The Federal Role in the Financing of Multifamily Rental Properties

Multifamily properties—those with five or more units—provide shelter for approximately one-third of the more than 100 million renters in the United States and account for about 14 percent of all housing units. Mortgages carrying an actual or implied federal guarantee have been an important source of financing for acquiring, developing, and rehabilitating multifamily properties, particularly after the collapse in house prices and credit availability that accompanied the 2008–2009 recession. According to the Federal Reserve, the share of outstanding multifamily mortgages carrying such a guarantee increased by 10 percentage points, from 33 percent at the beginning of 2005 to 43 percent at the end of the third quarter of 2014. (A slightly larger increase of about 16 percentage points occurred in the federal government’s market share of the much larger single-family market.) Such guarantees are made by a variety of entities, and some policymakers are looking for ways to make the federal government’s involvement more effective. Other policymakers have expressed concern about that expanded federal role and are looking at ways to reduce it.

What Are the Pros and Cons of Federal Support of the Multifamily Mortgage Market?

The federal government’s support of the multifamily mortgage market is one of many federal policies aimed at providing support for rental housing for low- and moderate-income families. Mortgage guarantees made by the Federal Housing Administration (FHA), the Rural Housing Service (RHS), and Fannie Mae and Freddie Mac—the two large government-sponsored enterprises (GSEs) that have been operating under federal control since 2008—increase the availability of mortgage credit. They do that primarily by insuring investors who buy mortgage-backed securities (MBSs) against losses that they would incur if the mortgage borrowers—whose mortgage loan payments provide the cash flows for those securities—defaulted. Those loan guarantees help to provide liquidity and stability to the market for those mortgages, particularly during periods of stress. They also lower the cost of financing multifamily housing projects slightly, because the fees the borrowers pay for their guarantees are lower than what a private guarantor would charge.

Providing such benefits through federal credit guarantees has several drawbacks, however. Loan guarantees could expose the federal government to potentially large losses if many multifamily loans defaulted. In addition, actual or implied federal loan guarantees may encourage excessive risk taking by insulating lenders and investors from losses on investments they would not make without the guarantee. Such guarantees slightly increase the large subsidies that favor housing over other types of investment, resulting in a slightly less productive allocation of capital resources in the economy. Furthermore, although the lower development costs made possible by government guarantees may be passed along to renters in the form of lower rental rates, the reduction is probably small.

What Are the Budgetary Costs of Federal Guarantees of Multifamily Mortgages?

Federal loan guarantees can generate very different budgetary effects depending on the accounting method used to value them. The Congressional Budget Office estimates that new loan guarantees issued by FHA and RHS in 2016 for multifamily mortgages will generate budgetary savings of about $386 million over their lifetime if the projected budgetary effects are calculated using the procedures delineated in the Federal Credit Reform Act of 1990 (FCRA) or costs of about $334 million if they are calculated on a fair-value basis.

Both estimates are based on the same projections of cash flows for those credit programs; the difference between them stems from the discount rates used to convert the projected cash flows to a present value. FCRA requires that a program’s cash flows be discounted using the rates on Treasury securities of comparable maturity to the cash flows of the program. Fair-value estimates also include the cost of market risk, which can be expressed as an adjustment to the discount rate to reflect the premium one would have to pay an investor to take on the market risk of a loan guarantee. The fair-value estimate approximates the price that the federal government would need to pay a private insurer to make loan guarantees on the same terms as FHA’s and RHS’s. Because those fair-value estimates incorporate a charge for market risk, they provide a more comprehensive measure of the costs of guarantees than do FCRA estimates.

CBO also projects federal budgetary costs for Fannie Mae and Freddie Mac. Since 2008, when those two GSEs were placed into conservatorship by the government, CBO has treated them as governmental for budgetary purposes and estimated the cost of their credit guarantees as if they were provided directly by the federal government. Unlike explicitly federal credit programs, whose budgetary effects are estimated under FCRA, the budgetary effects of the GSEs’ activities are estimated on a fair-value basis in CBO’s budget projections.

On that basis, the GSEs will generate a budgetary cost to the government of about $129 million in 2016 for their multifamily loan guarantees, CBO estimates, reflecting a subsidy rate (the loans’ lifetime cost divided by the amount of credit extended) of 0.2 percent. In comparison, CBO projected a fair-value subsidy rate of 0.4 percent for the GSEs’ single-family guarantee operations for 2016 in its August 2015 baseline. The fair-value subsidy rate for the GSEs’ multifamily loan guarantees is smaller because the fees they charge are estimated to be closer to those charged in comparable private transactions. In addition, in their multifamily operations, the GSEs require private lenders and investors to assume a larger share of any losses incurred on the mortgages they guarantee through a variety of risk-sharing mechanisms.

How Might the Federal Role in the Multifamily Mortgage Market Be Changed?

Policymakers could attempt to make the federal role in the multifamily market more efficient or reduce it by shrinking or eventually closing the GSEs’ multifamily operations and modifying the federal government’s explicit credit programs for that market. In a previous report, CBO identified four broad approaches to modifying the federal role in the single-family mortgage market.  Those same approaches could be applied in the multifamily market.

• Fully Federal Agency. A federal agency would replace the role played by the GSEs and provide an explicit federal guarantee for all credit losses on the mortgages that it insures.
• Hybrid Public/Private Approach. Private firms would cover initial losses and a federal agency would absorb the remainder of losses by providing a partial guarantee on some of the mortgages issued in the multifamily market.
• Federal Guarantor of Last Resort. A federal agency would provide a full credit guarantee to a small share of the market during normal economic conditions and to a larger share of the market during periods of economic stress.
• Largely Private Approach. A federal agency would provide a full guarantee on only a portion of FHA’s and RHS’s loan guarantees projected under current law; all other multifamily mortgages would be provided by private firms without federal guarantees.

In each of those approaches, Fannie Mae and Freddie Mac could have the multifamily portion of their operations made either explicitly public under the aegis of a federal agency or explicitly private, through privatization or liquidation. The remaining federal agencies (some of which might be consolidated) could target some or all of their guarantees toward multifamily rental properties for households with low incomes, as they do today. Because the single-family and multifamily loan guarantees offered by FHA, RHS, and the GSEs are designed to support different policy objectives—homeownership in the case of single-family guarantees and adequate, affordable rental properties for multifamily loan guarantees—-policymakers might modify the two markets differently.

The impact of the four approaches on the availability of mortgage credit during periods of economic stress would depend largely on the degree to which each relied on a federal guarantee. A fully federal agency would provide the most federal support to the market, and the largely private approach would provide the least. The effect of a government guarantee on credit availability under the hybrid public/private and federal guarantor of last resort approaches would be more complex. In normal economic conditions, the government guarantor in a hybrid public/private approach would offer guarantees on more loans than a guarantor of last resort. However, as the guarantor of last resort, the government would increase its share of guarantees in times when that guarantee could be most valuable to the market and most costly to the government.

What Would Each Approach Cost?

The cost of each approach relative to what would be expected under current law could vary considerably depending on the details of its implementation and the budgetary treatment used. To provide illustrative estimates of the cost of each approach, CBO analyzed the budgetary effects in 2020, by which time CBO assumed the transition would be complete. CBO estimated the costs of mortgage guarantees under two approaches: using a fair-value basis consistently for all guarantees and using current budgetary procedures (a FCRA basis for all explicitly federal programs and a fair value basis for the GSEs).

For the implementation of the fully federal agency, CBO assumed that mortgages that formerly carried a partial guarantee from the GSEs would instead carry a full federal guarantee with loan terms and subsidy rates similar to those of a representative loan in one of FHA’s programs. The higher subsidy rate and the availability of longer-term fixed-rate financing would increase the market share of mortgages carrying a federal guarantee. Thus, on a fair-value basis, estimated total federal costs would increase relative to estimated costs under current law. However, if the costs of existing federal agencies and the new agency were estimated under FCRA, total estimated budgetary costs under the new system would be lower than under current law, primarily because GSE-guaranteed mortgages that have a cost on a fair-value basis would be replaced with federal mortgage guarantees that produce projected savings under FCRA.

For the hybrid public/private approach, CBO assumed that guarantee fees and loss coverage on loans formerly guaranteed by FHA and RHS would be reduced to match those for the GSEs’ existing partial guarantee, which offers recipients a less generous subsidy. That change would slightly reduce the projected dollar volume of mortgages carrying a government or GSE guarantee. The change would produce estimated savings under both budgetary treatments, but the savings under FCRA would be considerably larger because of the switch in treatment for loans formerly guaranteed by the GSEs.

For the federal guarantor of last resort and the largely private approaches, a significantly smaller share of mortgages would carry a federal guarantee. The fees and terms on those guarantees would be similar to those in existing credit programs under current law. That reduction in federally guaranteed mortgages would produce estimated savings on a fair-value basis but estimated net costs under the current budgetary treatment, because the FCRA savings from loans guaranteed by FHA and RHS under current law would be reduced. For the guarantor of last resort, CBO’s estimates of the dollar volume and cost include the potential effect of an increase in the share of mortgages that may be guaranteed by the agency in a crisis; the government would increase its share of guarantees significantly once every 40 years and moderately once every 10 years, CBO projects. In all other years, CBO projects, the agency would maintain a 5 percent share of the total market. On average, the government would maintain a 10 percent share.