It's a perfect time to think about the fundamental restructuring of the world famous, now broke, Government-Sponsored Enterprises (GSEs), Fannie Mae and Freddie Mac.
Last summer, as we could all observe, the giant GSE risk turkey, weighing in at $5 trillion, came to roost in the dome of the U.S. Capitol. Like Edgar Allan Poe's celebrated raven, it won't go away. It roosts there, so the elected representatives of the people can't forget the mistakes they made in fattening it up.
But while Poe's raven croaked "Nevermore," does the GSE risk turkey gobble "Evermore"? Is this risk a permanent burden on the public finances? Should it be? My answer is No.
A recent libertarian critique of the GSEs' failure by Sheldon Richman has the title, "Bailing Out Statism." Richman says, "The key to understanding the saga of Fannie Mae and Freddie Mac is this: They were created--intentionally--to distort the housing and mortgage markets. Government planners were not content to let voluntary exchange configure those industries. So they intervened."
That is an accurate statement. The ongoing bailout of Fannie and Freddie is without question a government intervention to save a previous intervention.
Richman continues, "The collapse of Fannie and Freddie is government social engineering predictably gone bad."
But here is something very interesting: Fannie and Freddie went bad in a way nobody predicted--namely: from credit risk, or put simply, bad loans.
There were hundreds of articles and speeches either attacking Fannie and Freddie or defending them, many of them by the distinguished members of this panel, but everybody thought that the GSEs' credit risk function was all right, that if anything, it demonstrated monopoly profits.
Nobody predicted that credit risk would take Fannie and Freddie down. But it did.
Now they are both insolvent, if you don't count the preferred stock bought by the government. Their common stock has lost 99% of its value and represents a mere option on their possible future resuscitation.
Many people have suggested that the way to fix Fannie and Freddie is to limit them to mortgage securitization--in other words, to guaranteeing credit risk. But the fact that they failed through credit risk must make us dubious about this notion. (Many people also assert that securitization was the "original" purpose of the GSEs. For Fannie, this is distinctly wrong: Fannie had existed for more than 40 years before it did its first securitization.)
You could solve the problem of GSE credit risk by further limiting Fannie and Freddie securitization only to prime, high quality mortgage loans. But this would certainly render unhappy the many politicians who want them to finance riskier loans and make this course politically difficult, perhaps impossible. Of course, trying to make these same politicians happy with them contributed significantly to Fannie and Freddie's credit risk expansion, resulting losses and current insolvency.
More fundamentally: whatever may have been the case a generation ago, does the 21st century need GSEs in order to have securitization of prime loans? No, it doesn't. It is true that private securitization of prime, conforming loans has not previously existed, but this is only because no private firms could compete with the government-granted advantages and economic subsidies enjoyed by Fannie and Freddie.
Once past the panic and the bust, and without GSEs, there would be a private market for securitization of high quality loans. The financial technology is not very hard and well known. Such a market would work best if mortgage loan originators kept a significant, life-of-the-loan interest in the credit performance of the loans they originate--a key design improvement for future.
So what should be done with Fannie and Freddie going forward?
The answer must address two phases: first, a transition period during the current bust; then a long-term restructuring.
In this transition period, Fannie and Freddie have already become part of the government, for all intents and purposes. They have almost entirely changed from being GSEs to being government housing banks. They are therefore available to be directed by the government to fund Hope for Homeowners loans, for example, or in general to participate in refinancing the mortgage bust based on political policy, not profit. [This was made clear by their assigned role in the new administration's "Homeowner Affordability and Stability Plan," announced February 18, 2009, subsequent to our conference.]
This transition role could be made more efficient by completing Fannie and Freddie's conversion to temporary government housing banks by adding a legal Treasury guaranty of their debt. The government, as agent for the involuntary taxpayer participants, is already fully and in fact on the hook for all the debt and MBS obligations of Fannie and Freddie. But this is not legally explicit, however real.
The formal documentation of the government's relationship with Fannie and Freddie specifies explicitly that their debt is not explicitly guaranteed. When the lawyers for international investors discovered this and pointed it out to their investor clients, it made them all nervous. So the government could say that the debt was "effectively guaranteed," which is true, but had to admit that it was still not "explicitly guaranteed."
Why have this bond market uncertainty, since the public has all the risk anyway? Using Fannie and Freddie to address the bust would be cheaper and easier if Congress would simply enact an explicit guaranty of their debt, reflecting the de facto reality.
At the same time, the remaining common shares of Fannie and Freddie could be removed from private hands, definitively ending GSE status. Some of my colleagues argue for doing this by receiverships, which would wipe out the shareholders, but you could also have a government tender for all remaining shares. This would not cost very much money at current prices, and make Fannie and Freddie into 100% government-owned corporations--exactly how Fannie started out in 1938.
This transition government housing bank status should have a firm sunset safely on the other side of the bust: in five years. At that point, the long term restructuring of Fannie and Freddie would be effected.
I suggest for your consideration that this long term restructuring should divide Fannie and Freddie into three parts:
1. Prime mortgage securitization
2. Mortgage portfolio investing
3. Government activities
Fannie and Freddie's prime loan securitization businesses should be privatized--sent out into the world to compete like anybody else. They would become not GSEs, but private companies, sink or swim, flourish or fail.
Some have suggested that such companies could be organized as cooperatives, not ordinary corporations. The problem with the cooperative idea is that the industrial structure of the mortgage business has profoundly changed. From the historical pattern of thousands of small, mostly local savings and loans and mortgage banks, it has shifted to a concentrated industry with four dominant companies: Bank of America, Wells Fargo, Morgan-Chase, and Citi. After the further concentration the mortgage crisis has engendered, these four now represent 63% of all U.S. mortgage servicing. Would we want these four as the dominant "cooperators" in charge of a cooperative form? I don't think so. We want a more competitive outcome: regular, for-profit corporations.
As for the business of owning a leveraged portfolio of mortgage loans and MBS, this should be moved into a banking charter or a mortgage real estate investment trust (REIT), and then carry on life like any other bank or REIT, as part of the private market, once again, sink or swim, with a fresh start. The existing government-backed debt and assets should be put in a liquidating trust and liquidated over time (as was done in the privatization of Sallie Mae).
The final element of the former Fannie and Freddie consists of those activities which are properly those of the government: such as providing housing subsidies in one form or another and providing non-market financing of risky loans. These would stay in the government. Their funding would have to be appropriated by Congress in a transparent way, instead of escaping appropriations discipline by being hidden in the GSEs.
These governmental functions of Fannie and Freddie should be merged into the structures of the Department of Housing and Urban Development-FHA-Ginnie Mae.
The total result of the three steps: No GSEs are left--a consummation devoutly to be wished.
Alex J. Pollock is a resident fellow at AEI.