[Footnote 6] The analysis considers the historical 
performance of the existing loans in the Fund, projected future 
economic conditions, loss given claim rates, and projected mortgage 
originations. The Fund has met the capital ratio requirements since 
1995, and the single-family mortgage insurance program has maintained a 
negative overall credit subsidy rate, meaning that the present value of 
estimated cash inflows from premiums and recoveries exceeds estimated 
cash outflows for claim payments (excluding administrative costs). 
However, in recent years, the subsidy rate has approached zero.

A few single-family mortgage insurance programs are insured as 
obligations under either the General Insurance or Special Risk 
Insurance Funds. These programs are Section 203(k) rehabilitation 
mortgages, which enable borrowers to finance both the purchase (or 
refinancing) of a house and the cost of its rehabilitation through a 
single mortgage; Section 234(c) insurance for the purchase of a unit in 
a condominium building; and reverse mortgages under the Home Equity 
Conversion Mortgage (HECM) program, which can be used by homeowners age 
62 and older to convert the equity in their home into a lump sum 
payment, monthly streams of income, or a line of credit to be repaid 
when they no longer occupy the home.

Trends in the Mortgage Market and Their Impact on FHA:

Two major trends in the conventional mortgage market have significantly 
affected FHA. First, in recent years, members of the conventional 
mortgage market increasingly have been active in supporting low-and no- 
down-payment mortgages, increasing consumer choices for borrowers who 
may have previously chosen an FHA-insured loan. Subprime lenders, in 
particular, have offered mortgage products featuring flexible payment 
and interest options that allowed borrowers to qualify for mortgages 
despite a rise in home prices.[Footnote 7] Second, to help assess the 
default risk of borrowers, particularly those with high LTV ratios, the 
mortgage industry increasingly has used mortgage scoring and automated 
underwriting systems. Underwriting refers to a risk analysis that uses 
information collected during the origination process to decide whether 
to approve a loan, and automated underwriting refers to the process by 
which lenders enter information on potential borrowers into electronic 
systems that contain an evaluative formula, or algorithm, called a 
scorecard. The scorecard algorithm attempts to measure the borrower's 
risk of default quickly and objectively by examining data such as 
application information and credit scores. (Credit scores assign a 
numeric value generally ranging from 300 to 850 to a borrower's credit 
history, with higher values signifying better credit.) The scorecard 
compares these data with specific underwriting criteria (e.g., cash 
reserves and credit requirements) to predict the likelihood of default. 
Since 2004, FHA has used its own scorecard called Technology Open to 
Approved Lenders (TOTAL). FHA lenders now use TOTAL in conjunction with 
automated underwriting systems to determine the likelihood of default. 
Although TOTAL can determine the credit risk of a borrower, it does not 
reject a loan. FHA requires lenders to manually underwrite loans that 
are not accepted by TOTAL to determine if the loan should be accepted 
or rejected.

Further, as we noted in a recent report, the share of home purchase 
mortgage loans insured by FHA has fallen dramatically, from 19 percent 
in 1996 to 6 percent in 2005, with almost all the decline occurring 
since 2001.[Footnote 8] The combination of (1) FHA product restrictions 
and a lack of process improvements relative to the conventional market 
and (2) product innovations and expanded loan origination and funding 
channels in the conventional market--coupled with interest rate and 
house price changes--provided conditions that favored conventional 
mortgages over FHA products. Conventional subprime loans, in 
particular, emerged as an alternative to FHA-insured mortgages but 
often at a higher ultimate cost to certain borrowers.

At the same time, FHA's financial performance has worsened. As we noted 
in a recent testimony, one reason for deteriorating loan performance 
has been the increase in FHA-insured loans with down-payment assistance 
from nonprofit organizations funded by home sellers.[Footnote 9] Down- 
payment assistance programs provide cash assistance to homebuyers who 
cannot afford to make the minimum down payment or pay the closing costs 
involved in obtaining a mortgage. From 2000 to 2006, the total 
proportion of FHA-insured home purchase loans with down-payment 
assistance from nonprofits (the large majority of which received 
funding from property sellers) increased from about 2 percent to 
approximately 33 percent.

Legislative Proposals for FHA Modernization:

To help FHA adapt to recent trends in the mortgage market, in 2006 HUD 
submitted a legislative proposal to Congress that included changes that 
would adjust loan limits for the single-family mortgage insurance 
program, eliminate the requirement for a minimum down payment, and 
provide greater flexibility to FHA to set insurance premiums based on 
risk factors. HUD's proposal, as it currently stands, reflects 
revisions made by the Expanding American Homeownership Act of 2006, 
which was passed by the House of Representatives in July 2006. 
Specifically, as shown in figure 1, the proposal would increase the 
loan limit for FHA-insured mortgages from 95 to 100 percent of the 
local median home price. It would also raise the loan limit floor in 
low-cost areas from 48 to 65 percent of the conforming loan limit, and 
the ceiling in high-cost areas from 87 to 100 percent of the conforming 
limit.[Footnote 10] The proposal would also repeal the 3 percent 
minimum cash investment requirement and allow FHA to set premiums 
commensurate with the risk of the loan[Footnote 11]. FHA would 
establish a premium structure allowing either a combination of upfront 
and annual premiums or annual premiums alone, subject to specified 
maximum amounts.

Figure 1: Proposed Changes to FHA's Loan Limits:

[See PDF for image]

Source: GAO, HUD. 

[End of figure]

In addition to these three major changes, the modernization proposal 
also contained other provisions, including:

* Permanently eliminating the limit on the number of HECM (reverse) 
mortgages that can be insured, setting a single nationwide loan limit 
for HECMs, and authorizing a HECM program for home purchases.[Footnote 
12]

* Extending the permissible term of FHA-insured mortgages from 35 to 40 
years.

* Moving HECMs, Section 203(k) rehabilitation mortgages, and Section 
234(c) condominium unit mortgages from the General Insurance and 
Special Risk Insurance Funds to the Mutual Mortgage Insurance Fund. 
Moving the condominium program to the Fund would simplify the 
origination and underwriting process for these loans because they would 
no longer be subject to more complex requirements for multifamily 
housing loans.

While FHA's planning has reflected revisions made to its original 
proposal by the House of Representatives in the 109th Congress, new 
bills introduced in the 110th Congress could further affect FHA's 
planning.[Footnote 13]

Modernization Efforts Have Streamlined FHA Processes and Likely Would 
Affect Program Participation and Costs:

FHA's modernization efforts, which include completed administrative and 
proposed legislative changes, have streamlined the agency's insurance 
processes and likely would affect program participation and costs. 
According to FHA and mortgage industry officials with whom we spoke, 
FHA's recent administrative changes have resulted in efficiency 
improvements, making FHA products more attractive to use. FHA's 
proposed legislation would grant the agency new leeway to help address 
challenges, such as adverse selection, resulting from innovations and 
increased competition in the mortgage market. If passed, the 
legislative changes likely would have a number of program and budgetary 
impacts. For example, we estimate that raising the FHA loan limits 
could increase demand for FHA-insured loans, all other things being 
equal. The risk-based pricing proposal would decrease premiums for 
lower-risk borrowers, increase them for higher-risk borrowers, and 
disqualify other potential borrowers. In addition, FHA estimates that 
the legislative proposals would have a favorable budgetary impact.

Mortgage Industry Officials Report That FHA's Recent Administrative 
Changes Have Increased the Efficiency of Loan and Insurance Processing:

FHA has taken a number of steps to make the loans it insures easier to 
process and bring the agency more in line with the conventional market. 
For example, in January 2006, FHA introduced the Lender Insurance 
Program, which enables higher-performing lenders to endorse all FHA 
loans except HECMs without a prior review by FHA.[Footnote 14] Prior to 
that time, all lenders were required to mail loan case files to FHA for 
review by contract staff before the loan could be endorsed for 
insurance. If the contractor found a problem with the case file, FHA 
would mail the file back to the lender for correction. Under the new 
program, approved lenders are allowed to perform their own pre- 
endorsement reviews and submit loan data electronically to 
FHA.[Footnote 15] If the loan data pass checks for accuracy and 
completeness, the lender is able to endorse the loan automatically. As 
of December 31, 2006, 405 (31 percent) of the 1,314 FHA lenders 
eligible for the program had been approved to participate. Between 
January 1, 2006, and December 31, 2006, 46 percent of FHA-insured loans 
were endorsed through the program.

In addition to implementing the Lender Insurance Program, FHA revised 
its appraisal protocols and closing cost guidelines to align them more 
closely with conventional standards. Specifically, the agency 
simplified the appraisal process by adopting appraisal forms used in 
the conventional market and eliminating the requirement that minor 
property deficiencies be corrected prior to the sale of the property. 
Under the revised procedures, FHA limits required repairs to those 
necessary to protect the health and safety of the occupants, protect 
the security of the property, or correct physical deficiencies or 
conditions affecting structural integrity. Examples of property 
conditions that must be repaired include inadequate access to the 
exterior of the home from bedrooms, leaking roofs, and foundation 
damage. The agency requires the appraiser to identify minor property 
deficiencies (such as missing handrails, cracked window glass, and 
minor plumbing leaks) on the appraisal form, but no longer stipulates 
that they be repaired. These changes went into effect for all 
appraisals performed on or after January 1, 2006. In January 2006, FHA 
also eliminated its list of "allowable" and "non-allowable" closing 
costs and other fees that may be collected from the borrower. The 
agency made this change because FHA lenders had advised the agency that 
home sellers sometimes balked at accepting a sales contract from a 
homebuyer wishing to use FHA-insured financing because its guidelines 
differed from standard practice and did not consider regional 
variations. Lenders may now charge and collect from borrowers those 
customary and reasonable costs necessary to close the mortgage.

According to FHA lenders and industry groups, these changes have 
increased the efficiency of loan processing, making FHA products more 
attractive to use. Representatives of a mortgage industry group told us 
that feedback from the group's members on the Lender Insurance Program 
had been positive.[Footnote 16] Similarly, the FHA lenders we 
interviewed stated that the program had resulted in efficiency 
improvements, such as reduced processing times or costs. For example, 
one large FHA lender estimated that participating in the program had 
reduced the time it took to process an FHA-insured loan by about 35 
percent (or 15 to 20 days). The same FHA lender also estimated that 
participation in the program had reduced the operating costs (mostly 
printing and shipping costs) for its FHA business by about 25 percent. 
Additionally, the FHA lenders we interviewed and representatives of a 
real estate industry group noted that FHA's revised appraisal protocols 
and closing costs had made it easier to originate FHA loans. 
Representatives of the industry group noted that the revisions had 
shortened the time it took to close an FHA loan, which was important in 
a competitive market. Finally, the lenders we interviewed estimated 
that the administrative changes had contributed, at least in part, to 
recent modest increases in the number of FHA-insured loans they had 
made.

According to FHA officials, the Lender Insurance Program also has 
reduced the time it takes FHA to process insurance endorsements and led 
to cost savings. They estimated that it takes FHA from 2 to 3 days to 
endorse applications for insurance on loans that are not part of the 
program. For loans endorsed through the program, they noted that 
approval is virtually instantaneous if the loan passes quality checks. 
In addition to reducing insurance processing times, the program has 
resulted in cost savings for FHA. During the first year of the program, 
FHA realized a reduction in contracting costs of more than $2 million, 
as its contractors were required to perform fewer pre-endorsement 
reviews. FHA also saved more than $70,000 in mailing costs during the 
first 9 months of the program. FHA estimates that contract costs will 
continue to decline as the program is expanded to include the HECM 
program.

Raising Loan Limits Likely Would Increase Demand for FHA Loans, but the 
Effect of Other Major Proposals on FHA Loan Volume Is Uncertain:

Our analysis indicates that raising FHA's loan limits likely would 
increase the number of loans insured by FHA by making more loans 
eligible for FHA insurance. In some areas of the country, particularly 
in parts of California and the Northeast, median home prices have been 
well above FHA's maximum loan limits, reducing the agency's ability to 
serve borrowers in those markets. For example, the 2005 loan limit in 
high-cost areas was $312,895 for one-unit properties, while the median 
home price was about $399,000 in Boston, Massachusetts; about $432,000 
in Newark, New Jersey; $500,000 in Salinas, California; and about 
$646,000 in San Francisco, California. If the limits were increased, 
FHA insurance would be available to a greater number of potential 
borrowers. Our analysis of HMDA data indicates that the agency could 
have insured from 9 to 10 percent more loans in 2005 had the higher 
mortgage limits been in place.[Footnote 17] The greatest portion of 
this increase resulted from raising the loan limit floor in low-cost 
areas from 48 to 65 percent of the conforming loan limit. In 
particular, 82 percent of the new loans that would have been insured by 
FHA and 74 percent of the dollar amount of those loans in our analysis 
occurred in areas where the loan limits were set at the floor. Only 14 
percent of the new loans (22 percent of the dollar amount of new loans) 
would have resulted from increasing the loan limit ceiling. Our 
analysis also found that the average size of an FHA-insured loan in 
2005 would have increased from approximately $123,000 to about $132,000 
had the higher loan limits been in place.

The effect of the other major legislative proposals on the demand for 
FHA-insured loans is difficult to estimate. Although FHA has not 
estimated the effect on demand, FHA officials expect that risk-based 
pricing would enable them to serve more borrowers. By reducing premiums 
for relatively lower-risk borrowers, FHA expects to attract more of 
these borrowers. However, increased premiums for higher-risk borrowers 
could reduce these borrowers' demand for FHA products. Additionally, 
some high-risk borrowers who previously would have qualified for FHA 
insurance would not qualify under risk-based pricing. The effect of 
lowering down-payment requirements on demand for FHA-insured loans is 
also difficult to estimate. FHA expects a new zero-down-payment product 
to attract borrowers who otherwise would have used down-payment 
assistance from nonprofit organizations funded by home sellers. 
However, underwriting restrictions could limit the number of borrowers 
who would qualify for the product.

Developments in the subprime market also may affect the demand for FHA 
loans. Since 2001, FHA's share of the mortgage market has declined as 
the subprime market has grown. However, relatively high default and 
foreclosure rates for subprime loans and a contraction of this market 
segment could shift market share to FHA. For example, one major lender 
we interviewed said that FHA's continued modernization efforts combined 
with a weakening subprime market likely would result in renewed demand 
for FHA products as simplified processes make it easier for lenders to 
originate FHA-insured loans.

Risk-Based Pricing Could Help Address Adverse Selection but Would 
Affect the Cost and Availability of FHA Insurance for Some Borrowers:

To help address the problem of adverse selection, FHA has sought 
authority to price insurance premiums based on borrower risk, which 
would affect the cost and availability of FHA insurance for some 
borrowers. Currently, all FHA-insured borrowers pay an up-front premium 
of 1.5 percent of the original insured loan amount, and annual premiums 
of 0.5 percent of the remaining insured principal balance. Under this 
flat pricing structure, lower-risk borrowers subsidize higher-risk 
borrowers. In recent years, innovations in the mortgage market have 
allowed conventional mortgage lenders and insurers to identify and 
approve relatively low-risk borrowers and charge fees based on default 
risk. As relatively lower-risk borrowers in FHA's traditional market 
segment have selected conventional financing, FHA has been left with 
more high-risk borrowers who require a subsidy and fewer low-risk 
borrowers to provide that subsidy.

Partly due to this trend, the President's fiscal year 2008 budget 
stated that, in the absence of risk-based pricing, FHA would need to 
raise premiums to avoid the need for a positive subsidy. FHA officials 
told us that they would have to raise premiums for all borrowers to 
1.66 percent up front and 0.55 percent annually. Raising premiums for 
all borrowers could exacerbate FHA's adverse selection problem by 
causing even more lower-risk borrowers to opt for more competitive 
conventional products rather than FHA-insured loans, leaving FHA with 
even fewer lower-risk borrowers to subsidize higher-risk borrowers. 
Rather than raise premiums for all borrowers, FHA has proposed risk- 
based pricing as a solution to the adverse selection problem. Under 
risk-based pricing, some future FHA borrowers would pay more than the 
current premiums while others would pay about the same or less. As 
previously noted, discounting premiums could make FHA a more attractive 
option for relatively lower-risk borrowers.

As of May 2007, FHA's risk-based pricing proposal established six 
different risk categories, each with a different premium rate, for 
purchase and refinance loans.[Footnote 18] FHA used data from its most 
recent actuarial review to establish the six risk categories and 
corresponding premiums based on the relative performance of loans with 
various combinations of LTV ratio and credit score. Borrowers in 
categories with higher expected lifetime claim rates would have higher 
premiums than those in categories with lower claim rates. Premiums 
would range from 0.75 percent up front and 0.50 percent annually for 
the lowest-risk borrowers, to 3.00 percent up front and 0.75 percent 
annually for the highest-risk borrowers. Although the premiums that FHA 
would charge borrowers in the six risk categories would be more 
commensurate with the risks of the loans, lower-risk borrowers would 
continue to subsidize higher-risk borrowers to some extent.

If FHA were granted the authority to implement its risk-based pricing 
proposal, the agency would publish a pricing matrix that would allow 
borrowers to identify their likely premiums based on their credit 
scores and LTV ratios. As shown in figure 2, lower borrower credit 
scores and higher LTV ratios would result in higher insurance premiums. 
However, FHA would use its TOTAL mortgage scorecard to make the final 
determination of a borrower's placement in a particular risk category. 
While TOTAL takes into account more borrower and loan characteristics 
than LTV ratio and credit score (such as borrower reserves and payment- 
to-income ratio), it was designed to predict the probability of claims 
or defaults that would later result in claims within 4 years of loan 
origination rather than lifetime claim rates. Therefore, FHA rescaled 
the TOTAL scores to reflect lifetime claim rates. Because of the 
additional risk characteristics considered by TOTAL, a borrower's TOTAL 
score could indicate that a borrower belongs in a higher risk category 
than would be suggested by LTV ratio and credit score alone. FHA has 
not produced a formal estimate of how often this would occur, but plans 
to include this caveat in its pricing matrix.

Figure 2: Impact of Borrower Credit Scores and LTV Ratios on Insurance 
Premiums under FHA's Risk-Based Pricing Proposal:

[See PDF for image]

Source: GAO, FHA.

[End of figure]

Our analysis of how the proposed pricing structure would affect home 
purchase borrowers similar to those insured by FHA in 2005 found that 
approximately 43 percent of borrowers would have paid the same or less 
while 37 percent would have paid more. As discussed more fully later, 
20 percent would not have qualified for FHA insurance had the risk- 
based pricing proposal been in effect. These percentages hold true 
whether comparing the proposed risk-based premiums to the current 
premiums of 1.5 percent up front and 0.5 percent annually or the higher 
premiums of 1.66 percent up front and 0.55 percent annually that, 
according to FHA, would be needed to maintain a negative subsidy rate 
in fiscal year 2008. As shown in figure 3, risk-based pricing would 
have had a similar impact on first-time and low-income homebuyers FHA 
served in 2005.

Figure 3: Impact of FHA's Risk-Based Pricing Proposal on Borrowers' 
Premiums, Including First-Time and Low-Income Homebuyers:

[See PDF for image]

Source: GAO, SFDW.

Note: We analyzed SFDW data on 2005 home purchase borrowers. The figure 
shows how these borrowers would have fared under FHA's risk-based 
pricing proposal. Low-income homebuyers are those whose incomes are 
less than or equal to 80 percent of the area median income. The figure 
excludes the approximately 2 percent of borrowers for whom SFDW did not 
contain either an LTV ratio or credit score (the two variables FHA 
would use to determine risk-based premiums).

[End of figure]

Among FHA's 2005 borrowers, 47 percent of white borrowers and 40 
percent of Hispanic borrowers would have paid the same or less under 
the new proposed risk-based pricing structure than they did under the 
present pricing structure, while 28 percent of black borrowers would 
have paid the same or less. A little more than one-third of borrowers 
in each racial category would have paid more (see fig. 4). FHA 
officials concluded, in their analysis of an earlier version of the 
risk-based pricing proposal, that any disparate impacts of risk-based 
pricing using consumer credit scores would be based on valid business 
reasons. Specifically, they noted that, although some racial 
differences do exist in the distribution of credit scores and LTV 
ratios, these variables are strongly associated with claim rates and 
have become the primary risk factors used for pricing credit risk in 
the conventional market.

Figure 4: Impact of FHA's Risk-Based Pricing Proposal on Premiums Paid 
by Different Racial Groups:

[See PDF for image]

Source: GAO, SFDW. 

Note: We analyzed SFDW data on 2005 home purchase borrowers. The figure 
shows how these borrowers would have fared under FHA's risk-based 
pricing proposal. It excludes the 2 percent of borrowers for whom SFDW 
did not contain either an LTV ratio or credit score (the two variables 
FHA would use to determine risk-based premiums) and the 2.9 percent of 
borrowers for whom race was not disclosed. Percentages do not add to 
100 due to rounding.

[End of figure]

Risk-based pricing would also affect the availability of FHA insurance 
for some borrowers. Approximately 20 percent of FHA's 2005 borrowers 
would not have qualified for FHA mortgage insurance under the 
parameters of the risk-based pricing proposal we evaluated. FHA 
determined that the expected claim rates of these borrowers were higher 
than it found tolerable for either the borrower or the Fund. Those 
borrowers who would not have qualified had high LTV ratios and low 
credit scores. Their average credit score was 584, and their expected 
lifetime claim rates are more than 2.5 times higher than the average 
claim rate of all FHA loans.[Footnote 19] FHA officials stated that 
setting risk-based premiums for potential future FHA borrowers with 
similar characteristics would require prices higher than borrowers may 
be able to afford.

The overall distribution of 2005 FHA borrowers (by income, first-time 
borrower status, or race) would not have changed substantially had the 
policy not to serve borrowers with these higher expected lifetime claim 
rates been in place that year (all other things being equal). If the 20 
percent of borrowers with the higher expected claim rates were removed 
from FHA's 2005 borrower pool, our analysis found that low-income 
homebuyers would have remained about 51 percent of the pool. First-time 
homebuyers would have constituted about 78 percent of the pool, 
compared with 79 percent when all borrowers are included. Similarly, 
the overall racial distribution of borrowers would have changed 
modestly (see fig. 5). The percentage of Hispanic borrowers would have 
remained about 14 percent, black borrowers would have decreased from 13 
to 11 percent, and white borrowers would have increased from 69 to 70 
percent.

Figure 5: Impact of FHA's Risk-Based Pricing Proposal on Racial 
Distribution of FHA Borrowers:

[See PDF for image]

Source: GAO, SFDW.

Note: We analyzed SFDW data on 2005 home purchase borrowers. The figure 
shows how these borrowers would have fared under FHA's risk-based 
pricing proposal. Percentages do not add to 100 due to rounding.

[End of figure]

All other things being equal, implementing the legislative proposals 
likely would have had a slightly negative impact on FHA's ability to 
meet certain performance measures related to the types of borrowers it 
serves. HUD's strategic plan for fiscal years 2006 to 2011 calls for 
the share of first-time minority homebuyers among FHA home purchase 
mortgages to remain above 35 percent. Our analysis shows that 34 
percent of fiscal year 2005 home purchase mortgages were for first-time 
minority home buyers. Under risk-based pricing, a slightly lower 
percentage, 32 percent, would have been first-time minority home 
buyers. The strategic plan also calls for the share of FHA-insured home 
purchase mortgages for first-time homebuyers to remain above 71 
percent. Our analysis shows that 79 percent of fiscal year 2005 FHA 
home purchase borrowers were first-time home buyers. Under risk-based 
pricing, 77 percent would have been first-time home buyers.

Legislative Proposals Likely Would Have a Beneficial Budgetary Impact:

According to FHA's estimates, the three major legislative proposals 
would have a beneficial impact on HUD's budget due to higher estimated 
negative subsidies. According to the President's fiscal year 2008 
budget, the credit subsidy rate for the Fund would be more favorable if 
the legislative proposals were enacted. Absent any program changes, FHA 
estimates that the Fund would require an appropriation of credit 
subsidy budget authority of approximately $143 million. If the 
legislative proposals were not enacted, FHA would consider raising 
premiums to avoid the need for appropriations. If the major legislative 
proposals were passed, FHA estimates that the Fund would generate $342 
million in negative subsidies.[Footnote 20]

FHA's subsidy estimates for fiscal year 2008 should be viewed with 
caution given that FHA has generally underestimated the subsidy costs 
for the Fund. To meet federal requirements, FHA annually reestimates 
subsidy costs for each loan cohort dating back to fiscal year 
1992.[Footnote 21] The current reestimated subsidy costs for all except 
the fiscal year 1992 and 1993 cohorts are higher than the original 
estimates. For example, the current reestimated cost for the fiscal 
year 2006 cohort is about $800 million higher than originally 
estimated. As discussed more fully later in this report, FHA has taken 
some steps to improve its subsidy estimates.

FHA Has Enhanced Tools and Resources Important to Implementing 
Proposals but Does Not Intend to Mitigate Risks by Piloting New 
Products:

FHA has enhanced the tools and resources it uses that would be 
important to implementing the legislative proposals, but has not always 
used industry practices that could help the agency manage the risks 
associated with program changes. To implement risk-based pricing, FHA 
would rely on historical loan-level data, models that estimate loan 
performance, and its TOTAL mortgage scorecard. Although FHA has 
improved the forecasting ability of its models by adding variables 
found to influence credit risk, the agency is still addressing 
limitations in TOTAL that could reduce its effectiveness as a pricing 
tool. FHA also has identified changes in information systems needed to 
implement the legislative proposals and requested additional staff to 
help promote new FHA products but faces long-term challenges in these 
areas. However, the legislative proposals would introduce new risks and 
challenges such as the difficulty of pricing loans with very low or no 
down payments whose risks may not be well understood. While other 
mortgage institutions use pilot programs to manage the risks associated 
with changing or expanding their product lines, FHA has indicated that 
it does not plan to pilot any no-down-payment product it is authorized 
to offer.

Credit Score Information Has Enhanced the Data FHA Would Use to 
Implement Proposals:

Mortgage institutions use detailed information on the characteristics 
and performance of past loans to help predict the performance of future 
loans and price them correctly. Like other mortgage institutions we 
contacted, FHA has extensive loan-level data. These data are contained 
in the agency's SFDW, which FHA implemented in 1996 to assemble 
critical data from 12 single-family systems.[Footnote 22] SFDW is 
updated monthly and currently contains data on approximately 33 million 
FHA-insured loans dating back to fiscal year 1975. These data include 
information on the borrower (such as age, gender, race, income, and 
first-time home buyer status) and the loan (including whether it is an 
adjustable-or fixed-rate mortgage, the source and amount of any down- 
payment assistance, interest rate, premium rate, original mortgage 
amount, and LTV ratio).

FHA has added information on borrower credit scores to the loan-level 
data that it plans to use to assess risk and set insurance premiums if 
the legislative proposals were enacted. Research has shown that credit 
scores are a strong predictor of loan performance--that is, borrowers 
with higher scores experience lower levels of default. FHA started 
collecting credit score data in the late 1990s when it began allowing 
its lenders to use automated underwriting systems and mortgage 
scorecards. Upon approving the use of Fannie Mae and Freddie Mac's 
mortgage scorecards in fiscal year 1998, FHA began receiving credit 
score data for loans underwritten using these scoring tools. To develop 
its own mortgage scorecard, FHA purchased archived credit scoring data 
for loan origination samples dating back to 1992. Since implementing 
its TOTAL mortgage scorecard in May 2004, FHA has collected credit 
scores on almost all FHA borrowers.

FHA Has Made Some Improvements to Key Statistical Models, but 
Additional Challenges Remain:

FHA would rely on both its loan performance models and TOTAL mortgage 
scorecard to set insurance premiums if authorized to implement risk- 
based pricing. Although FHA has improved the forecasting ability of its 
loan performance models by incorporating additional variables found to 
influence credit risk, FHA is still in the process of addressing a 
number of limitations in TOTAL that could reduce its effectiveness for 
risk-based pricing. The agency's actuarial review contractor developed 
the loan performance models to estimate the economic value of the Fund 
for the annual actuarial review. The models estimate lifetime claim and 
prepayment (the payment of a loan before its maturity date) rates based 
on factors such as origination year, age, interest rate, mortgage 
product type, initial LTV ratio, and loan amount. FHA used the 
projected lifetime claim and prepayment rates from the most recent 
actuarial review as the basis for its proposed risk-based insurance 
premiums.[Footnote 23]

FHA has improved its loan performance models by adding factors that 
have been found to influence credit risk. In September 2005, we 
reported that FHA's subsidy reestimates, which use data from FHA's loan 
performance models, reflect a consistent underestimation of the costs 
of its single-family insurance program. We recommended that FHA study 
and report the impact (on the forecasting ability of its loan 
performance models) of variables that have been found in other studies 
to influence credit risk, such as payment-to-income ratios, credit 
scores, and the presence of down-payment assistance.[Footnote 24] In 
response, HUD indicated that its contractor was considering the 
specific variables that we had recommended FHA include in its annual 
actuarial review of the Fund. The contractor subsequently incorporated 
the source of down-payment assistance in the fiscal year 2005 actuarial 
review and borrower credit scores in the fiscal year 2006 review.

FHA also intends to use TOTAL to determine risk-based premiums, but we 
have identified weaknesses in the scorecard that could limit its 
effectiveness as a pricing tool. As previously noted, FHA plans to use 
TOTAL to make the final determination regarding premium rates if 
authorized to implement risk-based pricing. However, we reported in 
April 2006 that TOTAL excludes a number of important variables included 
in other mortgage scoring systems.[Footnote 25] For example, TOTAL does 
not distinguish between adjustable-and fixed-rate mortgages. However, 
adjustable-rate mortgages generally are considered to be higher risk 
than otherwise comparable fixed-rate mortgages because borrowers are 
subject to higher payments if interest rates rise. Unlike the mortgage 
scorecards of other institutions, TOTAL also does not include an 
indicator for property type (single-family detached homes or 
condominiums, for example).[Footnote 26] While currently a small 
component of FHA's business, FHA expects that it would insure more 
condominium loans if the condominium program were moved to the Fund, as 
set forth in its legislative proposal. Additionally, TOTAL does not 
indicate the source of the down payment. We have reported that the 
source of a down payment is an important indicator of risk, and the use 
of down-payment assistance in the FHA program has grown substantially 
since 2000. Finally, our April 2006 report noted that the data used to 
develop TOTAL were not current and FHA had no plans to update the 
scorecard on a regular basis.

Consistent with our recommendations concerning TOTAL, FHA developed 
policies and procedures that call for (1) an annual evaluation of the 
scorecard's predictive ability, (2) testing of additional predictive 
variables to include in the scorecard, and (3) populating the scorecard 
with more recent loan performance data. An FHA contractor is helping 
the agency to implement these procedures and is scheduled to issue a 
final report on its work in August 2007. After receiving the 
contractor's report, FHA will decide what changes to TOTAL are 
necessary. Because the magnitude of these changes has not yet been 
determined, FHA does not have a completion date for this effort. FHA 
officials indicated that they would initially implement risk-based 
pricing using the current version of TOTAL but would use the updated 
version when it became available.

FHA Has Identified Needed Changes in Information Technology but Faces 
Funding and Implementation Challenges:

FHA has identified changes needed in its information technology to 
implement the legislative proposals. FHA has divided these changes into 
two phases. The first phase consists of simpler changes that it can 
make in the short term, such as revising the system used to originate 
FHA-insured loans to allow for down payments of less than 3 percent. 
FHA also would need to make other changes to the system to support the 
new loan limits, such as allowing the loan amount to equal 100 percent 
of the conforming loan limit in applicable areas. The second phase 
includes modifications to the computer programs that calculate the up- 
front and annual insurance premiums to reflect risk-based pricing and 
revisions related to the proposed changes to the HECM and condominium 
programs.

FHA has not yet obtained some of the funding needed to make the 
technology changes and does not have estimates for how long it would 
take to complete all of the changes. In fiscal year 2006, the agency 
obligated $2.8 million of the $10.9 million it estimated was needed to 
make all anticipated changes. Specifically, FHA plans to use funds 
reprogrammed from HUD's salaries and expense account and other 
available funds to complete the first phase of changes. FHA estimates 
that most of this work could be completed in a few months. The 
President's fiscal year 2008 budget requests an additional $8.1 million 
to fund the second phase of changes needed to implement the legislative 
proposals. However, FHA officials told us that they did not have an 
implementation schedule for this phase and were waiting until the 
legislative proposals were approved and they had secured the funding to 
develop one.

Although FHA officials indicated that they could implement the 
legislative proposals after making these minor information technology 
changes, they also told us that major systems changes and integration 
would be needed to bring FHA's systems up to levels comparable with 
other mortgage institutions. Currently, over 40 systems support FHA's 
single-family business activity. While a thorough evaluation of large- 
scale systems changes was outside the scope of our review, FHA has 
indicated that its systems are poorly integrated, expensive to 
maintain, and do not fully support the agency's operations and business 
requirements.[Footnote 27] For example, the systems cannot easily share 
or provide critical information because they use different database 
platforms with varying capabilities; some of the older systems use an 
outdated programming language; and the creation of ad hoc systems that 
do not interface with other systems has resulted in duplicate data 
entry. However, FHA has limited resources to devote to the development 
of new systems for two main reasons. First, it has to compete with 
other divisions within HUD for information technology resources. Of the 
approximately $300 million that HUD has requested for information 
technology development and maintenance in fiscal year 2008, about 5 
percent would be for FHA's single-family operations. Second, FHA spends 
what resources it has primarily on systems maintenance. Of the $19 
million that FHA has budgeted for single-family information technology 
in fiscal year 2007, FHA officials estimate that $15 million would be 
devoted to systems maintenance.

In contrast with FHA, officials from other mortgage institutions with 
whom we spoke indicated that they devote substantial resources to 
developing new systems and enhancing existing systems that help them 
price products and manage risk. To illustrate, officials from one 
mortgage institution stated that they had a $15 million annual budget 
for capital improvements in information technology. Officials from 
another mortgage institution told us that 17 percent of the company's 
total expenses were related to information technology and that they 
recently spent about $15 million to develop a new system to price a 
mortgage product for the foreign market. These and other mortgage 
industry officials stressed that investments in state-of-the-art 
information systems were critical to operating successfully in the 
highly competitive mortgage market.

FHA Has Sought Limited Staff Increases to Help Implement Proposals, but 
Other Workforce Challenges Remain:

According to FHA officials, the legislative proposals would not 
fundamentally alter how the agency administers its single-family 
mortgage insurance program and, therefore, would not require major 
increases in staff above the approximately 950 single-family housing 
employees it had as of March 2007. Although implementing the 
legislative proposals would require considerable program analysis and 
monitoring, much of the analysis required to develop the proposals was 
performed primarily by staff from FHA's Offices of Finance and Budget 
and Single Family Housing with assistance from several contractors, who 
will continue to support the implementation. FHA officials told us that 
marketing any new products authorized and explaining program changes to 
lenders would be their next major challenge if the legislative 
proposals were passed. They also noted that successful implementation 
would require them to stay abreast of developments in the mortgage 
market. Therefore, the President's fiscal year 2008 budget requests an 
additional 21 full-time equivalent (FTE) positions to help promote new 
FHA products, analyze industry trends, and align the agency's single- 
family business processes with current mortgage industry practices.

Although a detailed assessment of FHA's staffing needs was outside the 
scope of our review, a HUD contractor's 2004 workforce analysis 
suggests that FHA faces broader challenges that could affect the 
agency's operations going forward.[Footnote 28] The analysis projected 
that FHA would have 78 fewer FTEs than needed to handle anticipated 
work demands by fiscal year 2008, assuming hires and transfers equal to 
the average numbers for 2001 through 2003. In addition to anticipated 
FTE shortfalls, the report also identified existing and projected 
deficits of FHA staff with certain important competencies such as 
technical credibility and knowledge of single-family programs, 
policies, and regulations.[Footnote 29] For example, the consultant 
projected a difference of 28 percentage points between the percentage 
of staff requiring technical credibility and the percentage that would 
meet this requirement in fiscal year 2008. FHA officials have 
acknowledged the agency's staffing challenges and have developed plans 
to address the projected gaps. In fiscal years 2005 and 2006, FHA 
gained 228 staff through hiring or transfers. However, the contractor 
had assumed gains of 362 staff during those years, which means that the 
projected fiscal year 2008 shortfall will be worse than originally 
estimated without substantial staff accessions in fiscal years 2007 and 
2008.

FHA also faces hiring and salary constraints that other mortgage 
institutions do not. FHA's hiring authority is limited by statute and 
congressional appropriations. Federal statute (Title 5 of the U.S. 
Code) restricts the amounts that FHA can pay staff, and each year's 
appropriation determines how many staff it can hire. Further, FHA must 
compete with other divisions within HUD for staffing resources and may 
not always receive its full request. Other mortgage institutions have 
greater flexibility in their ability to hire and compensate staff. For 
example, Fannie Mae and Freddie Mac are not subject to federal pay and 
hiring restrictions. These restrictions create challenges for FHA as it 
competes for qualified staff in the competitive mortgage labor market.

FHA's Prior Risk Management Did Not Always Utilize Common Industry 
Practices Such as Piloting, but Some Planned Actions Could Help Address 
New Risks and Challenges:

Although FHA has not always utilized risk-management practices that 
other mortgage institutions use, it plans to take some steps to help 
address the new risks and challenges associated with the legislative 
proposals. In November 2005, we reported that HUD needed to take 
additional actions to manage risks related to the approximately one- 
third of its loans with down-payment assistance from seller-funded 
nonprofits.[Footnote 30] Unlike other mortgage industry participants, 
FHA does not restrict homebuyers' use of such assistance. Our 2005 
analysis found that the probability that these loans would result in an 
insurance claim was 76 percent higher than for comparable loans without 
such assistance, and we recommended that FHA revise its underwriting 
standards to consider such assistance as a seller contribution (which 
cannot be used to meet the borrower contribution requirement).[Footnote 
31] Despite the detrimental impact of these loans on the Fund, FHA did 
not act promptly to mitigate the problem by adjusting underwriting 
standards or using its existing authority to raise premiums. However, 
in May 2007, FHA published a proposed rule that would prohibit seller- 
funded down-payment assistance.[Footnote 32]

In addition, as we reported in February 2005, other mortgage 
institutions limit the availability of or pilot new products to manage 
risks associated with changing or expanding product lines.[Footnote 33] 
We have previously indicated that, if Congress authorizes FHA to insure 
new products, it should consider a number of means, including limiting 
their initial availability, to mitigate the additional risks these 
loans may pose. We also recommended that FHA consider similar steps for 
any new or revised products. However, in response, FHA officials told 
us that they lacked the resources to effectively manage a program with 
limited volumes. We noted that if FHA did not limit the availability of 
new or changed products, the potential costs of making widely available 
a product with risks that may not be well understood could exceed the 
cost of a pilot program. With respect to its legislative proposal, FHA 
officials told us that they do not plan to pilot or limit the initial 
availability of any zero-down-payment product the agency was authorized 
to offer. They also indicated that they expected that a zero-down- 
payment product would perform similarly to loans with seller-funded 
down-payment assistance. While the experience of loans with this type 
of assistance is informative, a zero-down-payment product could be 
utilized by a different population of borrowers and may not perform the 
same as these loans.

Nevertheless, if the legislative proposals were to be enacted, FHA 
plans to take some steps to help address risks and challenges 
associated with (1) managing the risks of no-down-payment loans, (2) 
setting premiums to achieve a modestly negative subsidy rate, and (3) 
modifying oversight of lenders. First, loans with low or no down 
payments carry greater risk because of the direct relationship that 
exists between the amount of equity borrowers have in their homes and 
the risk of default. The higher the LTV ratio, the less cash borrowers 
will have invested in their homes and the more likely it is that they 
may default on mortgage obligations, especially during times of 
economic hardship or price depreciation in the housing market. No-down- 
payment loans became common in the conventional market when rapid 
appreciation in home prices helped mitigate the risk of these loans. 
However, if authorized to offer a zero-down-payment mortgage in the 
near future, FHA would be introducing this product at a time when home 
prices have stagnated or are declining in some parts of the country. 
And because FHA would continue to allow borrowers to finance some 
portion of closing costs and up-front insurance premiums, the effective 
LTV ratio for loans with very low or no down payments could be greater 
than 100 percent, further increasing FHA's insurance risk. To mitigate 
the risks associated with loans with no down payments, FHA plans to 
impose stricter underwriting criteria for such loans:

* FHA would limit the amount of up-front premium and closing costs that 
could be financed; therefore, all borrowers would be making some 
minimum cash contribution.

* FHA plans to require a minimum credit score of 640 to obtain FHA 
insurance on loans with no down payments.[Footnote 34]

* FHA would limit its zero-down-payment product to loans for owner- 
occupied, one-unit properties.

Second, FHA's legislative proposal would fundamentally change the way 
the agency manages the Fund in that FHA would set premiums to achieve a 
modestly negative overall subsidy rate, representing the weighted 
average of the subsidy rates for the different risk-based pricing 
categories. The President's budget for fiscal year 2008 estimates that 
the weighted average subsidy rate would be -0.6 percent (meaning that 
the Fund would generate negative subsidies amounting to 0.6 percent of 
the total dollars insured for loans originated that year).[Footnote 35] 
Achieving a modestly negative credit subsidy rate would depend on FHA's 
ability to price new products whose risks may not be well understood, 
although risk-based pricing could help FHA be more precise in setting 
and adjusting premiums for different segments of its portfolio. FHA 
officials told us that they would monitor the proportion of loans in 
its two highest-risk categories and consider raising premiums or 
tightening underwriting standards if unexpectedly high demand exposed 
FHA to excessive financial risk. Fannie Mae, Freddie Mac, and the four 
private mortgage insurers we interviewed noted that they carefully 
monitor their portfolios to make sure that they do not have too many 
loans in any given risk category and take similar steps when they 
determine that this is the case.

Third, FHA may need to modify the way that it oversees lenders if the 
legislative proposals were enacted. FHA has indicated that its 
legislative proposals would help the agency to expand service to higher-
risk borrowers in a financially sound manner. However, FHA may need to 
revise its Credit Watch program if it is to achieve this end. Under 
Credit Watch, FHA terminates the loan origination authority of any 
lender branch office that has a default and claim rate on mortgages 
insured by FHA in the prior 24 months that exceeded both the national 
average and 200 percent of the average rate for lenders in its 
geographic area. Because termination currently is based on how a 
lender's loans perform relative to other lenders in its geographic 
area, lenders that chose to make loans to higher-risk borrowers could 
suffer in comparison with lenders that served only lower-risk 
borrowers. To encourage lenders to serve borrowers in the higher-risk 
categories, FHA officials told us that they would consider taking into 
account the mix of borrowers in the various risk categories when 
evaluating a lender's performance. Because higher-risk loans can be 
expected to incur higher default and claim rates, they stated that FHA 
would not want to penalize lenders with larger shares of these loans as 
long as the loans were performing within expected risk parameters. FHA 
also has improved the accuracy and timeliness of the loan performance 
data it uses to evaluate lenders by requiring lenders to update the 
delinquency status of their loans more frequently.

Congress and FHA Could Consider Other Administrative and Legislative 
Changes to Help FHA Adapt to Changes in the Mortgage Market:

Mortgage industry participants and researchers have suggested 
additional options that Congress and FHA could consider to help FHA 
adapt to changes in the mortgage market, but some changes could have 
budget and oversight implications. FHA already has authority to 
undertake some of these options. Other options would require additional 
authorities from Congress to increase the agency's operational 
flexibility. Congress also could consider alternative approaches to the 
provision of federal mortgage insurance such as converting FHA to a 
government corporation or implementing risk-sharing arrangements with 
private partners.

FHA Has Existing Authority to Make More Administrative Changes:

Although FHA already has made several administrative changes to 
streamline the agency's insurance processes, additional administrative 
changes within FHA's existing authority could alleviate, to some 
extent, the need for a positive subsidy in fiscal year 2008. More 
specifically, FHA could exercise its existing authority to raise up- 
front premiums up to 2.25 percent and, for borrowers with down payments 
of less than 5 percent, annual premiums to 0.55 percent.

To moderate the need for a positive subsidy in fiscal year 2008, FHA 
could use its existing authority to increase premiums in one of three 
ways: (1) FHA could raise premiums for all borrowers, as the 
President's fiscal year 2008 budget suggests will be necessary; (2) FHA 
could charge the higher 0.55 percent annual premium to borrowers with 
lower down payments; or (3) FHA could implement a more limited form of 
risk-based pricing than it has proposed by adjusting premiums within 
the current statutory limits. HUD's Office of General Counsel 
determined in March 2006 that FHA has the authority to structure 
premiums for programs under the Fund on the basis of risk. FHA could 
implement premium adjustments, either for all or some borrowers, 
through the regulation process. However, according to FHA officials, 
the current statutory limits on premiums are too low to allow FHA to 
implement a risk-based pricing plan that would allow the agency to set 
prices high enough to compensate for the expected losses from the 
highest-risk borrowers or a new zero-down-payment product. And while 
raising premiums for some higher-risk borrowers could improve the 
Fund's credit subsidy rate, raising premiums for all borrowers might 
exacerbate FHA's adverse selection problem. That is, FHA could lose 
higher credit quality borrowers, resulting in fewer borrowers to 
subsidize lower credit quality borrowers. This, in turn, could require 
FHA to raise premiums again.

Additional Authorities for Investment in Technology, Pay and Hiring, 
and Introduction of Products Could Increase FHA's Operational 
Flexibility:

According to mortgage industry participants and researchers, Congress 
also could consider granting FHA additional authorities to increase the 
agency's ability to invest in technology and staff or offer new 
insurance products. First, Congress could grant FHA specific authority 
to invest a portion of the Fund's current resources--that is, negative 
subsidies that accrue in the Fund's reserves--in technology 
enhancement. The congressionally-appointed Millennial Housing 
Commission (MHC) found that FHA's dependence on the appropriations 
process for budgetary resources and competition for funds within HUD 
had led to under-investment in technology, increasing the agency's 
operational risk and making it difficult for FHA to work efficiently 
with lenders and other industry partners.[Footnote 36] Because FHA's 
single-family insurance program historically has generated estimated 
negative subsidies, FHA and some mortgage industry officials have 
suggested that the agency be given the authority to use a portion of 
the Fund's current resources to upgrade and maintain its technology.

One benefit of this option is that the technology enhancements could 
improve FHA's operations. As previously noted, FHA has more than 40 
single-family information systems that are poorly integrated, expensive 
to maintain, and do not fully support the agency's business 
requirements. However, according to FHA, the option would require a 
statutory change to allow FHA to use the Fund's current resources to 
pay for technology improvements. Also, the Fund is required by law to 
operate on an actuarially sound basis. Because the soundness of the 
Fund is measured by an estimate of its economic value--an estimate that 
is subject to inherent uncertainty and professional judgment--the 
Fund's current resources should be used with caution. Spending the 
Fund's current resources would lower the Fund's reserves, which in turn 
would lower the economic value of the Fund. As a result, the Fund's 
ability to withstand severe economic conditions could be diminished. 
Also, using the Fund's current resources would increase the federal 
budget deficit unless accompanied by corresponding reductions in other 
government spending or an increase in receipts.

Second, Congress could consider allowing FHA to manage its employees 
outside of federal pay scales. Some federal agencies, such as the 
Securities and Exchange Commission, the Office of Thrift Supervision, 
and the Federal Deposit Insurance Corporation, are permitted to pay 
salaries above normal federal pay scales in recognition of the special 
skills demanded by sophisticated financial market operations.[Footnote 
37] The MHC and mortgage industry officials have suggested that FHA be 
given similar authority. This option could help FHA to recruit 
experienced staff to help the agency adapt to market changes. Like the 
authority to invest in technological enhancement, this option could be 
funded with the Fund's current resources but would have similar 
implications for the financial health of the Fund and the federal 
budget deficit.

Third, Congress could authorize FHA to offer and pilot new insurance 
products without prior congressional approval. A variety of new 
mortgage products have appeared in the mortgage market in recent years, 
but FHA's ability to keep pace with market innovations is limited. For 
example, the MHC found that the statutes and regulations to which FHA 
is subject dramatically increase the time necessary to develop and 
implement new products. The MHC and mortgage industry officials have 
recommended that Congress expressly authorize FHA to introduce new 
products without requiring a new statute for each. Such authority would 
offer FHA greater flexibility to keep pace with a rapidly changing 
mortgage market. However, Congress would have less control over FHA's 
product offerings and, in some cases, it might take years before a new 
product's risks were well understood.

To manage the risks of new products, mortgage institutions may impose 
limits on the volume of the new products they will permit and on who 
can sell and service those products. Limits on the availability of new 
or revised FHA mortgage insurance products are sometimes set through 
legislation and focus on the volume of loans that FHA may insure. In a 
prior report, we recommended that FHA consider using pilots for new 
products and making significant changes to its existing 
products.[Footnote 38] Since FHA officials questioned the circumstances 
in which they could use pilots or limit volumes when not required by 
Congress, we also recommended that FHA seek the authority to offer new 
products on a limited basis, such as through pilots, if the agency 
determines it currently lacks sufficient authority. However, FHA has 
not sought this authority. Furthermore, while piloting could help FHA 
manage the risks associated with implementing new products, FHA 
officials told us that they lack the resources to manage a program with 
limited volumes effectively.[Footnote 39]

Finally, Congress could authorize FHA to insure less than 100 percent 
of the value of the loans it guarantees. Unlike private mortgage 
insurers, which offer several levels of insurance coverage up to a 
maximum of 40 or 42 percent (depending on the company) of the value of 
the loan, FHA insures 100 percent of the value of the loan. But since 
most FHA insurance claims are offset by some degree of loss recovery, 
some mortgage industry observers have suggested that covering 100 
percent of the value of the loan may not be necessary. In prior work, 
we examined the potential effects of reducing FHA's insurance coverage 
and found that while lower coverage would cause a reduction in the 
volume of FHA-insured loans and a corresponding decline in income from 
premiums, it would also result in reduced losses and ultimately have a 
beneficial effect on the Fund.[Footnote 40] However, we also noted that 
partial FHA coverage could lessen FHA's ability to stabilize local 
housing markets when regional economies decline and may increase the 
cost of FHA-insured loans as lenders set higher prices to cover their 
risk.

Alternative Approaches for Providing Federal Mortgage Insurance Include 
Converting FHA to a Government Corporation:

The MHC, HUD officials, and other mortgage industry participants have 
suggested alternative approaches to provide federal mortgage insurance 
in a changing mortgage market. First, since the mid-1990s, several 
groups including HUD and the MHC have proposed converting FHA into 
either an independent or a HUD-owned government corporation--that is, 
an agency of government, established by Congress to perform a public 
purpose, which provides a market-oriented service and produces revenue 
that meets or approximates its expenditures. Government corporations 
operate more independently than other agencies of government and can be 
exempted from executive branch budgetary regulations and personnel and 
compensation ceilings. Therefore, converting FHA to a corporation could 
provide the corporation's managers with the flexibility to determine 
the best ways to meet policy goals set by Congress or HUD.

This option could have budgetary and oversight implications that would 
need to be considered when setting up the new corporation. For example, 
Congress would have to determine the extent to which (1) the 
corporation's earnings in excess of those needed for operations and 
reserves would be available for other government activities and (2) the 
corporation would be subject to federal budget requirements. Also, if 
the corporation were created outside of HUD, Congress would have to 
consider whether oversight of the corporation would require a new 
oversight institution or could be performed by an existing organization.

Alternatively, rather than maintaining all the functions of a mortgage 
insurer within a government entity, the MHC and private mortgage 
insurers have suggested that the federal government could provide 
mortgage insurance through risk-sharing agreements with private 
partners.[Footnote 41] FHA already works with partners to conduct 
various activities related to its operations. For example, FHA has 
delegated underwriting authority to approved lenders, and contractors 
perform many day-to-day activities (such as marketing foreclosed 
properties) that once were performed by FHA employees. A public-private 
risk-sharing arrangement would recognize that government has a better 
ability to spread risk, while private mortgage industry participants 
generally are more flexible and responsive to market pressures and 
better able to innovate and adopt new technologies quickly. There are 
many different possible ways to structure a risk-sharing approach, with 
variables such as the amount of insurance coverage provided, the number 
and type of risk-sharing partners, the degree of risk accepted by each 
partner, and the roles and responsibilities of the partners.

Whatever the structure, a risk-sharing approach could result in greater 
efficiency and allow FHA to reach new borrowers through new partner 
channels. However, risk sharing also could diminish the federal 
government's ability to stabilize markets if private partners lacked 
incentive to serve markets where economic conditions were 
deteriorating. Additionally, implementing risk-sharing arrangements 
might require more specialized expertise than FHA currently has among 
its staff. For example, careful analysis in both program design and 
monitoring would be needed to ensure that FHA's financial interests 
were adequately protected.

Finally, Congress and FHA could elect to make no changes at this time 
and allow the private market to play the definitive role in determining 
the future need for federal mortgage insurance. The recent decline in 
FHA's market share occurred at a time when interest rates were low, 
house price appreciation was high, and mortgage credit was widely 
available. However, changes in the mortgage market, such as higher 
interest rates and stricter underwriting standards for subprime loans, 
may lead to an increasing role for FHA in the future or at least a 
continued role for the federal government in guaranteeing mortgage 
credit for some borrowers. Therefore, even if Congress and FHA were to 
make no changes at this time, FHA's market share might increase due to 
the recent change in market conditions. Or it might eventually become 
so small as to indicate that there is no longer a need for a federal 
role in providing mortgage insurance. If FHA's market share continues 
to decline to such a level, FHA might be eliminated or critical 
functions reassigned to maintain a minimal federal role in guaranteeing 
mortgage credit.

Making no changes to FHA at this time would acknowledge the substantial 
role the private market now plays in meeting the mortgage credit needs 
of borrowers. However, some home buyers might find it more difficult 
and more costly to obtain mortgages if FHA were eliminated or its 
functions reduced and reassigned to another federal agency.[Footnote 
42] And allowing FHA to become too small could impact the federal 
government's ability to play a role in stabilizing mortgage markets 
during an economic downturn. Also, any option that might lead to the 
eventual elimination of FHA's single-family mortgage insurance program 
would have broader implications for FHA and its other programs, such as 
the multifamily mortgage insurance and regulatory programs, which this 
report does not address. Such implications would, therefore, require 
further study.

Observations:

Recent trends in the mortgage market, including the prevalence of low- 
and no-down-payment mortgages and increased competition from 
conventional mortgage and insurance providers, have posed challenges 
for FHA. FHA's market share has declined substantially over the years, 
and what was a negative subsidy rate for the single-family insurance 
program has crept toward zero. To adapt to market changes, FHA has 
implemented new administrative procedures and proposed legislation 
designed to modernize its mortgage insurance processes, introduce 
product changes, and provide additional risk-management tools. To its 
credit, FHA has performed considerable analysis to support its 
legislative proposal and has made or planned enhancements to many of 
the specific tools and resources that would be important to its 
implementation.

However, the proposals present risks and challenges and should be 
viewed with caution for several reasons. First, FHA has not always 
effectively managed risks associated with product changes, most notably 
the growth in the proportion of FHA-insured loans with seller-funded 
down-payment assistance. In that case, FHA did not use the risk- 
management tools already at its disposal to mitigate adverse loan 
performance that has had a detrimental impact on the Fund. Second, the 
proposal to lower down-payment requirements potentially to zero raises 
concerns given the greater default risk of loans with high LTVs, 
policies that could result in effective LTV ratios of over 100 percent, 
and housing market conditions that could put borrowers with such loans 
in a negative equity position. Sound management of very low or no-down- 
payment products would be necessary to help ensure that FHA and 
borrowers do not experience financial losses. Piloting or otherwise 
limiting the availability of new products would allow FHA the time to 
learn more about the performance of these loans and could help avoid 
unanticipated insurance claims. Despite the potential benefits of this 
practice, FHA generally has not implemented pilots, unless directed to 
do so by Congress. We have previously indicated that, if Congress 
authorizes FHA to insure new products, Congress and FHA should consider 
a number of means, including limiting their initial availability, to 
mitigate the additional risks these loans may pose. We continue to 
believe that piloting would be a prudent approach to introducing the 
products authorized by FHA's legislative proposal. Finally, FHA would 
face the challenge of setting risk-based premiums--potentially for 
products whose risks may not be well understood--to achieve a specific 
financial outcome, a relatively small negative subsidy. Because the 
estimated subsidy rate is close to zero and FHA has consistently 
underestimated its subsidy costs, FHA runs some risk of missing its 
target and requiring a positive subsidy. Additionally, limitations we 
have identified in FHA's TOTAL scorecard, which would be a key tool 
used in risk-based pricing, could reduce the agency's ability to set 
prices commensurate with the risk of the loans. Accordingly, it will be 
important for FHA to continue making progress in addressing these 
limitations.

Our recent report on trends in FHA's market share underscores the 
challenges that FHA has faced in adapting to the changing mortgage 
market. For example, we noted that FHA's share of the market for home 
purchase mortgages has declined precipitously since 2001 due in part to 
FHA product restrictions and a lack of process improvements relative to 
the conventional market. While FHA has taken some steps to improve its 
processes and enhance the tools and resources that it would use to 
implement the modernization proposals, additional changes may be 
necessary for FHA to operate successfully in the long run in a 
competitive and dynamic mortgage market. Other mortgage industry 
participants have greater flexibility to hire and compensate staff, 
invest in information technology, and introduce new products, enhancing 
their ability to adapt to market changes and manage risk. A number of 
policy options that go beyond FHA's modernization proposals would give 
FHA similar flexibility but would have other implications that would 
require careful deliberation.

Agency Comments and Our Evaluation:

We provided HUD with a draft of this report for review and comment. HUD 
provided comments in a letter from the Assistant Secretary for Housing- 
Federal Housing Commissioner (see app. II). HUD said that the draft 
report provided a balanced assessment but also that the report's 
concerns about FHA's risk management and emphasis on the need for 
piloting lower-down-payment products were unwarranted.

HUD said that it welcomed the draft report's acknowledgment of FHA's 
improvements in program administration and risk management but 
questioned the report's concerns about FHA's ability to understand and 
manage risk. HUD indicated that its proposal to diversify FHA's product 
offerings and pricing structure grew out of recognition that FHA was 
subject to adverse selection, as evidenced by the loss of borrowers 
with better credit profiles and growth in seller-funded down-payment 
assistance loans. In addition, HUD listed steps it had taken to curtail 
seller-funded down-payment assistance, including publishing a proposed 
rule in May 2007 that would effectively eliminate seller-funded down- 
payment assistance in conjunction with FHA-insured loans. Our draft 
report cited a number of improvements in FHA's risk management, such as 
enhancements to its loan performance models. However, we continue to 
believe that our concerns about FHA's ability to manage risk are 
warranted. As our draft report noted, FHA did not take prompt action to 
mitigate the adverse financial impact of loans with seller-funded down- 
payment assistance. Furthermore, our draft report identified additional 
steps, such as improvements to TOTAL scorecard, that would help address 
the risks and challenges associated with the legislative proposals.

With regard to piloting, HUD said that pilot programs are appropriate 
where a concept is untested but that the concept of zero-or lower-down- 
payments was well understood. HUD indicated that it had a firm basis 
for anticipating the performance of zero-and lower-down-payment loans 
as a result of its experience with mortgages with seller-funded down- 
payment assistance. HUD said it used this experience to establish risk- 
based insurance premiums and minimum credit scores for zero-and lower- 
down-payment borrowers. Additionally, HUD said that it had recently 
started to collect 30-day and 60-day delinquency data, giving the 
agency the capability to track performance trends for different 
segments of its loan portfolio on a monthly basis. HUD stated that, for 
these reasons, the risks of zero-or lower-down-payment loans were 
sufficiently well known or knowable to not warrant a pilot program.

As our draft report noted, we previously have reported that other 
mortgage institutions limit the availability of, or pilot, new products 
to manage the risks associated with changing or expanding their product 
lines and have recommended that FHA consider adopting this practice. 
Our draft report also acknowledged that FHA's experience with seller- 
funded down-payment assistance could inform assessment of how a zero- 
down-payment product would perform. However, we continue to believe 
that FHA should consider limiting the availability of a loan product 
with no down payment. In particular, our draft report discussed two 
factors that indicate the need for caution in introducing such a 
product. First, a zero-down-payment product could be utilized by a 
different population of borrowers than seller-funded down-payment 
assistance loans and may not perform similarly to these loans. Second, 
zero-down-payment loans became common in the conventional mortgage 
market when rapid appreciation in home prices helped mitigate the risks 
of these loans. If authorized to offer a zero-down-payment product in 
the near future, FHA would be introducing it at a time when home prices 
have stagnated or are declining in some parts of the country. Because 
of these risks and uncertainties, we continue to believe that a prudent 
way to introduce a zero-down-payment product would be to limit its 
initial availability such as through a pilot program.

We are sending copies of this report to the Chairman, Senate Committee 
on Banking, Housing, and Urban Affairs; Chairman and Ranking Member, 
Subcommittee on Housing and Transportation, Senate Committee on 
Banking, Housing, and Urban Affairs; Chairman and Ranking Member, House 
Committee on Financial Services; and Chairman and Ranking Member, 
Subcommittee on Housing and Community Opportunity, House Committee on 
Financial Services. We will also send copies to the Secretary of 
Housing and Urban Development and to other interested parties and make 
copies available to others upon request. In addition, the report will 
be made available at no charge on the GAO Web site at http:// 
www.gao.gov.

Please contact me at (202) 512-8678 or shearw@gao.gov if you or your 
staff have any questions about this report. Contact points for our 
Offices of Congressional Relations and Public Affairs may be found on 
the last page of this report. Key contributors to this report are 
listed in appendix III.

Signed by: 

William B. Shear: 
Director, Financial Markets and Community Investment:

[End of section]

Appendix I: Objectives, Scope, and Methodology:

The Ranking Member of the Senate Committee on Banking, Housing, and 
Urban Affairs and Senator Wayne Allard requested that we evaluate FHA's 
modernization efforts, which include administrative and proposed 
legislative changes. Specifically, we examined (1) the likely program 
and budgetary impacts of FHA's modernization efforts, (2) the tools, 
resources, and risk-management practices important to FHA's 
implementation of the legislative proposals, if passed, and (3) other 
options that FHA and Congress could consider to help FHA adapt to 
changes in the mortgage market and the pros and cons of these options.

To determine the likely program and budgetary impacts of FHA's 
modernization efforts, we reviewed FHA guidance on three administrative 
changes implemented in 2006: the Lender Insurance Program and revisions 
to the agency's appraisal protocols and closing cost guidelines. To 
determine the extent to which these administrative changes have 
affected the processing of FHA-insured loans, we interviewed 
representatives of Countrywide Financial, Wells Fargo, Bank of America, 
and Lenders One (a mortgage co-operative representing about 90 
independent mortgage bankers). We selected Countrywide Financial and 
Wells Fargo because they are large FHA lenders, Bank of America because 
it had recently decided to grow its FHA business, and Lenders One 
because some of its members make FHA loans. We also interviewed 
representatives of three mortgage and real estate industry groups-- 
Mortgage Bankers Association, National Association of Realtors, and 
National Association of Home Builders. To determine how the Lender 
Insurance Program has affected the processing of FHA insurance, we 
interviewed FHA officials and obtained documentation from them on the 
extent of lender participation in the program and its effect on 
insurance processing time and costs.

In evaluating the likely program impacts of FHA's proposed legislative 
changes, we focused on the proposals to raise FHA loan limits, 
institute risk-based pricing of mortgage insurance premiums, and lower 
down-payment requirements. To examine the effect of raising loan limits 
on demand for FHA-insured loans, we analyzed 2005 HMDA data (the most 
current available). Specifically, we analyzed the home purchase loans 
recorded in 2005 to determine the number of loans in each of 380 core 
based statistical areas (CBSA) and used that data to calculate FHA's 
market share in each CBSA.[Footnote 43] (These 380 CBSAs were those for 
which we had data and included one aggregate "nonmetro" category.) We 
then determined the number of additional loans that, based on their 
loan amounts, would have been eligible for FHA insurance in 2005 had 
the higher proposed loan limits been in effect. Finally, we estimated 
the percentage of the newly-eligible loans in each CBSA that FHA would 
have insured using the following range of assumptions: (1) that FHA's 
market share would have been approximately the same as it was among all 
loans in that CBSA under the actual 2005 loans limits, (2) that FHA's 
market share would have been approximately the same as its share of 
loans with loan amounts ranging from 70 to 100 percent of the actual 
2005 loan limits in that CBSA, (3) that FHA's market share would have 
been approximately the same as its share of loans with loan amounts 
ranging from 75 to 100 percent of the actual 2005 loan limits in that 
CBSA, and (4) that FHA's market share would be approximately the same 
as its share of loans with loan amounts ranging from 80 to 100 percent 
of the actual 2005 loan limits in that CBSA.[Footnote 44]

For each of these four scenarios, we calculated the total number and 
dollar amount of new loans across all 380 CBSAs that could have been 
insured by FHA had the higher loan limits been in effect. All four 
assumptions yielded similar results. After arriving at an estimate of 
an overall increase in the number of FHA-insured loans, we then 
determined the proportions of the increase that would have resulted 
from raising the loan limit floor in low-cost areas, raising the loan 
limit ceiling in high-cost areas, or raising the limits in areas that 
fell between the floor and the ceiling. Finally, we calculated the 
average FHA-insured loan amount in 2005, as well as the average loan 
amount that FHA might have insured had the loan limits been increased. 
We assessed the reliability of the HMDA data we used by reviewing 
information about the data, performing electronic data testing to 
detect errors in completeness and reasonableness, and interviewing a 
knowledgeable official regarding the quality of the data. We determined 
that the data were sufficiently reliable for the purposes of this 
report.

To estimate the effects of risk-based pricing on borrowers' eligibility 
for FHA insurance and the premiums they would pay, we reviewed FHA's 
risk-based pricing proposal and interviewed FHA officials regarding 
their plans to implement risk-based pricing, if authorized. We then 
analyzed SFDW data on FHA's 2005 home purchase borrowers to determine 
how they would have been affected by FHA's risk-based pricing proposal. 
(We focused on 2005 borrowers because that was the most recent year for 
which we had complete data, and we restricted our analysis to purchase 
loans because they comprise the bulk of FHA's business.) First, we 
assigned borrowers to one of seven categories (FHA's six proposed risk- 
based pricing categories and one category for those who would not have 
been eligible for FHA insurance) based upon their LTV ratio and credit 
score. Since FHA does not currently insure loans without a down 
payment, we identified borrowers with down-payment assistance and 
determined the source and amount of assistance to approximate borrowers 
with LTV ratios of 100 percent. We recalculated the LTV ratio of their 
loans by adding the amount of their assistance to the principal balance 
of their loan. We then examined the demographic characteristics (race, 
income, and first-time home buyer status) of borrowers in each of the 
six pricing categories, as well as those borrowers who would no longer 
qualify for FHA insurance. We assessed the reliability of the SFDW data 
we used by reviewing information about the system and performing 
electronic data testing to detect errors in completeness and 
reasonableness. We determined that the data were sufficiently reliable 
for the purposes of this report.

We also interviewed representatives of the following consumer advocacy 
groups to obtain their views on FHA's proposed legislative changes: 
Center for Responsible Lending, Consumer Action, Consumer Federation of 
America, National Association of Consumer Advocates, National Community 
Reinvestment Coalition, National Consumer Law Center, and National 
Council of La Raza. We examined the potential budgetary impacts of the 
legislative proposals by reviewing the President's fiscal year 2008 
budget and FHA cost estimates as shown in the 2008 Federal Credit 
Supplement. (The Federal Credit Supplement provides summary information 
about federal direct loan and loan guarantee programs, including 
current subsidy rates and reestimated subsidy rates.) To determine the 
tools, resources, and risk-management practices important to FHA's 
implementation of the legislative proposals, we interviewed and 
reviewed documentation from FHA officials regarding the agency's plans 
for implementing the legislative proposals, if passed. We focused on 
completed and planned enhancements to FHA's SFDW data, loan performance 
models, TOTAL mortgage scorecard, information technology, human 
capital, and risk-management practices. To help us evaluate the need 
for enhancements to FHA's tools, resources, and practices, we followed 
up on our past work on (1) FHA's development and use of TOTAL, (2) 
FHA's estimation of subsidy costs for its single- family insurance 
program, (3) practices that could be instructive for FHA in managing 
the risks of new mortgage products, and (4) FHA's management of loans 
with down-payment assistance.[Footnote 45] To obtain information on the 
tools and resources that other mortgage institutions use to set prices 
and manage risk, we interviewed Fannie Mae, Freddie Mac, the Mortgage 
Insurance Companies of America (the industry group that represents the 
private mortgage insurance industry), and four private mortgage 
insurance companies--AIG United Guaranty, Genworth Mortgage Insurance 
Company, Mortgage Guaranty Insurance Corporation, and PMI Mortgage 
Insurance Company.

To determine other options that FHA and Congress could consider and the 
pros and cons of these options, we reviewed relevant literature, 
including the report of the Millennial Housing Commission,[Footnote 46] 
articles discussing past FHA restructuring proposals,[Footnote 47] and 
our past work on various options for FHA.[Footnote 48] We also 
interviewed FHA officials, academic experts, FHA lenders, and private 
mortgage insurance companies.

We conducted this work in Washington, D.C., from September 2006 to June 
2007 in accordance with generally accepted government auditing 
standards.

[End of section]

Appendix II: Comments from the Department of Housing and Urban 
Development:

[See PDF for image].

[End of figure].

U.S. Department Of Housing And Urban Development: 
Washington, DC 20410-8000:

Assistant Secretary For Housing- 
Federal Housing Commissioner:

JUN - 8 2007:

Mr. William B. Shear: 
Director:
Financial Markets and Community Investments: 
United States Government Accountability Office: 
441 G Street, NW:
Washington, DC 20548:

Dear Mr. Shear:

Thank you for the opportunity to comment on the Government 
Accountability Office (GAO) report, "Federal Housing Administration: 
Modernization Proposals Would Have Program and Budget Implications and 
Require Continued Improvements in Risk Management," (GAO-07 708). I 
welcome GAO's acknowledgement that FHA has made substantial 
improvements in both program administration and risk management, but 
feel that GAO's remaining reservations about FHA's ability to manage 
risk are unwarranted, and I believe that its recommendation that FHA 
implement lower downpayment programs by pilot is not appropriate 
considering the large volume of data now available for this type of 
program. Let me elaborate.

FHA's modernization proposals grow out of internal analyses showing 
that FHA was being adversely selected. As a result of profound changes 
in primary and secondary mortgage markets, borrowers traditionally 
served by FHA-lower-income and minority first-time homebuyers 
increasingly had access to mortgage financing, but frequently at 
excessive cost. As a government agency closely bound by statutes and 
regulations, FHA was unable to respond to these changes in a flexible 
manner. Under its one-size fits all pricing structure, FHA witnessed 
the loss of borrowers with better credit profiles and the rapid growth 
of seller-funded downpayment assistance loans.

FHA recognized the need to diversify its product offerings and its 
pricing structure. It further recognized that borrowers using seller- 
funded downpayment assistance loans would be served at lower risk to 
themselves and to FHA's insurance fund by zero-or lower-downpayment 
loans that would remove the incentive of parties to the purchase 
transaction to inflate house prices and, consequently, mortgage 
amounts, monthly payments, and borrower risk. FHA has taken steps to 
curtail seller-funded downpayment assistance. It supported the Internal 
Revenue Service when it issued a ruling withdrawing nonprofit status 
from entities funneling seller-funded downpayment assistance to 
borrowers, and, on May 11, 2007, it published a proposed rule that 
would effectively eliminate seller-funded downpayment assistance from 
use with FHA-insured loans. FHA believes that other alternatives, 
including risk-based pricing and a variable downpayment program, would 
better serve the same types of borrowers who currently need such 
assistance.

In FY 2005, FHA included Zero Downpayment and Payment Incentives 
products in its budget proposal, but these initiatives were not enacted 
into law. In FY 2006, FHA submitted a comprehensive reform proposal to 
Congress that passed the House of Representatives with a 415-to-7 vote, 
but the legislation stalled in the Senate. In FY 2007, FHA is again 
promoting a reform proposal that would increase its flexibility in 
offering mortgage insurance products and would permit a wider range of 
mortgage insurance premiums based on objective indicators of risk. With 
this new authority, FHA seeks to increase the options available to its 
traditional borrowers-lower-income and minority first-time homebuyers- 
so that they can attain their goal of homeownership at reasonable cost. 
Pricing risk in a mortgage insurance premium instead of the mortgage 
interest rate gives borrowers access to prime interest rates, lowers 
their current and overall borrowing costs, and facilitates more 
transparent mortgage transactions.

While the GAO report shows commendable understanding of FHA, it 
continues to express concern about FHA's ability to understand and 
manage risk, and recommends that FHA pursue reform through the use of 
pilot programs. FHA recognizes that pilot programs are appropriate 
where a concept is untested. For example, in 1989, FHA piloted Home 
Equity Conversion Mortgages that today have become FHA's fastest 
growing program. But the concept of zero-or lower-downpayments is well 
understood. In fact, the National Association of Realtors polled 7,548 
consumers who bought homes between mid-2005 and mid-2006 and found that 
45 percent of first-time buyers financed 100 percent of the 
transaction; they made no downpayments whatsoever.

FHA already has a firm basis for anticipating the performance of zero- 
and lower-downpayment loans by virtue of its experience with seller- 
funded downpayment assistance. In its FY 2008 budget proposal, FHA used 
this experience to posit minimum FICO scores for zero-and lower- 
downpayment borrowers and premiums graded by potential risk based on 
historical experience. As GAO is aware, FHA uses conditional claim and 
prepayments rates produced by an independent annual actuarial review in 
state-of-the-art cash flow models to re-examine loan performance and 
reset premiums on an annual basis. Having recently started to collect 
30-and 60-day as well as 90-day delinquency data, FHA now has the 
capability of tracking trends by book of business or other segment of 
the portfolio on a monthly basis. For these reasons, FHA believes that 
the risks of zero-or lower-downpayment loans are sufficiently known or 
knowable to that a a pilot program is unwarranted.

In conclusion, FHA is grateful for GAO's balanced assessment, but 
respectfully suggests that FHA is ready now to manage the risks of zero-
and lower-downpayment loans and so, to offer an alternative to 
borrowers who would otherwise be turned away or turn to higher cost 
alternatives.

Sincerely,

Signed by: 

Brian D. Montgomery: 
Assistant Secretary for Housing- 
Federal Housing Commissioner:

[End of section]

Appendix III: GAO Contact and Staff Acknowledgments:

GAO Contact:

William Shear (202) 512-8678 or shearw@gao.gov:

Staff Acknowledgments:

In addition, Steve Westley (Assistant Director), Steve Brown, Laurie 
Latuda, John McGrail, Barbara Roesmann, Paige Smith, and Richard 
Vagnoni made key contributions to this report.

FOOTNOTES

[1] The conventional market comprises mortgages that do not carry 
government insurance or guarantees. For more information on the decline 
in FHA's share of the mortgage market and the factors underlying this 
trend, see GAO, Federal Housing Administration: Decline in the Agency's 
Market Share Was Associated with Product and Process Developments of 
Other Mortgage Market Participants, GAO-07-645 (Washington, D.C.: June 
29, 2007).

[2] GAO, Federal Housing Administration: Proposed Reforms Will Heighten 
the Need for Continued Improvements in Managing Risks and Estimating 
Program Costs, GAO-06-868T (Washington, D.C.: June 20, 2006).

[3] HMDA requires lending institutions to collect and publicly disclose 
information about housing loans and applications for such loans, 
including the loan type and amount, property type, and borrower 
characteristics (such as ethnicity, race, gender, and income). These 
data are one of the most comprehensive sources of information on 
mortgage lending. Among other things, FHA's SFDW contains information 
on the borrower and loan characteristics of the mortgages FHA insures.

[4] Fannie Mae and Freddie Mac are government-sponsored private 
corporations chartered by Congress to provide a continuous flow of 
funds to mortgage lenders and borrowers by purchasing mortgages from 
lenders and re-selling them to investors. They purchase single-family 
mortgages up to the conforming loan limit, which for 2006 was set at 
$417,000.

[5] GAO, Mortgage Financing: HUD Could Realize Additional Benefits from 
Its Mortgage Scorecard, GAO-06-435 (Washington, D.C.: Apr. 13, 2006).

[6] The economic value of the Fund is the value of the Fund's assets 
minus its liabilities, plus the net present value of future cash flows 
of the outstanding portfolio. The capital ratio is the economic value 
divided by the amount of unamortized insurance-in-force (the total 
initial loan amounts of outstanding insured loans). The Omnibus Budget 
Reconciliation Act of 1990 mandated that the Fund achieve a capital 
ratio of at least 2 percent by fiscal year 2000 and maintain that level 
in all future years. See P.L. 101-508, Section 2105.

[7] Subprime borrowers typically have blemished credit, may have 
difficulty providing income documentation, and generally pay higher 
interest rates and fees than prime borrowers.

[8] GAO-07-645.

[9] GAO, Federal Housing Administration: Ability to Manage Risks and 
Program Changes Will Affect Financial Performance, GAO-07-615T 
(Washington, D.C.: Mar. 15, 2007).

[10] According to FHA, the existing loan limits are lower than the cost 
of new construction in many areas of the country and therefore do not 
allow buyers of new homes to use FHA products.

[11] The proposal to repeal the 3 percent minimum cash investment 
requirement would eliminate the complicated statutory formula used to 
calculate down payments. This formula considers multiple variables such 
as the average closing costs in the state.

[12] Under the program, seniors who wished to move from their current 
home could get a home purchase loan for a new dwelling and convert that 
loan into an HECM in a single transaction.

[13] See H.R. 1752, 110th Cong. (2007); H.R. 1852, 110th Cong. (2007); 
S. 947, 110th Cong. (2007).

[14] FHA defines higher-performing lenders as those with 2 years of 
acceptable default and claim rates (at or below 150 percent of the 
national average). Because FHA is phasing in the Lender Insurance 
Program, HECMs are not yet eligible for endorsement through the program.

[15] As was the agency's practice prior to the Lender Insurance 
Program, FHA will select a sample of each lender's mortgages for post- 
endorsement quality checks.

[16] Although the Lender Insurance Program has streamlined the 
processing of FHA-insured loans, the HUD Inspector General has 
expressed concerns that the program could increase the risk of fraud 
because the lenders, rather than FHA, maintain the records on loans 
insured through the program.

[17] Our analysis considered the number of additional loans that would 
have been eligible for FHA insurance if the loan limits in 2005 had 
been raised to 100 percent of area median income, with a floor in low- 
cost areas of $233,773 and a ceiling in high-cost areas of $359,650. 
For our assumptions about the share of newly eligible loans that would 
likely be insured by FHA, see appendix I.

[18] Different pricing would apply to refinances of existing FHA- 
insured mortgages.

[19] Additionally, the vast majority of these borrowers (90 percent) 
received down-payment assistance from nonprofits, most of which 
received funding from property sellers. 

[20] These figures do not reflect FHA's proposals to eliminate the 
limit on the number of mortgages insured under the HECM program and 
move the program from the General Insurance Fund to the Mutual Mortgage 
Insurance Fund. According to FHA's estimates, the HECM program would 
generate about $338 million in negative subsidies in fiscal year 2008. 
Therefore, moving the HECM program would result in negative subsidies 
totaling about $680 million for the Fund.

[21] Agencies are required to reestimate subsidy costs annually to 
reflect actual loan performance and expected changes in estimates of 
future loan performance. Essentially, a cohort includes the loans 
insured in a given year.

[22] These systems contain a wide variety of data that support FHA's 
administration of its single-family mortgage insurance program, 
including information on mortgage lenders and borrowers and the 
financial details and performance of the loans.

[23] More specifically, FHA developed index values--the ratio of the 
claim and prepayment rates for borrowers in different credit score and 
LTV ratio groupings to the claim and prepayment rate for FHA's average 
borrower. (FHA used loans with down-payment assistance from seller- 
funded nonprofit organizations as a proxy for loans with LTV ratios of 
100 percent.) FHA then applied these index values to the estimated 
lifetime claim and prepayment rates for the fiscal year 2008 book of 
business.

[24] See GAO, Mortgage Financing: FHA's $7 Billion Reestimate Reflects 
Higher Claims and Changing Loan Performance Estimates, GAO-05-875 
(Washington, D.C.: Sept. 2, 2005). While loan performance models are 
critical to subsidy cost estimation, other factors such as assumptions 
about the losses per insurance claim and economic conditions also 
influence subsidy estimates.

[25] GAO-06-435.

[26] FHA indicated that variables for adjustable-rate mortgages and 
property type were not included in TOTAL because the risk associated 
with them did not differ significantly in the data sample used to 
develop the model. However, the modeling effort may have failed to find 
significant effects for these variables because of the small numbers of 
loans with these characteristics in the development sample.

[27] These views are consistent with our October 2001 report on FHA's 
single-family information systems. See GAO, Single-Family Housing: 
Current Information Systems Do Not Fully Support the Business Processes 
at HUD's Homeownership Centers, GAO-02-44 (Washington, D.C.: Oct. 24, 
2001).

[28] LMI Government Consulting for the U.S. Department of Housing and 
Urban Development, Strategic Workforce Plan (McLean, Va.: November 
2004).

[29] The analysis defined technical credibility as demonstrating 
programmatic, financial, and technical knowledge and expertise that is 
commensurate with the demands of the position and understanding 
requirements for the administration of federal grants and loan 
guarantees.

[30] GAO, Mortgage Financing: Additional Action Needed to Manage Risks 
of FHA-Insured Loans with Down Payment Assistance, GAO-06-24 
(Washington, D.C.: Nov. 9, 2005).

[31] We reviewed a national sample of FHA-insured home purchase loans 
from 2000, 2001, and 2002.

[32] See 72 Fed. Reg. 27048 (May 11, 2007). FHA also has been 
anticipating a reduction in the number of loans with down-payment 
assistance from seller-funded nonprofit organizations as a result of 
actions taken by the Internal Revenue Service (IRS). IRS issued a 
ruling in May 2006 stating that these organizations do not qualify as 
tax-exempt charities, effectively making loans with such assistance 
ineligible for FHA insurance. According to FHA, as of June 2007, IRS 
had rescinded the charitable status of three of the 185 organizations 
that IRS is examining.

[33] GAO, Mortgage Financing: Actions Needed to Help FHA Manage Risks 
from New Mortgage Loan Products, GAO-05-194 (Washington, D.C.: Feb. 11, 
2005).

[34] Private mortgage insurers also set credit score thresholds for 
zero-down-payment loans.

[35] If the HECM program were moved to the Fund, as FHA has proposed, 
the weighted average subsidy rate would be -0.82 percent.

[36] The MHC, established by Congress in 2000, studied the federal role 
in meeting the nation's housing challenges and issued a report in 2002, 
which included recommendations for a variety of reforms to federal 
housing programs. See Meeting Our Nation's Housing Challenges: Report 
of the Bipartisan Millennial Housing Commission (Washington, D.C.: May 
30, 2002).

[37] In 1989, the Financial Institutions Reform, Recovery and 
Enforcement Act (P. L. 101-73) authorized certain financial regulators 
to determine their own compensation and benefits so that they could 
more effectively compete in the marketplace for qualified applicants. 
In 2002, the Investor and Capital Markets Fee Relief Act (P. L. 107- 
123) gave SEC similar authority as those federal banking regulatory 
agencies. These agencies are permitted by statute to pay salaries in 
excess of the Title 5 ceilings.

[38] GAO-05-194.

[39] FHA officials reported difficulties administering the HECM program 
initially as a demonstration for only 2,500 loans because of the 
challenges of selecting a limited number of lenders and borrowers.

[40] GAO, Homeownership: Potential Effects of Reducing FHA's Insurance 
Coverage for Home Mortgages, GAO/RCED-97-93 (Washington, D.C.: May 1, 
1997).

[41] FHA has implemented risk-sharing arrangements in its multifamily 
insurance program.

[42] In 1995, legislation was introduced in the House and Senate (but 
never enacted) that proposed to abolish FHA and replace FHA's single- 
family mortgage insurance program with a program in which risk would be 
shared between qualified mortgage insurers and a Federal Home Mortgage 
Insurance Fund within the Department of the Treasury. The federal 
government would have provided partial mortgage insurance on some 
single-family homes (and would no longer have insured multifamily 
mortgages).

[43] We excluded second liens and non-owner-occupied properties from 
our analysis because they are not a substantial part of FHA's business. 
As defined by the Office of Management and Budget, CBSAs are 
statistical geographic entities consisting of the county or counties 
associated with at least one core (urbanized area or urban cluster) of 
at least 10,000 population, plus adjacent counties having a high degree 
of social and economic integration with the core.

[44] Our analysis was based on an earlier FHA analysis. This analysis 
assumed that FHA would achieve a market share for newly eligible loans 
of (1) at least 50 percent of FHA's national market share for loans in 
areas with median home prices exceeding the 87 percent conforming limit 
and (2) 75 percent of FHA's current market share in an area that was 
not constrained by the 87 percent conforming loan limit.

[45] See GAO-06-435, GAO-05-875, GAO-05-194, and GAO-06-24.

[46] Meeting Our Nation's Housing Challenges: Report of the Bipartisan 
Millennial Housing Commission (Washington, D.C.: May 30, 2002).

[47] See, for example, Kerry D. Vandell, "FHA Restructuring Proposals: 
Alternatives and Implications," Housing Policy Debate, volume 6, issue 
2 (1995).

[48] See, for example, GAO/RCED-97-93.

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: