The future of the housing finance system will confront the new Congress. In particular, what should be done with Fannie Mae and Freddie Mac, or what should replace them? Like General Motors and Chrysler, these Government-Sponsored Enterprises, and the entire housing finance system along with them, have become wards of the federal government; unlike the auto companies, there have not yet been any serious efforts to determine what comes next.
Considering that the world financial crisis began with the startling collapse of Fannie Mae and Freddie Mac in July 2008, this is a remarkable omission by the Obama Administration and the last Congress. The 380,000-word Dodd-Frank Act included a large number of new requirements for individual mortgage lenders, but was silent about the structure of the overall housing finance system. The Administration asked for public responses to a series of questions about a future system this past spring and held a conference about those responses in August, but it has been silent since then.
The only activity last fall was a Congressional hearing by the House Financial Services Committee, called by committee Chairman Barney Frank (D.-Mass.). That September hearing was not encouraging to anyone wanting to avert a repetition of the debacle.
Five of the nine witnesses at the hearing were spokespersons for the housing industry, financial institutions, or Wall Street. They spoke with one voice: We need a steady supply of money for housing, and that means we need a federal government guarantee.
This certainly describes the GSEs as they have operated since they were established. It also describes the old savings and loan associations; they were required to concentrate on making mortgage loans, and they received a guarantee in the form of federal insurance on their deposits though the Federal Savings and Loan Insurance Corporation. These precedents should give lawmakers pause. The S&Ls collapsed in the course of a long period of accelerating inflation that began around 1965; legislation in the early 1980s to strengthen them through expanding their powers on both the asset and liability sides of their balance sheets proved unavailing, and finally several hundred hopelessly insolvent institutions were closed, beginning in 1989. The financial resolution of these "zombie" S&Ls cost American taxpayers about $150 billion, measured in today's dollars. Fannie Mae and Freddie Mac have already cost about that much, and serious estimates of the total loss to taxpayers range from $221 to $363 billion.
The S&Ls were the dominant sources of mortgage money for about 50 years; the GSEs lasted about 20. The industry witnesses at Mr. Frank's hearing seemed to be hoping that the third time would be the charm.
The other four witnesses were independent analysts. Most of them also favored a guarantee, with some reforms: the guarantee would be limited to mortgage-backed securities, the government would charge for the guarantee to cover losses, and it would be available to more than just Fannie Mae and Freddie Mac or their successors. There would be several, or many, issuers of mortgage securities; they would compete with each other. If they wanted to hold mortgages in their own portfolios, there would be no guarantee to back that investment.
This sounds like an improvement over the current GSE system, but it ignores the history of the GSEs. Neither Fannie Mae nor Freddie Mac was originally expected to hold portfolios of mortgages. Fannie Mae was set up to buy mortgages when money for housing was scarce and to sell them when money was more abundant. It was established as part of the federal government and limited to FHA and VA mortgages guaranteed by the government. Freddie Mac was created to provide the savings and loan associations with a GSE of their own; it was owned by the S&Ls and it specialized in issuing mortgage-backed securities. They both knew, however, that holding mortgages was more profitable than securitizing them, especially since their government guarantee meant they could borrow more cheaply than any bank or any other mortgage lender. Once Fannie Mae was spun off from HUD in 1968, and once Freddie Mac became independent of the S&Ls in 1989, they both rapidly built up their portfolios. That was where the money was. It was also where the risks were.
Limiting Fannie and Freddie or their successors to issuing mortgage-backed securities didn't work the last time. Nor is it likely to work this time. Even though we now know about the tremendous risks the GSEs have taken and the problems they have created for our economy, those memories will fade. Sooner or later, there will be problems in the mortgage market again. When they occur, the instinct of the industry and policy-makers will be to turn to the GSEs. For example, during the last severe recession in 1981, when rates on first mortgages averaged 15 percent, Fannie Mae was given the authority to buy second mortgages on a temporary basis. Mortgage rates came down within a couple of years, but Fannie Mae continued to buy second mortgages. In the recession of 1990 after the collapse of the S&Ls, the ability of banks to make construction loans to builders was sharply curtailed. In response to homebuilders' concern, Fannie Mae was allowed to set up a pilot program for construction loans. That program also became permanent. Both second mortgages and construction loans are much riskier than first mortgages to buy homes. Again and again, the GSEs have expanded their range of business activities, whenever the opportunity arose. That is one reason for their current problems.
One thing will be different for any new version of the GSEs. There is virtually unanimous agreement that they will not continue to have "affordable housing goals." In 1992, as part of the Federal Housing Enterprises Financial Safety and Soundness Act, Congress required the GSEs to devote part of their mortgage purchases to housing for families in the lower half of the income distribution, and also families living in areas where most residents are in the lower half of the income distribution. Specific percentage targets for these goals were set by Congress in the Act on a transition basis. The U.S. Department of Housing and Urban Development was required to revise the targets periodically, by regulation, which it has done three times, effective in 1996, 2001, and 2005, and the new regulator, the Federal Housing Finance Agency, issued new targets in 2009.
"Affordable" is often equated to "subprime," and blamed for the GSEs' collapse. GSE senior officials have seen them as a wonderful excuse, a way of shifting blame from their own mistakes and diverting attention from the accounting scandals that hit both GSEs in 2003 and 2004. The goals are also blamed by some conservatives, who see them as credit allocation.
But this convenient explanation doesn't fit the facts. The GSEs began investing heavily in subprime mortgages, and relaxing their underwriting standards, in 2002 - well before the affordable housing goals were increased in 2005. Indeed, they began to pull back from the subprime market after the new goals became effective. Their subprime mortgage purchases dropped from 2004 to 2005, measured both in number of loans and as a percentage of subprime mortgages, as shown in data compiled both by the GSE regulator and by Edward J. Pinto, a former senior Fannie Mae official. The GSEs were buying half of all subprime mortgage-backed securities in 2003, and only 20 percent by 2006.
If the GSEs are reconstituted with the same powers they now have but without responsibility for any affordable housing goals, the future of the housing finance system is likely to be a rerun of the recent past. Even if they are initially limited to issuing mortgage-backed securities and not allowed to buy mortgages for their own portfolio, they are likely to repeat the past, just taking a little longer. Either way, the taxpayers stand to lose again.