In a year when angry voters are demanding a reduced government role in the economy, it is remarkable that most of the ideas for supplanting Fannie Mae and Freddie Mac are just imaginative ways of keeping government in the business of housing finance.
While the Obama administration has not yet outlined its own proposal, Treasury Secretary Timothy F. Geithner has already signaled that the government will continue to have a role.
This is pretty astonishing. One would think that something might have been learned from the recent past, when two New Deal ideas for government housing support--the savings and loan industry and the government sponsored enterprises, Fannie Mae and Freddie Mac--failed spectacularly. It cost taxpayers $150 billion to clean up the first and may cost more than $400 billion to resolve the second.
The same old arguments are being trotted out: without government support there can't be a 30-year fixed-rate mortgage; interest rates in a nongovernmental system will make homes unaffordable; and inevitable disruptions in the credit markets may put a crimp in housing finance.
These bugaboos are mostly myths. A Google search reveals offers of 30-year fixed-rate jumbo mortgages--loans too large to be bought by the GSEs and provided by private lenders without government support. Studies by Federal Reserve economists in 2006 found no significant reduction in mortgage rates because of Fannie Mae and Freddie Mac's low-cost funding and secondary market activities.
As for market disruptions, insulating housing from financial downdrafts would be bad policy; it would eliminate incentives to be cautious in borrowing and investing and encourage the growth of bubbles. Moreover, the recent financial crisis--the worst since the Great Depression--affected securities backed by prime jumbo mortgages in the same way it affected other high quality asset-backed securities like credit cards or car loans: a sharp drop followed by a gradual recovery.
The major losses on mortgage securities occurred among those backed by subprime and other nontraditional loans, and the widespread failure of these weak loans caused a housing price collapse.
Accordingly, if our mortgage finance system were used only for conventional prime mortgages--those with a 10 percent minimum down payment, made to borrowers with good credit--there would be no need for government support. For example, in the housing bubble that ended in 1979, when almost all loans were traditional mortgages, foreclosure starts in the ensuing slump peaked at just 0.87 percent in 1983, according to the Mortgage Bankers Association's annual delinquency survey.
In the next bubble, which ended in 1992 and primarily involved traditional financing, foreclosure starts reached 1.32 percent in 1994.
But after 1992, when government housing policy encouraged the creation of 27 million subprime and other nontraditional loans--half of all mortgages--that result became destructive. When this bubble deflated in 2007, foreclosure starts jumped to a record 5.37 percent in 2009.
The lesson is that where mortgages are of good quality, the deflation of bubbles will not cause a serious financial disruption--certainly not enough to warrant continuous government involvement in housing. But government policy that deliberately degrades loan quality or creates moral hazard will eventually cause devastation in the housing market.
All this leads to several conclusions:
Government involvement in housing finance is an invitation to disaster. As illustrated by the S&Ls and GSEs, no matter how such a system is structured, government support will hide the real risks.
Arguments for government involvement are unpersuasive, amounting to unverifiable, or imaginary, dangers. We shouldn't forget that the vast majority of our economy functions successfully without direct government support.
As long as the residential mortgage system creates high- quality assets, occasional changes in credit availability can be handled without difficulty.
When the new Congress begins weighing alternatives to the GSEs, they should start with a fully private system. There are several options, and none necessarily excludes the others.
The George W. Bush administration proposed a covered bond system, in which banks would issue debt backed by a pool of mortgages that they hold on their balance sheets but are separate from other assets. This has worked well in Europe and could be adopted in the U.S.
Earlier this year, a House of Representatives committee sent a bill to the floor--sponsored by Scott Garrett, a Republican from New Jersey--that would create a system in which any financial firm could issue covered bonds.
The housing finance system in Denmark is also worth considering. Homeowners can buy back all or part of their mortgage debt at a discount when interest rates rise and they can refinance when rates fall. Only mortgages with loan-to-value ratios of 80 percent or lower are allowed into the system, a major reason for its long-term success.
A third option is securitization, which has provided low cost asset-backed financing for more than 30 years. While the GSEs are still in a government conservatorship, a transition to a fully private securitization system shouldn't be difficult.
As investors begin to return to the securitization market, the government can start reducing the size of mortgages that Fannie Mae and Freddie Mac are allowed to acquire. While that occurs, private securitizers should move in to take their place. We can observe whether the market is supplying all the credit that the housing industry needs, and stop the process if it isn't.
When the size of mortgages that Fannie Mae or Freddie Mac can buy approaches zero, they can either be shut or privatized. At the same time, a covered bond or Danish system, or both, could begin operations. If these alternatives function better than the securitization system, as determined by homebuyers and investors, they will supplant it.
The important point is that the housing market will finally be returned to parity with other financial markets and cease to be a ward of the government.
Peter J. Wallison is the Arthur F. Burns Fellow in Financial Policy Studies at AEI.