Proposed Housing Goals for Fannie Mae and Freddie Mac for 2005-2008
July 15, 2004
U.S. Department of Housing and Urban Development
Regulations Division, Office of the General Counsel, Room 10276
451 Seventh Street, SW
Washington, DC 20410
Re: Proposed Housing Goals for Fannie Mae and Freddie Mac for 2005-2008,
Docket No. FR-4790-P-01
Dear Mr. Secretary:
The NATIONAL ASSOCIATION OF REALTORS® is pleased to comment on the Proposed Rule establishing Fannie Mae and Freddie Mac housing goals for 2005-2008. Our 1 million strong Association supports efforts to expand affordable housing opportunities but is vitally concerned about the Proposed Rule and the potential impact on commercial, investment, and residential real estate and related activities that directly affect the rental and homeownership aspirations of Americans. While REALTORS® support setting ambitious housing goals for Fannie Mae and Freddie Mac, we are concerned that the Proposed Rule assumes a GSE market share and future economic conditions that are too optimistic. The assumptions that drive the Department’s aggressive goals and subgoals may cause the GSEs to take actions that will distort mortgage markets, with negative results for the FHA markets and middle-income home buyers, especially in high-cost markets. NAR recommends that HUD raise the goals to the levels proposed for 2005 and keep them at these levels through 2008.
The Department has proposed a major increase in the GSE housing goals and established subgoals based on the “recognition that there is substantial room for the GSEs to grow in certain market segments.” Under the Proposed Rule, the Department would establish higher goals, increasing gradually from 2005 to 2008, with subgoals under each of the housing goals for purchases of mortgages for owner-occupied housing in metropolitan areas and underserved areas. By setting goals as a percentage of all loans purchased by Fannie Mae and Freddie Mac, the Department expects the GSEs to “match the market” by 2008 in terms of purchasing prime and A- subprime mortgages within the conforming loan limit. The Department also maintains that establishing subgoals for home purchase loans under the three larger goals will encourage the GSEs “to take a leadership position in creating homeownership financing opportunities” in line with the housing goals required by Congress.U.S. Department of Housing and Urban Development
Docket No. FR-4790-P-01
Page 5
Appropriate Market Benchmarks are Difficult to Establish
The goals, as currently structured, could distort the market in a number of ways that would adversely affect housing. Goals that are set too high can be just as damaging as goals that are set too low. While it is incumbent on the Department to establish goals that will stretch the GSEs, the goals should be established in a way that avoids disruptions and creates the proper set of incentives.
As HUD acknowledges in its technical discussions, data limitations make it extremely difficult to measure the size of the “affordable market” in any given year. No single data source provides all of the requisite information, and even a combination of data sources leaves many critical issues unresolved. As a result, HUD must rely on a number of inherently arbitrary assumptions to estimate the share of originations that qualify for the various affordable housing goals. Among the most important unknowns are the annual dollar volume of multifamily originations; the average loan amount per unit (for both multifamily and 2-4 unit properties); the relative size of the single-family rental mortgage market; the share of single-family rental and multifamily units that meet the various goals; and the size of the subprime market. Unfortunately, none of these factors can be measured with any degree of precision, and relatively small differences in values can lead to relatively large differences in the estimated market share of affordable loans.
As great as they are, the uncertainties involved in measuring past performance are dwarfed by the uncertainties involved in predicting performance in future years which, in the end, must serve as the basis for establishing the GSE goals. Even if one accepts HUD’s estimates of previous market shares that meet the goals, it is unreasonable to assume that the economic conditions that produced these outcomes will characterize future markets. Increases in housing prices have exceeded income growth in the past few years, interest rates are on the rise, and rental markets are soft. And, according to a recent study, the credit scores of renters have declined significantly over time, reducing the number of qualified borrowers. R. Bostic, P. Calem, S. Wachter, “Hitting the Wall: Credit as an Impediment to Homeownership,” Nov. 2003. These trends suggest that further increases in homeownership rates will be more difficult to achieve in the next few years, and that the affordable goals established by HUD are unrealistically high and are caused by housing market conditions that cannot be appropriately addressed by the GSEs.
HUD acknowledges these uncertainties by testing the sensitivity of its market estimates using a variety of different assumptions and establishing a projected range of future market shares that meet the goals. These ranges, along with the annual goals proposed by HUD, are presented in Table 1 of the preamble to the Proposed Rule. As shown in Table 1, goals in 2007 and 2008 have been set at the upper end of projected shares under the assumption that the GSEs should lead the market. However, even if we assume that HUD’s methodology is sound, the Department’s analysis cannot make up for the fact that the underlying data is deficient. Given the uncertainties involved in measuring both past and future market performance, we believe that it is inappropriate to set the goals at the high end of the estimated ranges. While the GSEs should lead the market, their activities should not distort it. Setting goals that are inconsistent with likely future conditions will only do the latter.
Selected Data from Table 1 of the Preamble
| HUD's Proposed Housing Goals | HUD's Projected Market Estimates | ||||
| 2005 | 2006 | 2007 | 2008 | ||
| Low- and Moderate-Income | 52% | 53% | 55% | 57% | 51-57% |
| Underserved Areas | 38% | 39% | 39% | 40% | 35-40% |
| Special Affordable | 22% | 24% | 26% | 28% | 24-28% |
Impact on Rental Markets
Setting the goals too high could have unintended consequences for rental housing. If market conditions make it difficult to achieve the goals through the purchase of single-family mortgages—for example, if there is a major recession or another refinancing wave—GSEs would most likely respond by buying additional multifamily mortgages regardless of the market’s underlying need. Since rental units are much more likely to meet the goals, this is the surest and most cost-effective way for the GSEs to increase their goal performance.
Given the current softness of rental markets, a response of this kind could well lead to over-investment in the rental sector and threaten the recovery that is now underway. The GSEs have increased their multifamily purchases significantly in the past few years, in large part due to pressures in meeting their housing goals during a refinancing wave. While the GSE share of the multifamily market remains below the level achieved in the single-family sector, there is little evidence that the multifamily mortgage market is currently underserved. Indeed, given the high volume of originations that has occurred in the past two years—and the fact that most multifamily mortgages have five year lock-in periods—demand in the next few years is likely to be relatively low.
Setting the goals too high would thus appear ill-advised. As we have seen too often in the past, too much investment in rental housing can exacerbate the cyclical nature of the market and have devastating consequences on the long-term health of the sector.
Impact on FHA
Setting the goals too high also threatens to jeopardize FHA, which provides a critical source of financing for low- and moderate-income homebuyers, including minority and first-time homebuyers. Recent years have witnessed a dramatic decline in the relative performance of FHA loans. Delinquency rates on FHA mortgages have risen dramatically in the past few years, and according to the Mortgage Bankers Association, now exceed the delinquency rates reported on subprime mortgages. In contrast, the delinquency rates on conventional mortgages have remained relatively low, creating a large and growing gap in the performance of conventional and FHA mortgages. Foreclosure rates show a similar pattern. The foreclosure rate on conventional loans has remained relatively flat since the early 1990s. In contrast, the foreclosure rate on FHA loans has more than doubled, rising from 0.43 percent in 1990 to 0.87 percent in Q1 2003 despite aggressive loss mitigation efforts by HUD.
These patterns suggest that FHA has been losing its better quality loans to the conventional market in recent years. Technological innovations in the private sector, coupled with a heightened focused on affordable lending, have driven conventional lenders into markets previously dominated by FHA. A dramatic increase in the GSE goals will only accelerate this process, causing further deterioration in the quality of the FHA portfolio that will eventually result in higher fees to FHA borrowers or government subsidies to pay claims, effectively making FHA the lender of last resort.
Such an outcome would pose a significant threat to the future of FHA. FHA has historically relied on lower risk loans to cross-subsidize its riskier borrowers. Without these lower risk loans, or an explicit subsidy, insurance premiums will eventually have to rise to keep the Mutual Mortgage Insurance Fund (MMIF) financially sound. This will initiate a continuing cycle of rising fees and fewer lower-risk loans going to FHA. If this occurs, FHA’s ability to fulfill its public mission will be seriously compromised.
Thus, in the end, unrealistically high goals for the GSEs could ultimately be counterproductive to the Administration’s efforts to expand the number of low-income and minority homeowners. A strong FHA has been vital to the financing needs of underserved communities. While the GSEs can and should do more, this should not occur at the expense of FHA.
Impact on Middle-Income Markets
REALTORS® are also concerned about the consequences of the Proposed Rule for middle income families. The clear intention of the Rule is to get the GSEs to purchase more low- and moderate-income home mortgages. The objective is laudable, but the execution may require the GSEs to reduce purchasing loans to middle-income families—such as teachers and other public servants, especially those living in high-cost areas—and loans at the upper end of the conforming loan limit. This consequence conflicts with the statutory requirements of the Fannie Mae and Freddie Mac mission to provide liquidity and stability in the secondary mortgage market. The GSE housing goals were established to extend the established benefits of the secondary market to low- and moderate-income borrowers, not to eliminate them for middle-income borrowers.
Recommendations
Given the uncertainties about the size of the “affordable market,” the possibility the goals will distort the mortgage market, the prospect that the GSEs may be forced to limit purchases of middle-income mortgages to meet the housing goals, and the risk to FHA, NAR believes that HUD’s decision to set the goals at the higher end of the estimated market is ill-advised. Instead, we recommend that the Department establish the goals at the lower end of the probable range and apply the goals proposed for 2005 to the remaining years. This recommendation stems from our strong belief that unrealistically high goals would ultimately be bad for housing and the very renters and homebuyers that HUD intends the GSEs to serve.
The prospect that the GSEs may be forced to purchase fewer mortgage loans made to middle-income families—the “bread and butter” of the real estate industry and the economy—is especially troubling given the significant GSE mission to provide liquidity for the secondary mortgage market. First-time home buyers and the homeownership aspirations of low- and moderate-income homebuyers should not be pitted against the homeownership aspirations of middle-income households. This is a potential unintended consequence that cannot be left unaddressed in the final rule
As an alternative to setting goals at the 2005 levels, the Department could reconsider the decision to discontinue using bonus points to provide incentives for GSE mortgage purchases. Bonus points would enable the Administration to target GSE activities to those specific segments of the market that remain underserved and increase homeownership, including minority homeownership. If desired, the need for bonus points could be reviewed on an annual basis to determine their continued appropriateness.
Another reason for maintaining the goals at the proposed 2005 levels is the very real possibility that Congress may revamp the mechanisms by which the GSEs are required to meet their affordable housing mission. Legislative proposals under consideration this year, and expected to be revisited by the next Congress, could have a dramatic impact on how the GSEs achieve this mission. Congressional committees have expressed interest in a variety of ideas, from using a percentage of GSE profits to support affordable housing to a model based on the Federal Home Loan Bank system. HUD's regulatory revision of the goals may be superseded by the legislative proposals before Congress, in effect rendering these goals transitional. We urge the Department to be mindful of these legislative deliberations as it considers how to proceed in setting new housing goals.
On behalf of the NATIONAL ASSOCIATION OF REALTORS®, I urge you to consider our comments and make the appropriate adjustments to proposed housing goals before issuing the Final Rule.
Sincerely,
Walt McDonald
President
