The GSEs: Even Yoda cannot see their fate

This is a bond market blog. Fannie Mae and Freddie Mac are dominant players the bond market. Not only are they the biggest non-Treasury issuers of straight debt securities, mortgage-backed bonds guaranteed by the two mortgage giants represent over 30% of the total taxable bond market. Therefore the problems at the GSEs probably touches just about every investor directly in a way few other companies would.

While delinquencies on their guarantee portfolio remain relatively small (0.81% for Freddie Mac and 1.22% for Fannie Mae), the fact is that both companies employ tremendous leverage, and therefore losses even mildly above historic norms are likely to put huge pressure on the company's equity. In addition, Freddie Mac is yet to complete the $5.5 billion capital raise they promised in May, and given market conditions, this will be all but impossible without government intervention. So I'm not here to challenge the plunging share price of either Fannie Mae or Freddie Mac.

I also don't feel like commenting on the bailout plan, other than to say that its a sad day for free markets. I see the Treasury as between a rock and a Depression, and has selected the rock. I'd have done the same. I don't blame Treasury so much as I lament that its come to this. Exactly who to blame for this or what could have been done differently in the past is a discussion for another time.

There are two somewhat unrelated things I want to comment on. First is this stupidity about FAS 140. FAS 140 has nothing to do with anything. Or at least, it should have nothing to do with anything. An upcoming revision to FAS 140 will tighten the rules about off-balance sheet accounting. Based on how both GSEs currently operate, this would probably require both Fannie Mae and Freddie Mac to bring their guarantee portfolio on balance sheet. Under current statues, the GSEs minimum capital required is 0.45% of of their guarantee portfolio and 2.5% of their aggregate on-balance sheet assets. So taken literally, the GSEs capital requirements would increase dramatically if their entire $4.5 trillion guarantee portfolio were brought on balance sheet. The commonly reported number is $43 billion for Fannie Mae and $38 billion for Freddie Mac.

Of course, its ridiculous to think that regulators would allow an accounting rule change to dictate the GSEs minimum capital. Economically the GSEs are no more or less safe whether this stuff is on balance sheet or off. Its especially ridiculous given that a capital infusion of that level would be impossible to achieve in almost any market, much less now. And the GSEs alternative would be to sell massive amounts of MBS on their books, which would benefit exactly no one. Indeed OFHEO director James Lockhart has said point blank that FAS 140 should not apply to the GSEs and that accounting changes would not drive a capital change.

The second is where we should go from here with the GSEs. I will ignore what the Libertarian in me would like to see and thinking about what could happen. In other words, we should take it as a given that the base mission of the GSEs will remain as is. Congress wants to see some entity out there which can promote home ownership and help stabilize the mortgage market.

My idea would be to break up Fannie Mae and Freddie Mac into several, smaller entities. Say we make it 10 different companies, which I'll just call the "Macs." The $4.5 trillion guarantee portfolio would be divided equally among the Macs, not by geography, but more or less randomly, with some effort made to be sure the quality of each loan portfolio is roughly the same.

All of the Macs would be given a credit line with the Treasury equal to 10% of their guarantee portfolio. Any amounts drawn on the line would be at some small spread to LIBOR, say 3-month LIBOR + 20bps. This would in essense ensure liquidity at each of the Macs while at the same time giving them a strong incentive to use private funding, which would almost certainly be at a better rate than LIBOR +20. Note that historically, Fannie Mae and Freddie Mac debt has traded at LIBOR minus 20 or so.

The Macs capital adequacy could be managed more like a bank's. And there would be a stipulation that if the Macs capital fell to a certain level, the Treasury would take over operations, not unlike the FDIC taking over a failing bank. There could then be something similar to a bankruptcy procedure, but with the Treasury credit line ensuring that some mass contagion did not ensue.

What's the advantages here? First of all, no one Mac would be too big to fail. If one of the Macs f'ed up their hedging or some such, the government could liquidate shareholders relatively easily. Second, there would be a more known procedure for what happens in event of a Mac failure. Third, having several Macs would encourage competition among them, which would probably drive mortgage rates down. As it is, Freddie and Fannie are dictating terms of the mortgage lending business.

Now I know this isn't a complete plan, but I'm interested to hear comments on what we should do with the GSEs. I encourage everyone to remain within the realm of realism. There will be such a thing as a GSE in the future. So let's consider what they will look like.