Transitioning to Alternative Structures for Housing Finance

More than six years after the federal government took control of Fannie Mae and Freddie Mac, policymakers are weighing a comprehensive overhaul of the mortgage finance system that could shrink or eventually close the two entities and create a system with more private capital. Fannie Mae and Freddie Mac were originally chartered as government-sponsored enterprises (GSEs) to ensure a stable supply of credit for residential mortgages nationwide. They operate in the secondary (or resale) market where they buy mortgages from the financial institutions that make the loans (thus ensuring that those institutions have a source of funds to originate new mortgages). Fannie Mae and Freddie Mac then pool those loans to create mortgage-backed securities (MBSs), which they guarantee against defaults on principal and interest payments by borrowers and sell to investors.

Through its financial commitment to the two GSEs and its other mortgage programs, the federal government now directly or indirectly insures over 70 percent of all new residential mortgages. Loans guaranteed by Fannie Mae and Freddie Mac account for over two-thirds of those mortgages (about 50 percent of the total amount of mortgages), and loans insured by the Federal Housing Administration (FHA) make up most of the remaining federal share. Such government dominance was not always the case—in the 20 years before the financial crisis that began in 2007, roughly half of all mortgages were financed without backing from the federal government or either of the GSEs.

As the effects of the financial crisis have receded and as the housing market has recovered, policymakers have taken some initial steps toward returning to a secondary mortgage market with more private-sector involvement. Those steps include raising the fees that Fannie Mae and Freddie Mac charge for their guarantees to levels that private firms may be better able to compete with. CBO expects that the steps already taken, together with pending changes to financial regulations, will reduce the two GSEs’ share of the mortgage market over the next 10 years.

This report examines various mechanisms that policymakers could use to attract more private capital to the secondary mortgage market. The report also addresses how those mechanisms could be combined in different ways to help the market make the transition to a new structure during the coming decade. CBO analyzed transition paths to four alternative structures that involve choices about whether the government would continue to guarantee payment on mortgages and MBSs and, if so, what form and prices those guarantees would have. Under those different structures, the government’s activities would range from providing full or partial guarantees for a large share of the mortgage market to playing a minimal role in a largely private market (except perhaps during a financial crisis). Any transition to a new type of secondary market would also require decisions about what to do with the existing operations, guarantee obligations, and investment holdings of Fannie Mae and Freddie Mac.

CBO’s analysis has three key findings:

• A transition to a new structure for housing finance that emphasized private capital could reduce costs and risks to taxpayers. One drawback to such a transition is that mortgages could become somewhat less available and more expensive to borrowers. Thus, over the longer term, it could also result in a modest shift of the economy’s resources away from housing toward other activities.
• Although the transition to a new structure could significantly decrease the number of borrowers who received mortgages backed by Fannie Mae or Freddie Mac, additional private capital would replace most of the lost funding. Borrowers would probably not face significant increases in interest rates because the two GSEs’ current pricing is not too far below market pricing. Consequently, a gradual transition would probably exert only modest downward pressure on house prices.
• Because policymakers have already raised the guarantee fees charged by Fannie Mae and Freddie Mac close to those that CBO estimates would be charged by private insurers, the budgetary costs of the two GSEs’ activities over the next 10 years are expected to be small. As a result, the budgetary savings would also be small under any of the transition paths to a more private system that CBO considered. Thus, the choice between the different market structures probably rests primarily on considerations other than budgetary costs. (Those findings depend on the accounting framework that CBO uses for Fannie Mae and Freddie Mac, as described below.)

CBO’s projections of budgetary costs and the size of the federal role in the mortgage market involve considerable uncertainty. In particular, because the market for mortgages is now dominated by large government-backed entities, the price that private investors would charge to bear the risks of mortgage guarantees and how that price might evolve over time are highly uncertain. CBO based its projections on its assessment of the middle of the distribution of estimates of that price.

What Options Would Attract More Private Capital?

CBO expects that the role of Fannie Mae and Freddie Mac in the secondary mortgage market will shrink over the next decade under current policy. If policymakers wanted to reduce that role further and lessen the advantages given to the two GSEs, they could use various mechanisms:

• Raise guarantee fees on mortgage-backed securities further to bring the two GSEs’ pricing closer to pricing in the private market. Even small increases in those fees would nearly eliminate budgetary costs for Fannie Mae and Freddie Mac (using the accounting framework described below) and allow private firms to capture some of the GSEs’ current business.
• Change the limits on the maximum size of mortgages that Fannie Mae and Freddie Mac are allowed to guarantee. The limit is currently $625,500 in areas with high housing costs and $417,000 in the rest of the country, although the average size of new mortgages guaranteed by the two GSEs is only about $200,000 in 2014.
• Share the credit risk of mortgages (the risk of loss when a borrower defaults) with private investors—for example, offer compensation to induce private investors to assume responsibility for the initial losses on GSE-guaranteed loans.
• Auction off a limited number of new guarantees by Fannie Mae and Freddie Mac to the highest bidders rather than requiring the GSEs to continue to guarantee all eligible mortgages submitted by lenders for preset fees, thus reducing the size of the two GSEs’ guarantee business. Auctions would determine the market prices of those guarantees and allocate them to the lenders who most valued them.

How those mechanisms were used, whether alone or in combination, would depend on which new structure for the secondary mortgage market policymakers wanted to encourage. Changes in the GSEs’ guarantee fees and loan limits could be useful for many types of restructuring, whereas risk sharing and auctions would be more appropriate for a transition to a smaller, but continuing, federal presence in the market. (The Federal Housing Finance Agency, which regulates the GSEs, currently has the authority to employ any of those mechanisms, although CBO’s budget projections for Fannie Mae and Freddie Mac are based on the assumptions that their present fee levels and loan limits will continue and that risk sharing with private investors will remain limited.)

What Structures for the Secondary Market Did CBO Consider, and What Would a Transition to Them Involve?

For this analysis, CBO packaged the aforementioned policy mechanisms into illustrative transition paths that, between 2015 and 2024, would move the secondary mortgage market from dominance by two large government- sponsored enterprises to one of four alternative structures:

• A market with a single, fully federal agency that would carry out the two GSEs’ main function of buying eligible mortgages and turning them into securities that are guaranteed against losses from default on the underlying mortgages. The transition to such an agency would require little or no change to the structure of the GSEs’ guarantees, the fees charged for them, or the GSEs’ loan limits because no significant amounts of new private capital would be required beyond those that are expected to be invested under current policy. By the end of the transition period, the federal agency would have a smaller share of the market than Fannie Mae and Freddie Mac have today.
• A hybrid public-private market with federal guarantees against catastrophic losses. Under that structure, the government and private investors would share credit losses on eligible MBSs, with federal guarantees covering catastrophic risks (those associated with severe downturns in the housing market) for a significant share of mortgages. As a result, taxpayers would bear most of the losses during a crisis, but private investors would bear most of the losses in other periods. The main policy mechanism used to transition to that structure would be sharing credit risk with private investors.
• A market with the federal government as “guarantor of last resort,” in which private companies would guarantee most new mortgages in normal times, but the government would fully guarantee most new mortgages during financial crises. (In normal times, the government would guarantee a small sample of mortgages of all sizes to ensure that it is capable of doing so in times of crisis.) The major policy actions taken to establish the new structure would be auctioning off the GSEs’ new guarantees and raising their loan limits.
• A largely private market with no explicit federal guarantees of MBSs (other than those provided by the Government National Mortgage Association, which securitizes and guarantees mortgages insured by other federal agencies, such as FHA and the Department of Veterans Affairs). That structure would minimize the explicit credit risk borne by taxpayers. The main policy changes made during the transition would be raising guarantee fees and lowering loan limits until the GSEs no longer guaranteed new mortgages.

Those alternative market structures share some common features (although those features might be altered through policy changes that lie beyond the scope of this report). The government would guarantee only mortgages that met certain eligibility criteria, and private financial institutions would provide most other mortgages. FHA would continue to provide assistance to low-income homebuyers. Under all of the structures, the portfolios of mortgages that Fannie Mae and Freddie Mac hold as investments would be reduced. Depending on the new structure, the two GSEs could be incorporated into a single federal agency, liquidated (with their operating systems sold to private investors), or privatized. Any transition would involve legal and regulatory issues that would necessarily take some time to resolve, which is why CBO examined transitions that would take place over a 10-year period.

How Would the Transition to Those Structures Affect Borrowers, the Housing Market, and the Federal Budget?

If policymakers reduced the role of Fannie Mae and Freddie Mac, borrowers would probably face somewhat higher interest rates on mortgages, and house prices would probably decline modestly. The increases in interest rates that borrowers faced, however, would probably be smaller than the fluctuations in market interest rates that occur during a typical year. Borrowers would most likely continue to have access to 30-year fixed rate mortgages as long as the market for converting mortgages into MBSs was large and the securities were easily traded, whether or not that market had government backing. Lending standards would most likely be higher on privately backed mortgages, and during a financial crisis, the availability of those private loans could be sharply disrupted, causing their costs to rise significantly.

Under current policy, CBO expects that Fannie Mae and Freddie Mac will guarantee, on average, about $1 trillion of new mortgages per year over the next decade but that their overall share of the mortgage market will decline from over 50 percent now to about 40 percent by 2024. CBO projects that the present value of the government’s total income and payments over the life of those mortgages will translate to costs of about $19 billion for taxpayers over that period (estimated on a fair-value basis, as described below). Those budgetary costs would be significantly reduced in a transition to any of the market structures that CBO analyzed except in the transition to a market with a single, fully federal agency. (In that transition, costs over the next 10 years would rise slightly.)

Under all of the illustrative transition paths, some borrowers who would have had GSE-backed mortgages under current policy would shift to FHA-insured loans rather than to privately backed loans. CBO estimates that the increase in the volume of mortgages for single-family homes guaranteed by FHA would range from relatively small—in the transition to a market with a single, fully federal agency—to significant—in the transition to a largely private market. Because the two GSEs and FHA are accounted for differently in CBO’s estimates of the federal budget (as described below), the shift in guarantees from Fannie Mae and Freddie Mac to FHA would have the effect of increasing the budgetary savings projected for the transition paths, even though the risk borne by taxpayers would be little changed.

How Does CBO Account for Fannie Mae, Freddie Mac, and FHA in Its Budget Estimates?

The estimates of budgetary costs in this report depend on the accounting treatment that CBO uses for the two GSEs and FHA. CBO accounts for the costs of Fannie Mae’s and Freddie Mac’s activities on a fair-value basis, in which estimated costs represent an approximation of the price that the federal government would need to pay a private insurer to make loan guarantees on the same terms as the GSEs. Because those fair-value estimates incorporate a charge for market risk—the additional compensation that private investors demand to invest in risky assets such as mortgages—they provide a more comprehensive measure of the costs of guarantees than do projections of the net cash costs associated with guarantees.

That choice of accounting method has two consequences for CBO’s analysis:

• Transactions that occurred at market prices in liquid and orderly markets would have no fair-value costs. Thus, regardless of which structure for the secondary market was ultimately adopted, the cost or savings to the government from transferring the GSEs’ existing mortgage assets or guarantee obligations to private investors would probably be small under fair-value accounting. If, instead, CBO accounted for the GSEs’ activities on a cash basis, as the Administration does, those transactions would result in costs for the government because the GSEs would lose future streams of income, which would include at least some compensation for market risk.
• The fair-value approach would probably show small savings from a transition that reduced the volume of new guarantees by Fannie Mae and Freddie Mac and increased the role of the private sector in the secondary mortgage market. If, instead, CBO accounted for the GSEs’ activities on a cash basis, that same transition would probably result in large estimated costs to the government. Specifically, the GSEs’ current activities are expected to produce cash savings for the government because Fannie Mae and Freddie Mac charge enough for their loan guarantees to more than cover the projected losses on of those guarantees (though not enough to cover the risks that a competitive private insurance company would factor in when charging for the same guarantees).

In contrast to CBO’s treatment of Fannie Mae and Freddie Mac, guarantees by FHA, like the activities of most other federal credit programs, are accounted for in the budget using the procedures specified in the Federal Credit Reform Act of 1990. Unlike the fair-value approach, those procedures do not incorporate market risk. As a result of those differences in accounting, additional savings would be reported in the budget as some borrowers shifted from GSE financing to FHA guarantees.