Homebuyers who finance and homeowners who refinance benefit from the activity of two privately owned but federally chartered corporations, Fannie Mae and Freddie Mac. These government-sponsored enterprises (GSEs) helped establish an efficient secondary market for home mortgages, which in turn has reduced regional differences in mortgage interest rates and lowered mortgage interest rates in general by making mortgages easier to buy and sell.

The GSEs benefit in several ways from having a Federal charter, the most important of which is "agency status," an informal designation that allows Fannie Mae and Freddie Mac to borrow at lower interest rates because the markets perceive lower risks due to the government connection. In the Federal Housing Enterprises Financial Safety and Soundness Act of 1992, Congress imposed certain obligations on the GSEs in exchange for the benefits they receive. In particular, Congress charged the U.S. Department of Housing and Urban Development (HUD) with setting annual goals to ensure that Fannie Mae and Freddie Mac use the benefits received from the Federal Government to help lower income and minority households and that underserved areas receive adequate mortgage financing.¹

HUD set interim goals for 1993 through 1995 and, in December 1995, issued formal goals for the period 1996 through 1999. Early next year, HUD will issue for public comment new proposed goals for the period starting in year 2000. This article presents some of the data that HUD will use and discusses some of the issues HUD will confront in setting those goals.²

Background

The GSEs' purchases accounted for a rising share of the single-family conventional conforming market between 1980 and 1993 (see figure 1). In the early 1980s, the GSEs' share ranged from 12 to 34 percent of the dollar volume of originations in the single-family conforming conventional market. The GSEs' share rose to about 45 percent during the high-volume origination years of 1986 and 1987, and then rose dramatically in the early 1990s to reach a peak of 71 percent during the high refinancing year of 1993. In 1994 the GSEs' share of the conforming conventional market dropped to 55 percent, due primarily to a greater volume of conventional adjustable-rate mortgages as interest rates rose after the first quarter. The GSEs' market share fell further to 43 percent in 1995 and remained between 40 and 50 percent in 1996 and 1997.

Figure 1. GSEs’ Purchases as Share of the Single-Family Conventional Conforming Market, 1980–97 Source: Fannie Mae Economics Department; preliminary estimate for 1997

The housing goals set by HUD for 1996 through 1999 are:

• A low- and moderate-income goal, which targets mortgages on housing for families with less than median income and was set at 40 percent of total units financed by each GSE in 1996 and at 42 percent for each year from 1997 through 1999.

• A geographically targeted goal, which targets mortgages on properties located in "underserved areas," defined as low-income and high-minority census tracts but excluding high-income, high-minority tracts and was set at 21 percent of total units financed by each GSE in 1996 and at 24 percent in each year from 1997 through 1999.

• A special affordable goal, which targets mortgages on housing for low-income families and very low-income families living in low-income neighborhoods and was set at 12 percent of total units financed by each GSE in 1996 and at 14 percent during each year from 1997 through 1999.

• A special affordable multifamily subgoal, for rental units in properties with five or more units that are affordable to very low-income tenants or that are in low-income neighborhoods and affordable to low-income tenants, which was set at $1.29 billion per year for Fannie Mae and $988 million per year for Freddie Mac for each year from 1996 through 1999.

Current Goals and Recent Performance

Both GSEs reported that they exceeded all of the goals for the past 2 years. For example, 45.5 per-cent of Fannie Mae's purchases and 42.9 percent of Freddie Mac's purchases were reported as qualifying for the low- and moderate-income goal in 1997, surpassing HUD's target of 42 percent. The geographically targeted goals were all reportedly surpassed by at least two percentage points in 1996-97 and the special affordable goals were reportedly exceeded by at least one percentage point.

Table 1. Overview of GSEs’ Housing Goal Performance for 1993–97 and Housing Goals for 1996–99 1 Source: HUD analysis of data submitted by the GSEs for 1993–95; as reported to HUD by the GSEs in their Annual Housing Activities Reports for 1996 and 1997 1 Percentages of dwelling units in properties whose mortgages were purchased by the GSEs that would have qualified for each goal in 1993–97 and goals for 1996–99. 2 Abbreviated definitions of goals: Low- and Moderate-Income: Households with income less than or equal to area median income (AMI). Geographically Targeted: Dwelling units in metropolitan census tracts with (1) median family income less than or equal to 90 per-cent of AMI or (2) minority concentration of at least 30 percent and tract median income less than or equal to 120 percent of AMI; dwelling units in nonmetropolitan counties with (1) median family income less than or equal to 95 percent of the greater of State or national nonmetropolitan median income or (2) minority concentration of at least 30 percent and county median income less than or equal to 120 percent of State nonmetropolitan median income. Special Affordable: Households with income (1) less than or equal to 60 percent of AMI or (2) less than or equal to 80 percent of AMI and located in low-income areas. 3 Goals and performance in billions of dollars.

The special affordable multifamily dollar-based subgoals were set at a time when Freddie Mac was just reentering the multifamily business, so they have been surpassed by Fannie Mae but a challenge for Freddie Mac. Freddie Mac's annual multifamily special affordable purchases increased steadily between 1993 and 1997.

For the period as a whole, the GSEs have made gains in the income-targeted goals, with combined special affordable performance doubling between 1993 and 1997 and combined low- and moderate-income purchases increasing from 32 percent in 1993 to 44 percent in 1997. Fannie Mae has consistently surpassed Freddie Mac in goal performance, but Freddie Mac's relative performance has increased -- from 72 percent of Fannie Mae's special affordable record in 1993 to 80 percent in 1997 and from 88 percent of Fannie Mae's low- and moderate-income performance in 1993 to 94 percent in 1997.

Gains have also been made on the geographically targeted goal, with average performance rising by 25 percent between 1993 and 1997. In 1994 Fannie Mae's performance jumped by 6.1 percentage points in that year alone and has since remained at the 1994 level. Freddie Mac has shown more gradual, but relatively steady, gains in this area, though its performance relative to Fannie Mae's was lower in 1997 than in 1993.

Characteristics of GSE Mortgage Purchases

Table 2. Loan and Borrower Characteristics of Single-Family Mortgages Purchased by the GSEs, 1996 and 1997 Source: HUD analysis of GSEs’ loan-level data on mortgages on owner-occupied one-unit properties 1 From Table 9 of Fannie Mae’s and Freddie Mac’s Annual Housing Activities Reports for mortgages on all single-family owner-occupied properties. Percentages refer to home purchase mortgages only.

As housing has become more affordable, first-time homebuyers have played a more significant part in the market. These are typically people in the 25- to 34-year-old age group who purchase modestly priced homes. Although the proportion of the population in this age group decreased from 18 percent in 1985 to 15 percent in 1996, first-time homebuyers increased from about 40 percent of all buyers in the 1980s to 45 percent or more in recent years.⁴ Compared to FHA, the GSEs lag in support for first-time homebuyers. In 1997, 34.5 percent of Fannie Mae's home purchase loans were for first-time homebuy-ers, somewhat above Freddie Mac's first-time home-buyer share (28.4 percent). Both GSEs' shares were well below FHA, where in recent years first-time homebuyers have accounted for approximately three-fourths of its home purchase loans.

Most of the mortgages that the GSEs purchase are originated with high downpayments. As shown in table 2, 68 to 69 percent of the loans acquired by Fannie Mae in 1996-97 had loan-to-value ratios (LTVs) at or below 80 percent. Such low LTV (high downpayment) loans accounted for 70 percent of Freddie Mac's 1997 purchases, though the share of loans with very high downpayments (LTVs at or below 60 percent) was slightly larger for Fannie Mae (19 percent) than for Freddie Mac (18 percent). Since 1995 Freddie Mac has entered the very high LTV (greater than 95 percent) market, though such loans still amounted to only 0.6 percent of its loans in 1997, compared with 2.2 percent of Fannie Mae's loans last year.

Other Factors Considered in Setting Goals

In carrying out its congressional mandate, HUD will consider how the goals match performance elsewhere in the conventional mortgage finance industry. Table 3 uses Home Mortgage Disclosure Act (HMDA) data to compare the affordable lending performance of Fannie Mae and Freddie Mac with conventional lenders. The most appropriate comparisons for the GSEs are with the retained portfolios of depository institutions [banks and savings associations (SAs)] also operating in the conforming market. However, data from FHA are also presented to provide a complete picture of the mortgage market.

Using data from sources other than HMDA, table 3 excludes from the conforming market totals loans by 7 lenders that originate primarily manufactured home loans and loans by 41 lenders that originate primarily below-investment-grade "B&C," quality loans. A longstanding issue concerns whether conventional manufactured home loans should be included in market data that are being compared with GSE data. Because manufactured home loans generally do not meet the GSEs' underwriting standards, Fannie Mae and Freddie Mac contend that manufactured home loans should not be considered in such comparisons. Similar concerns have been raised by the GSEs about B&C loans, which are also included in HMDA data. Because manufactured home and B&C loans are a growing source of affordable lending, the socioeconomic characteristics of these borrowers are provided separately in table 3.⁶

The relevant comparison for the GSEs is with the non-GSE portion of the conventional conforming market, which consists mainly of portfolio lenders, such as banks and SAs. In general, loans sold to the GSEs in 1996 were less likely to be for underserved borrowers and their neighborhoods than were loans retained by portfolio lenders, especially those held by commercial banks. For example, very low-income borrowers accounted for 8.3 percent of loans purchased by the GSEs, compared with 13.5 percent of loans retained by banks. African-American borrowers accounted for 3.6 percent of loans sold to Freddie Mac and 4.4 percent of loans sold to Fannie Mae, compared with 4.9 percent of loans retained by banks. On the other hand, Hispanic borrowers accounted for a much larger share of loans sold to Fannie Mae (6.6 percent) than loans retained by banks (4.8 percent). The neighborhood data in table 3 show similar differentials between the GSEs and banks. Low-income census tracts account for 7.4 percent of loans sold to Freddie Mac and 8.5 percent of loans sold to Fannie Mae, compared with 12.2 percent of loans retained by banks. High-concentration African-American neighborhoods account for 3 percent of Freddie Mac's loans and 3.6 percent of Fannie Mae's loans, compared with 4.8 percent of bank loans.

Table 3. FHA and Conforming Conventional Home Purchase Mortgages by Borrower and Census Tract Characteristics in 1996 Note: Tabulation limited to home purchase mortgages in metropolitan areas. Loans originated by lenders that originate primarily B and C quality or manufactured home loans have been excluded from the conforming market totals. 1 AMI refers to the median income of the metropolitan area. Mortgages with loan amounts less than $15,000 (except for manufac-tured homes) or loan-to-income ratios greater than 6 are excluded. 2 Census tract median income less than or equal to 80 percent of AMI. 3 African-American (Minority) composition of census tract is equal to or greater than 30 percent. 4 Metropolitan census tracts with (1) median income less than or equal to 90 percent of AMI or (2) minority concentration greater than or equal to 30 percent and tract median income less than or equal to 120 percent of AMI. 5 Depository originations (including originations of mortgage company subsidiaries) that were not sold during the calendar year. The abbreviation SAs stands for savings associations.

A recent study by the Federal Reserve Board examined the degree to which different mortgage market institutions -- the GSEs, FHA, depositories, and private mortgage insurers -- are taking on the credit risk associated with funding affordable mortgages.⁷ The Fed study combined downpayment and market share data such as those reported above with data on projected foreclosure rates to estimate the credit risk assumed by each institution for each borrower group. The Fed study found that Fannie Mae and Freddie Mac together provided only 4 to 5 percent of the credit support for lower income and minority borrowers and their neighborhoods. This relatively small role in providing credit support is due to the GSEs' low level of funding for these groups and to the fact that they purchase mainly high down-payment loans.

More Analysis To Come

• National housing needs.

• Economic, housing, and demographic conditions.

• The performance and effort of the GSEs toward achieving the goal in previous years.

• The size of the conventional mortgage market targeted under the goal relative to the size of the overall conventional mortgage market.

• The ability of the GSEs to lead the industry in making credit available for mortgages targeted under the goal.

• The need to maintain the GSEs' sound financial condition.⁸

Notes

1. Congress also established an independent office within HUD, the Office of Federal Housing Enterprise Oversight, with the separate responsibility of determining whether the GSEs have sufficient capital and other resources to meet their financial obligations in unfavorable economic conditions.

2. This article draws heavily on two forthcoming articles from HUD's Housing Finance Working Paper series: Characteristics of Mortgages Purchased by Fannie Mae and Freddie Mac: 1996-97 Update, by Paul B. Manchester; and The GSEs' Funding of Affordable Loans: A 1996 Update, by Harold L. Bunce and Randall M. Scheessele.

3. The analysis in Tables 1 and 2 is based on the loan-level data that the GSEs submit annually to HUD about the mortgages that they purchase.

4. See Who's Buying Homes in America, Chicago Title and Trust Co., 1996, for data on first-time homebuyers.

5. U.S. Department of Housing and Urban Development, Privatization of Fannie Mae and Freddie Mac: Desirability and Feasibility, A HUD Report, 1996.

6. Data on manufactured home and B&C loans are presented separately to highlight characteristics of the borrowers taking out these loans. Our separate treatment of these loans does not necessarily mean that HUD agrees with the GSEs' arguments for excluding these loans from the conforming market definition with analyzing their affordable lending performance. See Randall M. Scheessele, "The 1996 HMDA Data: A Closer Look," U.S. Housing Market Conditions, November 1997, for the source of these data and a discussion of the effects on HMDA data of the recent increase in HMDA reporting by manufacturing home and B&C lenders.

7. Glenn B. Canner, Wayne Passmore, and Brian J. Surrette, "Distribution of Credit Risk Among Provision of Mortgages to Lower Income and Minority Homebuyers," Federal Reserve Bulletin, December 1996.

8. For the special affordable goal, factors 2 and 4 are omitted and data submitted to the Secretary in connection with this goal for previous years are included as a factor for consideration.