Housing Enterprises: Assessment of OFHEO's Interest Rate Analysis (Stmnt.
for the Rec., 07/31/96, GAO/T-GGD-96-173).
GAO discussed the Office of Federal Housing Enterprise Oversight's
(OFHEO) analysis of the potential effects of a large and rapid interest
rate increase on the financial conditions of the Federal National
Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage
Corporation (Freddie Mac). GAO noted that: (1) OFHEO analyzed its own
financial projection model as well as those used by Fannie Mae and
Freddie Mac; (2) the financial models assumed severe financial
conditions, including a sustained interest rate increase, high credit
losses, enterprise debt cost increases, and guarantee fee increases,
over a 10-year stress period; (3) the three financial models differed in
both the assumptions and the relationships embodied in the models, and
time constraints limited OFHEO analysis of the models and their
projections; and (4) although Fannie Mae and Freddie Mac officials
believed that the OFHEO assumptions were not reasonable, GAO believes
that the assumptions, while not necessarily likely or probable, were
reasonable for evaluating the potential effects of severe financial
conditions.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: T-GGD-96-173
TITLE: Housing Enterprises: Assessment of OFHEO's Interest Rate
Analysis
DATE: 07/31/96
SUBJECT: Econometric modeling
Federal aid for housing
Interest rates
Inflation
Financial analysis
Government sponsored enterprises
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Before the Subcommittee on Capital Markets,
Securities and Government Sponsored Enterprises
Committee on Banking and Financial Services
House of Representatives
Not to be Released Before
10:00 a.m. EDT
Wednesday
July 31, 1996
HOUSING ENTERPRISES - ASSESSMENT
OF OFHEO'S INTEREST RATE ANALYSIS
Statement for the Record
James L. Bothwell, Director
Financial Institutions and Markets Issues
General Government Division
GAO/T-GGD-96-173
GAO/GGD-96-173T
(233505)
Abbreviations
=============================================================== ABBREV
OFHEO - x
GAO - x
STATEMENT
============================================================ Chapter 0
Mr. Chairman and Members of the Subcommittee:
On July 5, 1996, the Subcommittee on Capital Markets, Securities and
Government Sponsored Enterprises of the House Committee on Banking
and Financial Services requested that the Office of Federal Housing
Enterprise Oversight (OFHEO) analyze the potential effects of a large
and rapid increase in interest rates, accompanied by otherwise
unfavorable economic conditions, on the financial condition of Fannie
Mae and Freddie Mac (the enterprises). The Subcommittee requested
that we review OFHEO's analysis and assess the reasonableness of the
approach and key assumptions that OFHEO used in its analysis.
OFHEO used two approaches to address the Subcommittee's request.
First, OFHEO asked the enterprises to use their own financial models
to project the impacts of interest rate changes and credit conditions
specified by OFHEO. Second, OFHEO used its financial models, which
are under development to respond to a statutory mandate to establish
risk-based capital standards for the enterprises. This mandate was
contained in the Federal Housing Enterprises Financial Safety and
Soundness Act of 1992 (the 1992 Act). OFHEO analyzed two scenarios
with each approach, one called a "wind down" scenario with no new
business over a period of 10 years and a second scenario with
specified projections of new business.
To monitor and assess OFHEO's analysis, we met extensively with OFHEO
staff. In these meetings, we discussed the assumptions and the
workings of OFHEO's financial models. We also discussed the
enterprises' financial models with both OFHEO and enterprise staff.
We did not verify the data used in nor the structure of the financial
models used by OFHEO, Fannie Mae, or Freddie Mac. To help us judge
the reasonableness of assumptions, we also consulted economic and
financial literature. We did our work in June and July in
Washington, D.C., in accordance with generally accepted government
auditing standards.
Both approaches made projections over a 10-year period and required
the use of key assumptions. The Subcommittee had specified a 600
basis point\1 increase in interest rates on U.S. Treasury securities
over a period of 18 months. OFHEO assumed that after the 18 month
increase in interest rates, these Treasury rates would remain
constant for the remaining 8-1/2 years of the 10-year stress period.
In addition to the sustained 18-month interest rate increase, the
Subcommittee asked that the analysis assume that credit losses were
simultaneously high or severe, as might be the case if the higher
interest rates were not accompanied by an increase in the general
inflation rate. OFHEO specified which particular credit loss
assumptions would be used in the analysis. Generally, the credit
loss assumptions OFHEO supplied to the enterprises specified losses
on existing mortgages purchased by the enterprises that would exhaust
minimum capital, loan loss reserves, and guarantee fee income, on a
present-value basis, over the 10-year stress period. Credit losses
on any new business over the 10-year period were assumed to be
consistent with credit losses on existing business. Both OFHEO and
Freddie Mac officials told us that the loss assumptions specified by
OFHEO would be equivalent to about 30 basis points in credit losses
on purchased mortgages.
OFHEO also directed the enterprises to assume an additional 50 basis
point increase in enterprise debt costs, beginning in the third year
of the stress period, to reflect potential investor concerns about
possible default. In addition, OFHEO specified that annual increases
in guarantee fees, starting in the third year of the stress period,
could not exceed 2 basis points. Under these assumptions, net
interest margins on new enterprise business and the growth of new
business could not be the basis for alleviating or reversing the
projected impacts of the interest rate increase and credit losses on
existing business.
The financial models used by OFHEO, Fannie Mae, and Freddie Mac
differed in both the assumptions and the relationships embodied in
the models. For example, the three models contain different
assumptions about the impact of an interest rate increase on
prepayment rates by borrowers.\2 In addition to differences in the
three sets of financial models used in OFHEO's two approaches, each
approach had its own limitations. The approach that relied on the
enterprises' models was constrained because OFHEO was dealing with
the enterprises' models through the enterprises rather than directly.
This resulted in a significant amount of interchange with enterprise
analysts over a short period of time to establish detailed
assumptions on items such as general and administrative expenses and
net income on short-term investments. OFHEO was limited in the
second approach because its own modeling work for the statutory
stress test, including the normalization of enterprise accounting
data, had not been completed. This exercise was the first
statistical use of OFHEO's financial models.
Enterprise officials stated that, in their view, the combination of
higher interest rates and the credit stress specified by OFHEO in the
first approach was not reasonable. In particular, both enterprises
believed that the increase in interest rates would likely be
accompanied by increased inflation, greater home price appreciation,
and subsequent declines in credit losses. For example, Freddie Mac
officials told us that a credit stress with 10 basis points in credit
losses would be more appropriate for the Subcommittee's request than
that specified by OFHEO, which was equivalent to a 30 basis point
loss. These officials also told us that current losses from credit
risk on Freddie Mac mortgages were equivalent to about 10 to 11 basis
points. Fannie Mae officials told us that the OFHEO-specified
contraction in earnings on new business was circular; the additional
50 basis point increase in borrowing costs was based on a stressful
situation that was not realistic.
We believe, in light of the constraints and limitations discussed
above, that the approach undertaken by OFHEO to comply with the
Subcommittee request was reasonable. We also believe that the key
assumptions on interest rate movements and accompanying credit
losses, while severe, appear reasonable for achieving the purpose of
a stress test. Such tests are designed to estimate the effects of
adverse conditions, which are not necessarily the most probable or
likely conditions.
There are at least two reasons why we believe it was reasonable for
OFHEO to assume that an interest rate increase need not be
accompanied by a general increase in prices. First, past experience
has shown that interest rates can rise by more than the inflation
rate. Second, home prices do not necessarily increase at the general
rate of inflation. In addition, the Subcommittee specifically asked
that the interest rate increase not be assumed to result from a
general increase in inflation.
We also believe that OFHEO's specification of an additional 50 basis
point increase in enterprise debt costs is reasonable because it is
consistent with the experiences of Fannie Mae during the early 1980s
and the Farm Credit System during the mid-1980s when each experienced
financial difficulty.\3 It also appears reasonable for OFHEO to
assume that the enterprises could not increase guarantee fees by more
than 2 basis points annually in response to an increase in credit
losses, because of competition from other providers of residential
mortgage credit.
We believe the analysis requested by the Subcommittee and carried out
by OFHEO should provide useful information about how the enterprises
would fare under the specified assumptions. However, we agree with
OFHEO and enterprise officials that the combined interest rate and
credit stresses specified by OFHEO in this analysis may differ from
the statutory stress tests mandated in the 1992 Act. Once the
results of those tests are available, OFHEO is to develop capital
requirements that are consistent with an acceptable level of taxpayer
risk. OFHEO expects to complete its proposed risk-based capital rule
in Spring 1997.
--------------------
\1 A basis point is one one-hundredth of a percentage point.
\2 A prepayment occurs when the borrower pays off the remaining
principal on a loan before the full term of the loan (e.g., 30 years)
has been reached.
\3 See Government-Sponsored Enterprises: The Government's Exposure
to Risks (GAO/GGD-90-97, Aug. 1990) p. 87.
*** End of document. ***