AIG on the Potomac


The future of government mortgage support

Today’s Financial Times articles: White House seeks wind down of Fannie and Freddie (Feb 11, 2011), Debate on Fannie and Freddie’s fate looms (Feb 11, 2011)

The stockholders of AIG were wiped out because that company sold insurance on mortgage-backed securities that fell in value. Not only did it price its insurance too cheaply, but it also neglected to allocate capital reserves to the contracts that it wrote.

More or less the same was true of Fannie and Freddie, but without the associated capital resources of a global insurance conglomerate. The ultimate stockholder of Fannie and Freddie, so it turned out, was the U.S. Treasury, and hence the U.S. taxpayer.

Now comes the Treasury report “Reforming America’s Housing Finance Market” which proposes (as Option 3) a system of “catastrophic reinsurance behind significant private capital” that would combine the private capital buffer of the AIG system with a scaled-back version of the public insurance system of Fannie and Freddie.

The idea seems to be that tail risk insurance is appropriately a public good, albeit one that needs to be priced, and reserved, more appropriately than in the past. First loss should be taken by the borrower, in the form of increased down payments. Second loss should be taken by the lender, in the form of increased capital reserve requirements. Only when all these buffers are exhausted would government insurance kick in.

This proposal represents a major pull back from governmental involvement in housing finance; actual funding of mortgages is to be largely eliminated. By my count, the word “private” appears sixty times in the thirty two pages of the report. Fannie and Freddie, the twin sources of the bulk of taxpayer losses in the crisis, are to be wound down as quickly as possible, with the speed limited only by the depressed state of housing markets.

Notwithstanding the dramatic transformation of government involvement, from outright intermediation to backstop reinsurance, in other respects the proposal rather proudly upholds American exceptionalism in housing finance. Securitization is here to stay, as are other distinctive features such as the pre-payable 30 year fixed mortgage. The report mentions the mortgage interest tax deduction only in passing; the non-recourse character of most American mortgages is not mentioned at all.

The future envisioned by the report is a restoration of the private securitization market, after fixing all of the (many) problems revealed by the crisis. The historic transformation from a bank lending-based credit system to a capital market-based credit system is thus to continue.

From a larger money view perspective, the most significant feature of the mortgage loan is that it amounts in effect to a kind of capitalization of future wage income; the house serves as collateral for the loan, but provides no cash flow toward paying back the loan. The mortgage market might better be viewed as a kind of human capital market.

Securitization was invented to allow humble households to tap capital markets, previously accessible only to government and the strongest corporate borrowers. It worked. In 2006, at the peak of the housing bubble, households accounted for an unprecedented 44.3% of total outstanding non-financial credit in the United States, more than business (32.1%) and more than government (23.6%).

The Treasury proposal would make it a bit harder for individual households to borrow against anticipated future wage income, but in the cause of making the resulting financial instrument more readily acceptable in world capital markets.

The way the global system worked before the crisis, the U.S. was a net purchaser of goods from the rest of the world, and it paid for those goods with promises to pay in the future. Foreigners wouldn’t take mortgage loans, but they would accept mortgage-backed securities, at least before the crisis.

Perhaps the Treasury proposal will convince foreigners to dip their toes into our waters once again.

Share your perspective