Housing Finance: Implications of Alternative Methods of Adjusting the
Conforming Loan Limit (Letter Report, 10/05/94, GAO/RCED-95-6).
The Housing and Community Development Act of 1980 caps the size of
mortgages that can be purchased by either Fannie Mae or Freddie Mac. The
act allows the conforming loan limit to be adjusted annually so that
Fannie Mae and Freddie Mac can respond to changing conditions. For 1994,
the conforming loan limit is $203,150. This report reviews the
methodology used to adjust the conforming loan limit. GAO (1) assesses
the effect on the loan limit of using alternative adjustment methods,
(2) determines the implications of Fannie Mae's and Freddie Mac's
decisions not to adjust the loan limit for 1994, and (3) provides
information on how users of the Finance Board's data view the data's
accuracy.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: RCED-95-6
TITLE: Housing Finance: Implications of Alternative Methods of
Adjusting the Conforming Loan Limit
DATE: 10/05/94
SUBJECT: Mortgage loans
Federal aid for housing
Mortgage programs
Data integrity
Government guaranteed loans
Lending institutions
Demographic data
Economic analysis
Mortgage interest rates
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. *
* *
* A printed copy of this report may be obtained from the GAO Document *
* Distribution Facility by calling (202) 512-6000, by faxing your *
* request to (301) 258-4066, or by writing to P.O. Box 6015, *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time. *
**************************************************************************
Cover
================================================================ COVER
Report to Congressional Requesters
October 1994
HOUSING FINANCE - IMPLICATIONS OF
ALTERNATIVE METHODS OF ADJUSTING
THE CONFORMING LOAN LIMIT
GAO/RCED-95-6
Conforming Loan Limits
Abbreviations
=============================================================== ABBREV
FHA - Federal Housing Administration
GAO - General Accounting Office
HUD - U.S. Department of Housing and Urban Development
NAR - National Association of Realtors
Letter
=============================================================== LETTER
B-257364
October 5, 1994
Congressional Addressees
The Federal National Mortgage Association (Fannie Mae) and the
Federal Home Loan Mortgage Corporation (Freddie Mac)--federally
chartered corporations--channel funds between mortgage lenders and
capital market investors. While these organizations do not originate
mortgage loans, by purchasing mortgages from lenders, Fannie Mae and
Freddie Mac provide liquidity to lenders, thereby making additional
credit available to qualified borrowers.
The Housing and Community Development Act of 1980 requires a limit
(conforming loan limit) on the size of mortgages that can be
purchased by either Fannie Mae or Freddie Mac. Fannie Mae and
Freddie Mac may not purchase mortgages that exceed the conforming
loan limit--called "jumbo" loans. Rather, lenders either hold these
loans in their portfolio or sell them to private investors.\1 For
borrowers, recent studies have found that conforming loans carry
somewhat lower interest rates than jumbo loans. The act provides
that the conforming loan limit be adjusted annually so that Fannie
Mae and Freddie Mac can respond to changing conditions. For 1994,
the conforming loan limit is $203,150.
To adjust the conforming loan limit, Fannie Mae and Freddie Mac are
required to use data on home sales prices published by the Federal
Housing Finance Board (Finance Board) in its "Monthly Interest Rate
Survey." The Finance Board's survey is based on the average price of
homes sold in the last 5 days of the month.\2 To calculate the new
loan limit, the percentage change in the average price of homes sold
is determined using data from the Finance Board's survey for the
month of October versus the previous October. Then, the previous
loan limit is increased by this percentage change. Some critics of
the loan limit have suggested alternatives to this method of
adjusting the conforming loan limit.
This report, mandated by the Housing and Community Development Act of
1992, reviews the methodology used to adjust the conforming loan
limit. Specifically, the report (1) assesses the effect on the loan
limit of using alternative adjustment methods, (2) determines the
implications of Fannie Mae's and Freddie Mac's decisions not to
adjust the loan limit for 1994, and (3) provides information on how
users of the Finance Board's data view the data's accuracy.
--------------------
\1 Lenders can originate both conforming and jumbo loans. According
to data from the Federal Housing Finance Board, in 1993 about 47
percent of jumbo loans were originated by mortgage companies, 32
percent by savings and loan institutions, and about 20 percent by
commercial banks.
\2 Actually, the data are based on the sales price of homes sold that
closed during the last 5 days of the month. For this report, we
refer to homes sold.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
In comparison with the current method of adjusting the loan limit,
alternative methods do not substantially change the resulting 1993
loan limit or the share of the conventional mortgage market that
would be below the conforming loan limit. For three of the four
alternatives we tested, the 1993 conforming loan limit would be
within 7 percent of the actual conforming loan limit. The greatest
difference in the loan limit--over 14 percent--would occur if the
change in home prices, rather than the percentage change in home
prices, is simply added to the previous conforming loan limit. With
regard to the share of the conventional mortgage market that would
fall below the conforming loan limit, 93 percent of all conventional
loans were at or below the actual conforming loan limit in 1993. In
comparison, between 87 and 91 percent of all conventional loans were
at or below the loan limits derived from alternative methods.
Fannie Mae's and Freddie Mac's decision to maintain the same loan
limit in 1994 as in 1993, while the index of home prices declined 3
percent, is authorized by law. This decision should allow both
companies to, at a minimum, serve the same segment of the mortgage
market that they had served the previous year and ease any potential
disruption to lenders. For some borrowers, maintaining the same
conforming loan limit in 1994 could mean lower interest rates
associated with conforming loans.
However, because the law also requires that adjustments to the
conforming loan limit be based upon increases in home prices over a
1-year period, the 1995 loan limit may be adjusted upward without
reflecting the percentage change in home prices that occurred during
the entire 2 years since the limit was last adjusted. This would
result in the loan limit no longer following the long-term pattern of
growth in home prices, thus imparting a permanent upward bias in the
conforming loan limit. Furthermore, should similar circumstances
arise in the future, the conforming loan limit would be further
biased upward.
While for some borrowers, a higher loan limit could provide the
benefits of the lower interest rate associated with conforming loans,
the permanent upward bias in the loan limit could increase the number
of loans that would be eligible for Freddie Mac and Fannie Mae to
purchase, while reducing the number of jumbo loans. This would
particularly affect those lenders that specialize in originating and
holding jumbo loans. Similarly, an upward bias in the limits for
home loans insured by the Federal Housing Administration (FHA) may
occur should the FHA loan limits be indexed to the conforming loan
limit, as has been proposed in recent legislation.
Finally, from a national perspective, users of the Finance Board's
survey find the data to be generally accurate. Some users did
question the accuracy of data for local areas--which is used for
purposes other than setting the national conforming loan limit.
Furthermore, the Finance Board's data remain as the only
comprehensive source of national data on housing price changes for
both new and existing homes.\3
--------------------
\3 Home price data on existing homes are available from the National
Association of Realtors, and data on new homes are available from the
Bureau of the Census' Current Construction Reports.
BACKGROUND
------------------------------------------------------------ Letter :2
The conforming loan limit is a legislative restriction on the size of
loans that Fannie Mae and Freddie Mac may purchase from lenders.
Specifically, the Housing and Community Development Act of 1980
requires that the maximum loan limit be adjusted annually by a
percentage equal to the percentage increase in the national average
price of houses as measured by the Finance Board. The legislation
also specifies that the time period for which an increase in average
home prices is measured is the 12-month period ending with the
previous October. This adjustment mechanism was provided to allow
Fannie Mae and Freddie Mac the capacity to respond to changing
conditions over time--presumably including the changing price of
homes sold over time.
Between 1980 and 1993, the conforming loan limit rose from $93,750 to
$203,150, as Fannie Mae and Freddie Mac adjusted the loan limit
annually on the basis of the Finance Board's index. For all but one
year, the loan limit increased--by as little as less than 1 percent
to as much as 15.6 percent. The loan limit declined by one-tenth of
1 percent in 1990--the only year the loan limit has declined. In
November 1993, the Finance Board reported that the national average
price of homes sold in October 1993 was 3 percent lower than the
national average price of homes sold the previous October. However,
because of concerns over the potential impact lowering the limit may
have on home buyers and lenders, and because the act only specifies
that the conforming loan limit be increased according to the Finance
Board's index, Fannie Mae and Freddie Mac decided not to change the
loan limit for 1994.
ALTERNATIVE METHODS FOR
ADJUSTING THE CONFORMING LOAN
LIMIT YIELD SIMILAR RESULTS
------------------------------------------------------------ Letter :3
A 1990 study prepared for the Department of Housing and Urban
Development (HUD) assessed several criticisms of the current method
for setting the conforming loan limit.\4
Among these criticisms were that the current method results in the
limit (1) being volatile from year to year, (2) rising more rapidly
than home prices, (3) not reflecting regional differences in home
prices, and (4) not accounting for the changing quality of homes
sold.\5
However, according to the Finance Board, its survey remains the only
comprehensive home price data for both new and existing homes. Also,
despite the volatility of this index, over the long term, the
conforming loan limit is in line with other house price indexes,
according to the Congressional Research Service.\6 The 1990 HUD study
also recognized this fact. Finally, the 1990 study also found that
adjusting limits for regional differences in home prices and
accounting for the changing quality of homes require more detailed
data than are now available.
While some critics of the current method believe that the limit has
risen more rapidly than home prices, the alternative methods that
they suggest would result in similar limits for 1993. In addition,
while one might expect that an inflated conforming loan limit would,
over time, result in proportionately more loans falling under the
limit, the share of loans under the limit has been fairly stable for
the past 13 years.
To test the effect of using alternative methods for setting the
conforming loan limit, we compared the actual 1993 conforming loan
limit with the loan limits derived from alternative methods contained
in the 1990 study. These alternative methods included using the same
Finance Board data in different ways--median and 3-month averages of
home sales prices and a simple addition of the absolute change in
home prices--and using alternative data from the National Association
of Realtors (NAR) and the Department of Commerce's Bureau of the
Census.\7 For all alternatives, we estimated the 1993 loan limit
using data for all years from 1980 through 1992. Using the median or
3-month averages of the Finance Board's home price data does not
significantly affect the amount of the conforming loan limit--between
$192,800 and $196,700 versus $203,150. The substitution of
alternative data, such as from the Bureau of the Census and the NAR,
for the Finance Board's data results in a somewhat larger change but
still is within 7 percent of the actual conforming loan limit. As
expected, the greatest difference in the loan limit--over 14
percent--would occur if the average change in home prices is simply
added to the previous conforming loan limit. (See fig. 1.)
Figure 1: 1993 Conforming Loan
Limit--Actual and Using
Alternate Methods
(See figure in printed
edition.)
\a October, November, and December.
\b August, September, and October.
Source: GAO's analysis of data provided by the Finance Board, Bureau
of the Census, and NAR.
The effect on the share of conventional loans below the loan
limit--and therefore eligible for purchase by Fannie Mae and Freddie
Mac--was relatively small for each of the alternative methods we
tested. For example, while about 93 percent of conventional loans
were below the conforming loan limit in 1993, 91 percent of
conventional loans were below the loan limits that would be set if
the median home price or a 3-month average method of adjusting the
loan limit were used instead. Even using alternative data from the
NAR and the Census Bureau results in about 90 percent of conventional
loans falling under the alternate loan limit in 1993. Again, the
greatest effect on the share of conventional loans below the loan
limit occurs if the loan limit is set by simply adding the average
change in home prices to the previous limit. (See fig. 2.)
Figure 2: 1993 Share of All
Conventional Loans at or Below
the Actual and Alternative
Conforming Loan Limits
(See figure in printed
edition.)
\a October, November, and December.
\b August, September, and October.
Source: GAO's analysis of data provided by the Finance Board, Bureau
of the Census, and NAR.
To assess the criticism that the Finance Board's index had risen
faster than actual home prices, we reviewed the share of conventional
loans that were below the conforming loan limit for the 1980 through
1993 period. If the conforming loan limit were rising faster than
the overall value of homes purchased with conventional mortgages,
then a growing share of the loans originated would fall below the
limit each year. The result would be an increase in the conforming
share of the market. In fact, we found that the share of
conventional loans below the conforming loan limit has been
relatively stable between 1980 and 1993. Specifically, the share of
loans below the conforming loan limit averaged about 91 percent for
the period and ranged from a low of 88 percent to a high of 94
percent. During the same period, the conforming loan limit rose from
$93,750 in 1980 to $203,150 in 1993. (See fig. 3.)
Figure 3: Conforming Loan
Limit and Share of Conventional
Loans Below Loan Limit (1980
Through 1993)
(See figure in printed
edition.)
Source: GAO's analysis of data provided by the Finance Board.
--------------------
\4 Effects of the Conforming Loan Limit on Mortgage Markets (Mar.
1990), prepared for HUD by ICF, Inc.
\5 The survey used to adjust the conforming loan limit includes data
on fully amortized, purchase money, conventional, first mortgage
loans. The survey does not include data on balloon loans,
refinancings, or FHA/Veterans Administration loans. Also, the survey
does not include data on some other loans, such as those secured by
structures with more than one unit.
\6 Housing Finance Debates: The "Conforming Loan" Limits of FNMA and
FHLMC, Congressional Research Service, (IB 87094, updated Jan.
1988).
\7 Freddie Mac advocates the use of an alternative methodology that
would be based upon an index of weighted repeat sales. Such an index
measures the sales price of the same homes over time. Fannie Mae
advocates the use of an index of transaction prices for adjusting the
loan limit, as is done now. In June 1994, Freddie Mac and Fannie Mae
introduced an index of weighted repeat sales, combining data from
both companies. According to Freddie Mac, this is the only such
index prepared on a national level but does not include data on new
homes sold. For this report, we did not compute the conforming loan
limit using a repeat-sales index. Rather, we limited our analysis to
those alternatives found in HUD's 1990 study.
IMPLICATIONS OF NOT ADJUSTING
DOWNWARD THE 1994 LOAN LIMIT
------------------------------------------------------------ Letter :4
For 1981 through 1993, Fannie Mae and Freddie Mac adjusted the
conforming loan limit annually according to the Finance Board's
index. For all but one year, the index caused the conforming loan
limit to increase--by as little as less than 1 percent and as much as
15.6 percent. The loan limit declined by one-tenth of 1 percent in
1990--the only year the loan limit has declined. In November 1993,
the Finance Board reported that the price of homes sold in October
1993 was 3 percent lower than the price of homes sold the previous
October. However, Fannie Mae and Freddie Mac decided for the first
time to make no changes to the limit for 1994. Among the reasons
cited by Fannie Mae and/or Freddie Mac were that (1) the act only
specifies that the conforming loan limit be increased according to
the Finance Board's index, (2) it was not clear that there was a real
decline in house prices in 1993, and (3) there was the need to
provide stability in the secondary market.
The Housing and Community Development Act of 1980 provides that the
conforming loan limit be adjusted annually using data on home prices
from the Finance Board. Specifically, in order to provide Fannie Mae
and Freddie Mac with the capacity to respond to changing conditions
over time, the act requires that the maximum loan limit be adjusted
each year by a percentage equal to the percentage increase in the
national average price of houses as measured by the Finance Board.\8
The act also specifies that the time period for which an increase in
average home prices is measured is the 12-month period ending with
the previous October. However, the act is silent on adjusting the
loan limit when there is a decline in the average sales price of
homes. Neither does the act's history explain whether Fannie Mae and
Freddie Mac are required to make adjustments to the loan limit when
the average price of homes declines. Like the act, its legislative
history speaks of adjusting the maximum allowable loan limit by
adding to the existing limit a percentage equal to the percentage of
increase, during the 12-month period ending with the previous
October, in the national average home price as measured by the
Finance Board. Accordingly, in this instance, the act does not
require Fannie Mae and Freddie Mac to lower the loan limit when the
price declines.
We do not address here whether or not Fannie Mae and Freddie Mac have
authority to maintain the conforming loan limit regardless of the
extent or duration of housing price declines. According to the
Senate report that accompanied the legislation which originally
instituted maximum loan limits, the purposes of the conforming loan
limit are "to reduce risk to [Fannie Mae and Freddie Mac] and to
encourage the flow of mortgage credit to low- and moderate-priced
housing." In the event housing prices declined drastically or
declined continuously over a period of years, a decision by Fannie
Mae and Freddie Mac to maintain the conforming loan limit at a level
equal to its highest level could contravene these purposes. It might
also result in an inappropriate increase in the share of the
secondary market held by the two organizations.
Fannie Mae and Freddie Mac also cited other reasons for their
decision--including a concern that a reduction in the limit would
hurt many middle-class home buyers, especially in high-cost markets
such as California. These borrowers would either have to come up
with a larger downpayment or seek a jumbo loan, which recent studies
find carry higher interest rates. Both Fannie Mae and Freddie Mac
also believe that a reduction in the loan limit would impose
operational burdens on lenders, as they would have to operate with
two different sets of loan limits for a period of time and have
controls in place to ensure that mortgages originated at the previous
higher limits are delivered within specified deadlines. For lenders
with insufficient volumes to make jumbo loans, a reduction in the
loan limit could mean a direct loss of business, according to Fannie
Mae.
In terms of the number of loans affected, we found that in 1993, only
about 2.4 percent of conventional loans were for amounts that were
within 3 percent below the 1993 loan limit--the equivalent of the
reduction in the limit that might have been made for 1994 based on
the reduction in the average sales price. The impact on the dollar
volume of loans is greater. For example, Fannie Mae estimates that
about 6 percent of its and Freddie Mac's total 1993 business was
within 3 percent below the conforming loan limit. Also, Fannie Mae
reports that borrowers in central cities and high-cost areas, such as
California, as well as minorities would be disproportionately
affected by a reduction in the conforming loan limit.\9 For example,
Fannie Mae estimates that more than 23 percent of the families
affected are in central cities, and 8 percent of the entire
California market would be affected. Also, 17 percent of the loans
in this range were made to minorities in 1993.\10
While not lowering the loan limit in 1994 is allowed by the law, and
should ease disruption to certain borrowers and lenders, the method
by which subsequent adjustments to the loan limit are made could
result in a continuing impact on the share of conventional mortgages
below the conforming loan limit. That is, should the loan limit be
adjusted upward in 1995 to reflect an increase in the average price
of homes sold between 1993 and 1994 as currently required by law,
regardless of whether there was an increase in the average price of
homes sold between 1992 and 1994, the resulting loan limit will be
biased upward, and a greater proportion of conventional loans would
fall below the conforming loan limit. For example, assuming that
average house prices rise by 5 percent this year, the 1995 conforming
loan limit would increase by 5 percent if only one year's data were
used as a basis for adjusting the limit. However, the 1995 limit
would increase by only 2 percent if two year's data were
used--reflecting the last point in time that the loan limit was
adjusted. Thus, the 1995 loan limit would be 3 percent higher than
the loan limit if the adjustment period reflected the entire period
since the loan limit was last adjusted--2 years. Consequently, we
believe that a greater number of conventional loans would fall under
this loan limit in 1995 and in all subsequent years. That is, the
number of conventional loans that Fannie Mae and Freddie Mac could
purchase could be greater, while the number of conventional loans
above the loan limit--jumbo loans--could be lower than otherwise
might be the case. Furthermore, if similar circumstances arose in
the future, the gap between what could be the loan limit would widen.
For some borrowers, a higher conforming loan limit would bring the
benefit of lower interest rates associated with conforming loans.
Finally, recently proposed legislation would provide for indexing the
loan limits for loans insured by FHA to the conforming loan limit.
Should the FHA limits be so indexed, any upward bias in the
conforming loan limit would result in a similar upward bias in the
FHA loan limits.
--------------------
\8 Freddie Mac commented that it and Fannie Mae have the ability to
set a loan limit below the maximum allowable loan limit.
Historically, however, both have adjusted the limit according to the
percentage increase in the Finance Board's index.
\9 Every metropolitan statistical area has at least one central city,
which is usually its largest city.
\10 According to HUD, about 25 percent of Fannie Mae's single-family
1993 purchases were on properties located in central cities. Thus,
the proportion of central city borrowers that might be affected by a
decline in the loan limit is roughly equal to the proportion of
central city borrowers that Fannie Mae serves.
COMMENTS FROM USERS ON THE
RELIABILITY OF THE FINANCE
BOARD'S DATA ON AVERAGE HOME
PRICES
------------------------------------------------------------ Letter :5
In response to the Finance Board's May 1990 request for comments on
proposed changes to its Monthly Interest Rate Survey, respondents
typically did not question how accurately data were input or how
reliably the data were processed. Some respondents did question the
accuracy of local data that are used for purposes other than
adjusting the conforming loan limit. Specifically, the respondents
most often suggested that the sample size be expanded so that
regional, state, and local data would be more reliable. Some
respondents suggested alternative methods of increasing sample size,
such as using more days of the month or the last days of the month.
Other respondents suggested adjustments to better reflect the mix of
lenders reporting and the geographic and size mix of homes included
in the sample.
In response to respondents' concerns, the Finance Board made several
technical modifications to its method for calculating the average
home sales price. To increase the sample size, in November 1991, the
Finance Board started using data for the last 5 days of the month,
instead of the first 5 days of the month, because more loans are
closed at the end of the month than at the beginning. In addition,
the Finance Board, in January 1992, implemented a new weighting
scheme that, according to the Finance Board, would allow for the
share of mortgages represented in the survey for each type of lender
to comport with aggregate lending patterns.\11 The principal effect
of this change, according to the Finance Board, is to increase the
statistical importance of loans originated by mortgage companies and
commercial banks and decrease the statistical importance of loans
originated by savings and loan associations.
While these changes address some of the respondents' concerns, we
have not evaluated their statistical significance. Regardless, the
Mortgage Interest Rate Survey remains the only comprehensive source
of home price data for both new and existing homes.
--------------------
\11 The Finance Board uses data from HUD's Survey of Mortgage Lending
Activity to adjust the weights it gives to different types of
lenders. However, the Finance Board questions the representation of
commercial lenders in the survey. Currently, HUD is evaluating the
reliability of its survey.
CONCLUSIONS
------------------------------------------------------------ Letter :6
Alternative methods of adjusting the conforming loan limit have
little effect on what the loan limit would be and the share of
conventional loans that would be conforming. Also, while the data
used to adjust the limit do not include data on all house sales, they
remain the only source for national data on both new and existing
homes.
The Housing and Community Development Act of 1980 provides that the
conforming loan limit be adjusted annually to reflect the percentage
increase in the national average price of homes purchased. The act
also specifically requires that adjustments be based on the 12-month
period preceding the adjustment. Given that Fannie Mae and Freddie
Mac decided not to adjust the conforming loan limit for 1994,
adjusting the limit next year on the basis of a 1-year increase in
the average home prices between 1993 and 1994 as currently required
by law will introduce an upward bias in the conforming loan limit.
This upward bias could result in a greater proportion of conventional
loans falling below the loan limit and being available for purchase
by Fannie Mae and Freddie Mac. Conversely, such an increase in the
loan limit could result in fewer jumbo loans, which would
particularly affect lenders that specialize in originating and
holding such loans. Moreover, should the index of home prices
decline again in the future, and subsequent adjustments be based on
only a 1-year change in home prices, the resulting loan limit would
be further biased upward. Each such event would further increase the
share of the mortgage market in which Fannie Mae and Freddie Mac can
operate. For borrowers that would have to obtain a jumbo loan if the
conforming loan limit was not increased, a higher limit could provide
the benefit of the lower interest rates associated with conforming
loans.
MATTER FOR CONGRESSIONAL
CONSIDERATION
------------------------------------------------------------ Letter :7
If the Congress intends that the conforming loan limit follow the
long-term pattern of growth in average home prices, it should amend
the legislation to require that adjustments be made on the basis of
the time period since the limit had last been changed rather than the
12-month period preceding the adjustment, as mandated now. For
example, given that the loan limit was not adjusted for 1994, the
loan limit for 1995 should be based upon the change in the average
home price between October 1992 and October 1994 rather than the
change in price between October 1993 and October 1994.
AGENCY COMMENTS AND OUR
EVALUATION
------------------------------------------------------------ Letter :8
We received written comments from the Finance Board, HUD, Fannie Mae,
Freddie Mac, and the Mortgage Bankers Association of America (see
apps. I through V). In addition, the Savings and Community Bankers
of America opted to provide oral comments. Overall, the Finance
Board, HUD, and the Savings and Community Bankers of America
generally agreed with our conclusions; Fannie Mae, Freddie Mac, and
the Mortgage Bankers Association agreed with our finding that Fannie
Mae's and Freddie Mac's decision to maintain the same loan limit in
1994 is authorized by the statute.
Three organizations commented on our matter for congressional
consideration. HUD agreed that the matter for congressional
consideration could eliminate the potential upward bias in the loan
limit described in the report. HUD also suggested an alternative
method that uses a cumulative index to remedy the upward bias. The
official from the Savings and Community Bankers of America said that
the loan limit should adjust downward as well as upward, but agreed
that our matter for congressional consideration, over the long term,
would result in the same loan limit. The Mortgage Bankers
Association said that the matter for congressional consideration was
inappropriate because it appeared to be based on the belief that
thrifts would be adversely impacted, that the limit should not be
used as a market allocation tool, and that the adverse affect is
overstated. Freddie Mac also said that the draft incorrectly stated
that thrifts would be particularly affected by loan limit increases
and that the loan limit was not intended to be a market allocation
device.
In response to HUD's suggestion of using a cumulative index, we agree
with HUD's assessment that it would produce outcomes identical to
those obtained with the procedure described in our matter for
congressional consideration. HUD officials described this cumulative
index as an index that would be set at 100 to correspond with the
Finance Board's data for October 1992. Such a cumulative index would
be increased by the percentage increase in the Finance Board's index
from October 1992, but in years when the Finance Board's index had an
annual decline, the cumulative index would remain at its previous
level. In effect, this suggestion is one way to adjust the
conforming loan limit consistent with our matter for congressional
consideration--on the basis of the time period since the loan limit
had last been adjusted.
With regard to the Savings and Community Bankers' suggestion of
following the Finance Board's index regardless of whether the index
rises or falls, while we agree that this would allow the loan limit
to follow the long-term pattern of growth in average home prices, it
might impose operational burdens on lenders when the loan limit
declines, as cited by Fannie Mae and Freddie Mac. Also, given that
declines in the Finance Board's index have been infrequent and that
the law is silent on declines in the index, we believe that our
matter for congressional consideration would not only ensure that the
limit follows the long-term pattern of home price appreciation but
also would alleviate any short-term disruption to lenders and
borrowers.
In response to the comments concerning the impact on thrifts, we have
deleted the specific reference to savings institutions' being
particularly affected by an increase in the conforming loan limit and
refer instead to those lenders that specialize in originating and
holding jumbo loans. In addition, we have added data to show that
mortgage companies originated 47 percent of the dollar volume of
jumbo loans in 1993 and savings and loan institutions originated
about 32 percent. In addition, we recognize that lenders may
originate both jumbo and conforming loans.
With regard to concerns that the loan limit not be used as a tool to
allocate market share, we note that the loan limit does, in effect,
define the market in which Fannie Mae and Freddie Mac may operate.
However, the basis for our matter for congressional consideration is
not to employ a market allocation tool but continues to be solely the
desire for the conforming loan limit to follow the long-term pattern
of changes in average home prices.
In its comments, Freddie Mac suggested that we consider whether it is
appropriate for the Finance Board to administer the survey or index
used to determine the maximum conforming loan limit because the
Finance Board is an advocate for the Federal Home Loan Banks and
suggested that other agencies perform this function. We did not
address the independence of the Finance Board because this was not
part of our mandate. In addition, while we did not assess the
reliability of the Finance Board's data, in reviewing comments to the
Finance Board on its proposed changes to the index, we found no
indication that users of the data believed that the Finance Board was
in any way manipulating the data to the advantage of the Federal Home
Loan Banks.
Finally, in response to these and other comments, we have added
information about the recently announced Freddie Mac/Fannie Mae home
price appreciation index, clarified our description of how the limit
is adjusted, added data on who particularly originates jumbo loans,
and added further detail of what data are included in the Finance
Board's index. Where appropriate, we have incorporated other
suggested clarifications to the report.
SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :9
To assess the methodology used to adjust the conforming loan limit,
we examined both the effect that using alternative methods of
adjusting the loan limit would have on the limit and market share and
the historical share of conventional loans that are below the
conforming loan limit. We limited the alternative methods we used
for comparison to those that were previously reported in the 1990
study on loan limits prepared for HUD. In comparing the current
method for determining the conforming loan limit with alternate
methods, we used the Federal Housing Finance Board's data for 1979
through 1992, as well as Census data on new home prices and NAR's
data on existing home prices. To assess the requirements for
adjusting the conforming loan limit, we reviewed the legislative
history of conforming loan limits and obtained the views of Fannie
Mae and Freddie Mac officials. We examined comments received by the
Finance Board in response to its 1990 request for comment on
improving the Monthly Interest Rate Survey. We did not assess the
reliability of the Federal Housing Finance Board's data. Our work
was conducted between April and September 1994 in accordance with
generally accepted government auditing standards.
---------------------------------------------------------- Letter :9.1
We are providing copies of this report to the Secretary of Housing
and Urban Development, the Director of the Office of Management and
Budget, and other interested parties. We will also make copies
available to others upon request. In response to the mandate, we
have also provided in a separate report information on how the
income, age, and race of borrowers of FHA-insured loans and the
location of their homes has changed since the 1970s.\12
Please contact me on (202) 512-7631 if you or your staff have any
questions about this report. Major contributors to this report are
listed in appendix VI.
Sincerely yours,
Judy A. England-Joseph
Director, Housing and Community
Development Issues
List of Addressees
The Honorable Paul S. Sarbanes
Chairman
The Honorable Christopher S. Bond
Ranking Minority Member
Subcommittee on Housing and
Urban Affairs
Committee on Banking, Housing,
and Urban Affairs
United States Senate
The Honorable Barbara A. Mikulski
Chair
The Honorable Phil Gramm
Ranking Minority Member
Subcommittee on VA, HUD,
and Independent Agencies
Committee on Appropriations
United States Senate
The Honorable Henry B. Gonzalez
Chairman
The Honorable Marge Roukema
Ranking Minority Member
Subcommittee on Housing and
Community Development
Committee on Banking, Finance and
Urban Affairs
House of Representatives
The Honorable Louis Stokes
Chairman
The Honorable Jerry Lewis
Ranking Minority Member
Subcommittee on VA, HUD,
and Independent Agencies
Committee on Appropriations
House of Representatives
(See figure in printed edition.)Appendix I
--------------------
\12 Housing Finance: Characteristics of Borrowers of FHA-Insured
Mortgages (GAO/RCED-94-135BR, Apr. 6, 1994).
COMMENTS FROM THE FEDERAL HOUSING
FINANCE BOARD
============================================================== Letter
Now footnote 11.
(See figure in printed edition.)
(See figure in printed edition.)COMMENTS FROM THE FEDERAL HOUSING
FINANCE BOARD
(See figure in printed edition.)
(See figure in printed edition.)Appendix II
COMMENTS FROM THE U.S. DEPARTMENT
OF HOUSING AND URBAN DEVELOPMENT
============================================================== Letter
(See figure in printed edition.)
(See figure in printed edition.)Appendix III
COMMENTS FROM FANNIE MAE
============================================================== Letter
(See figure in printed edition.)Appendix IV
COMMENTS FROM FREDDIE MAC
============================================================== Letter
(See figure in printed edition.)
(See figure in printed edition.)
(See figure in printed edition.)
(See figure in printed edition.)Appendix V
COMMENTS FROM MORTGAGE BANKERS
ASSOCIATION OF AMERICA
============================================================== Letter
(See figure in printed edition.)
Now on p. 11.
(See figure in printed edition.)
MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix VI
HOUSING AND COMMUNITY
DEVELOPMENT ISSUE AREA
-------------------------------------------------------- Appendix VI:1
Jacquelyn Williams-Bridgers
Robert Procaccini
Mathew J. Scire
John McGrail
C. Bernard Myers