Setting Affordable Housing Performance Goals That Are Meaningful for America’s Low-Income Neighborhoods
On February 16, 1995, the Department of Housing and Urban Development (HUD) published an extensive set of proposed regulations revising the affordable housing performance goals for Fannie Mae and Freddie Mac. The challenge for HUD, which oversees these two government sponsored enterprises (GSEs), was to set targets that were both realistic and appropriate, given the sheer size of the GSEs, their mission, and how they operate. HUD is now seeking comments from the public and especially from affected communities on the appropriateness of the goals.
An understanding of the new performance measures depends, in part, upon an understanding of how the GSEs operate and the nature of the special public benefits they receive and their corresponding statutory requirements. The GSEs are part-private, part-public corporations. As with any financial institution, the GSEs make their profits by lending at higher rates than they pay. The greater the spread, the greater the profits for Fannie and Freddie and their private shareholders. As government-sponsored corporations, the two GSEs enjoy a direct line of credit to the U.S. Treasury, exemption from securities regulations and all State and most local taxes, and the triple-A investment rating of government institutions. This quasi-governmental status enables them to raise capital at interest rates significantly below those paid by other financial institutions. In exchange for their unique status, these corporations were chartered to “support the secondary market for residential mortgages (including activities relating to mortgages on housing for low- and moderate-income families); and promote access to mortgage credit throughout the Nation.”
Some of the federal government’s initial purposes for establishing Fannie Mae and Freddie Mac certainly have been achieved. The two housing GSEs have helped to standardize and liquefy the housing finance markets. In many respects, the GSEs are sterling examples of government sponsored initiatives that have worked. Many home buyers have benefited from lower interest rates (estimated to be 1Ú4 to 1Ú2 of a percentage point lower than they otherwise would be) and through increased access to housing finance credit provided as a result of the GSEs. The GSEs are transforming the way this country finances its housing: no longer do local banks take local deposits and reinvest them through mortgages in the local housing stock. Most lenders now make loans in order to “sell” them to the GSEs or the “secondary market” and keep a small fee to cover administrative costs and profit.
But while the GSEs have increased mortgage credit availability in general, they have been much less successful in serving low-income and traditionally underserved housing markets. The GSEs have been making record profits for their shareholders but have failed to even keep pace with the private mortgage market in serving the nation’s low-income and predominantly minority communities. As a result, Congress enacted legislation in 1992 directing HUD to establish specific interim performance goals, for 1993 and 1994, to stimulate the GSEs’ role in mortgage lending for affordable housing. HUD was also to provide revised goals for 1995, 1996, and beyond. The affordable housing provisions were strongly supported by national housing advocates, community organizations, and community development intermediaries such as the National Low Income Housing Coalition, National Peoples’ Action, ACORN, Center for Community Change, the Housing Assistance Council, LISC, and the Enterprise Foundation.
The underlying purpose of the regulations proposed by HUD is to ensure that these corporate giants continue to produce clear public benefits. Despite their dominance in the housing finance market, the country still suffers from pervasive disparities in the access to credit for racial and ethnic minorities and residents of low-income inner city and rural areas. Furthermore, the country faces a huge shortfall in its supply of affordable housing. These needs form the backdrop for HUD’s proposed goals for the GSEs.
Table 1: Conventional Housing Mortgages For Low- And Moderate-Income Families
| Actual 1993 Performance | Proposed Goals | Estimated Market | ||||||
| Fannie | Freddie | 1995 | 1996 | |||||
| Owner-Occupied | 17% | |||||||
| 1-4 Unit Rental | 12% | |||||||
| Total single-family | 31.8% | 28.9% | 29% | |||||
| Multifamily | 3.8% | 0.3% | 21% | |||||
| Total Goal | 35.6% | 29.2% | 38.0% | 40.0% | 50 – 55% | |||
HUD’s Goals for the GSEs
After conducting considerable research to estimate the size of the potential market and the GSEs’ past performance, HUD has proposed three categories in its goals for the GSEs: mortgages for low- and moderate-income families; areas traditionally under-served by housing finance (low-income, high minority population urban and suburban areas, rural areas); and a “Special Affordable Housing” percentage goal for people with very low incomes. Purchases in each category are counted as a percentage of total activity for the GSEs based on the estimated market in each category.
Importantly, a single mortgage can count for all three of these goals. For example, a mortgage that finances a house for a low-income family in an underserved area counts under the special affordable housing goal, the low- and moderate-income housing goal, and the underserved areas goal. Under the GSEs’ performance for 1993, the majority of mortgages that qualified for one goal also qualified for a second goal.
The Center for Community Change and the National Low Income Housing Coalition have carefully examined the goals. While we commend HUD for the thorough and extensive efforts made to establish effective goals, we have three main concerns with HUD’s proposals.
Two-for-One Overlapping Goals
First is the degree to which HUD has allowed the goals to overlap. For example, the proposed low- and moderate-income housing goal allows purchases of mortgages for single-family home ownership, single-family rental, and multifamily rental housing (more than four dwelling units) to be counted towards the same numeric goal. Thus, the GSEs could satisfy their obligation to the multifamily rental housing market through merely an increased share in low-income homeownership activity. In fact, HUD’s analysis of the GSEs’ previous activities suggests that they met the general 1993 Low- and Moderate-Income Housing Benefit goal mainly through lending for single-family housing (see Table 1). Because of the important differences in these housing types and the severity of the need for affordable rental housing, HUD should re-specify the goals with distinct subgoals for priority needs like rental housing and very low income housing.
Similarly, HUD should separate its goals for urban and rural areas to ensure that the needs of each very distinct geographic area is served by the GSEs. As for the special affordable housing goal, the proposal allows the GSEs to fulfill the rental portion of that goal entirely through the purchase of single-family (1-4 unit) rental housing loans and not necessarily through the purchase of mortgages for larger, multifamily rental housing. A separate subgoal in this area is needed as well.
Apparently, HUD’s reluctance to establish subgoals stems from the lack of authority the agency believes it has to enforce the subgoal requirement. But we believe that such subgoals would help spur increased performance by the GSEs in these areas of greatest need.
Leading the Market
The second concern with the goals is that they do not ensure that the GSEs will “lead the market” in the production of affordable housing and housing in underserved areas. For example, HUD estimates that the notoriously under-performing private financial market originates 50 percent of its mortgages on housing affordable to low- and moderate-income families. However, the 1996 Low- and Moderate-Income goal for the GSEs is only 40 percent of the existing market. Fannie Mae will purchase over $170 billion in loans in 1995, but the amount of additional purchases required for it to meet the targets will be less than $1.5 billion. For Freddie, the volume projections are $130 billion, and the additional purchase requirements would be approximately $6 billion. The proposed goals are clearly not commensurate with the scope of the problem and the massive resources available to the GSEs.
Minority Community Access to Housing Finance
Lastly, the GSEs could play a far greater role in ensuring increased access to housing finance for members of minority communities. In 1993, the GSEs purchased 70 percent of all single-family mortgages, yet they have under performed in the market in serving minority households. In that same year, just 2.9 percent of all mortgages nationally went to African-Americans. According to HUD Secretary Cisneros, “In 1993, mortgages to African Americans accounted for only 2.3 percent of Fannie Mae’s purchases and 1.7 percent of Freddie Mac’s. Similarly, 3.4 percent of all loans went to Hispanics in 1993, but Fannie Mae’s purchases amounted to 2.7 percent and Freddie Mac’s 2.9 percent.” HUD’s proposed rules, however, do not directly address this issue. We think they should.
GSEs’ Past Performance and the Multifamily Market
“The lack of a strong secondary market for multifamily loans has made it more difficult to obtain debt financing for multifamily housing,” HUD reports. The Department estimates that 29 percent of conventional mortgages on housing for low- and moderate-income families will be for single-family (1-4 unit) housing, and 21 percent will be for multifamily units. HUD’s research shows that the GSEs’ market share in multifamily housing finance is far below their share of single-family finance.
In 1993, Fannie Mae purchased $4.6 billion in multifamily mortgages, while Freddie Mac purchased $191 million. This compares to almost $29 billion in total multifamily mortgage originations that year. Thus, the GSEs’ purchases amounted to about 17 percent of originations. Given that some of the GSEs’ purchases were older loans, their share of the current market is even smaller, according to HUD.
In 1993, the GSEs held or had securitized about 10 percent of outstanding multifamily mortgage debt. State and local housing finance agencies and insurance companies each held another 10 percent of outstanding debt. Depository institutions held 36 percent. Meanwhile, HUD estimates, the GSEs have provided financing for over half of all outstanding single-family mortgages.
Although improving, the GSEs’ – especially Freddie Mac’s – performance in the multifamily rental housing market is still weak, while the need for such housing is critical. According to 1995 statistics published by the Low Income Housing Information Service, the average rent for a one-bedroom unit is beyond the reach of at least one-third of renter households in every state. It is beyond the reach of more than one-half of all renter households in Hawaii, New York, and the District of Columbia. And two of five renter households in every state cannot afford the average rent for a two-bedroom unit. Two-thirds of the 14 million households with incomes below 30 percent of median paid more than 30 percent of their income for housing or lived in inadequate or crowded housing.
The large disparity in the GSEs’ performance in single- and multifamily housing finance has led some advocates to argue for the creation of a distinct category for multifamily lending to ensure that increased financing for multifamily housing becomes available.
Central Cities, Rural Areas, And Other Underserved Areas
The proposed goals for housing located in central cities, rural areas, and other underserved areas improves on existing regulations by developing a more rigorous definition of underserved areas. To define underserved areas, HUD used 1993 Home Mortgage Disclosure Act data and 1990 Census data, as well as reports and other research on the availability of mortgage credit and mortgage flows, to analyze mortgage application denial and origination rates throughout the country. HUD’s research showed that the availability of mortgage credit to an area is related to its minority concentration and income characteristics:
- Census tracts with higher percentages of minority residents have higher mortgage denial and lower loan origination rates than all-white or predominately white census tracts.
- Census tracts with lower incomes have higher denial rates and lower origination rates than higher income tracts.
The rule thus defines underserved areas as census tracts or non-metropolitan counties where minorities comprise 30 percent or more of the residents and the median income of families does not exceed 120 percent of the area median; or where the median family income does not exceed 80 percent of the area median. Rural areas are defined as any underserved area located outside any metropolitan statistical area so designated by the Office of Management and Budget. This refined definition of underserved areas appropriately focuses the GSEs’ goals in those areas that have the least amount of access to housing finance. (The Housing Assistance Council is pushing for further refinement of the definition of rural underserved areas.)
The GSEs regard the “central cities” portion of the goal as the most difficult. Fannie Mae reported under the old definition that it slightly exceeded the 30 percent goals for 1994, with 30.4 percent in the third quarter of 1994. But Freddie Mac reported that it again fell short under the old definition of this goal with only 24.8 percent (third quarter) of last year’s business in central cities.
Using the revised definitions, the proposed annual goal for the purchase of mortgages for housing located in central cities, rural areas, and other underserved areas is 18 percent for 1995 and 21 percent for 1996, while HUD estimates the total market for this category to be between 21 and 23 percent. According to HUD’s analysis of the GSEs’ 1993 activities, only 15.9 percent and 14.4 percent of Fannie Mae’s and Freddie Mac’s activities, respectively, were in underserved areas.
Special Affordable Housing goals
The proposed rule establishes the “Special Affordable Housing” goal at 11 percent for 1995, and at 12 percent for 1996. This goal has two parts. The first is homeownership for either low-income families living in low-income areas or homeownership for very low-income families. The second is rental housing for very-low-income families. Each subgoal is to receive 5.5 and 6 percent of the GSEs’ units financed during 1995 and 1996, respectively.
In this case as in the others, the GSEs are only following the market. According to HUD, “the market is originating many more loans for lower income homebuyers than the GSEs are purchasing.” The GSEs purchased a much smaller proportion of conforming mortgages originated for very-low income homebuyers than of mortgages originated for high income homebuyers (41 percent versus 55 percent), HUD’s research shows. In 1993, low- and moderate-income individuals received 16 percent of all mortgages made, yet of all loans purchased by Fannie Mae and Freddie Mac, only 14.4 percent were from low- and moderate-income borrowers.
The central question posed by HUD is: what is the appropriate extent to which Fannie Mae and Freddie Mac should be leading the housing finance industry to increase its lending to underserved segments of our population and disinvested communities? The GSEs contend they are already doing this, but research suggests they can do much, much better. The 1992 GSE Act is premised upon the belief that, while the GSEs cannot force mortgage originators to increase their lending in underserved markets, they can and should provide the necessary leadership to spur this outcome. Although the performance of the GSEs appears to have improved somewhat over the past two years, the bar still needs to be lifted further to ensure that they fulfill their public mandate.
About the Author
Comments