2003: Does Fannie Mae pose systemic risk to the economy?

AUDIENCE MEMBER: Hi. I’m John Golob from Kansas City. I have a follow up question on derivatives.

After the press zeroed in on the comments in your annual letter, the head of Fannie Mae got up and said, well, Mr. Buffett’s criticisms don’t apply to Fannie because, number 1, we have simple vanilla derivatives that are priced in the market. And secondly, we need derivatives to protect against interest rate risk.

And I guess, given what happened to the savings and loan industry back in the ’80s, that seems reasonable.

So my first question is what is your rejoinder to Mr. Raines?

And the second part of my question gets to your concern about the connection between derivatives and systemic risk. And that is, do Fannie and Freddie play a particularly prominent role in this concern? Thank you.

WARREN BUFFETT: Yeah. I have a lot of respect for Frank Raines. I think he’s done a good job at Fannie Mae. I don’t know the situation intimately.

The problem, as you mention, is that an operation like Fannie Mae or Freddie Mac, or savings and loans in the past, had this problem, which is inherent in the mortgage instrument, in matching — or coming close to matching — assets and liabilities.

And the reason they had that terrible problem, which did in many institutions, was the optionality in a mortgage instrument.

And in a mortgage instrument, particularly as the years have progressed, you buy a — you know, if you buy a mortgage — or somebody else takes out the mortgage, you own it — you have a 30-year instrument if it’s a bad deal and you have about a 30-minute instrument if it’s a good deal.

The buyer — the person who takes out the mortgage — can call off the deal at any time at relatively low cost. And the public has been sensitized to that more and more as time has gone by, so they’ve been quicker to refinance for very small differentials.

Now, many years ago, they had what they called due-on-sale clauses in California. I think — were they invalidated, Charlie, or what happened with those? So that there were ways of shortening up the mortgage expected maturities.

But it’s a fundamental problem when you are operating on borrowed money in a very big way, which is what S&Ls did and what Fannie does, and Freddie, that you have this very long-term instrument, and it — but it can be very short-term if it becomes advantageous to you, and rates go down and you want to keep it. Or it becomes very long-term if rates go up and nobody wants to refinance.

And under those circumstances, if you run a huge institution, or even a smaller one, but that has a high — highly leveraged, you are going to look for one way or another to try to match the duration of your liabilities as close as you can to the duration of your assets, and have various methods to protect yourself against the optionality that exists with the counterparty, in effect, your asset.

That’s not easy to do. And Fannie, and Freddie, and other institutions, attempt to do that through various types of derivative instruments, as well as other things, in terms of the kind of debt they issue themselves and so on.

And they’re very smart, and they do it — you know, my guess is they do a better job than Charlie and I could do at it, but it can’t, by its nature, be perfect.

And under some circumstances, where you get large gaps — the thing you worry about in financial markets, and it doesn’t happen very often, but your — the thing that really destroys people are what the academics would call six-sigma, or five-sigma, or seven-sigma events, which are things that are never supposed to happen, basically.

And sigma is a method of describing the probabilities that they will happen in any given period — the number of sigmas.

Financial markets don’t lend themselves well to modeling based on that. You know, they do most of the time, until it doesn’t work. And when it doesn’t work, you know, chaos reigns.

And there are more six-sigma events that happen in financial markets — or theoretical six- sigma events — than any study of probability curves would ever come up with.

And that’s — when you have gaps or discontinuities, when markets close, whatever it may be, those are what cause institutions to go out of business.

And derivatives, in my view, anyway, accentuate the possibility of it happening and the extent of the damage, if and when it should happen.

Again, we don’t think we mathematically can tell you what the probabilities are of something like that, and we think anybody that does tell you, you know, is kidding you.

And I’ve had managers of hedge funds sit down with me and tell me that they had, you know, a 28 percent probability of returns between 30 and 40 percent. Come up with all these exact figures. Anybody comes up with exact figures in finance, watch out.

So, I would say that if I were running Fannie or Freddie, I would be terribly conscious of what was happening in there, and I would understand this basic problem I have, which I can guarantee you, Frank Raines understands extremely well, of the optionality built into his assets.

And I would try to come as close to matching that. And if I did it through derivative instruments or whatever, I would try to match — I would try to reduce the troubles that could be produced by that optionality to a minimum.

And then I would get very worried about who the counterparties were, because anytime somebody promises to pay you a lot of money if something terrible happens to you, you better be very sure that they can and will pay, because the terrible thing that’s happening to you may be presenting terrible things to them.

You know, that will happen in the insurance industry from time to time. And that’s why reinsurance recoverables are a dangerous asset to have.

It will happen in the derivative markets.

When LTCM had troubles with one type of asset, they had troubles with a lot of types of assets, and everybody else they were doing business with was having a lot of troubles with those same things.

And that’s why the Federal Reserve stepped into something they never dreamt they would have stepped into. They stepped in to force, essentially, a solution, which may have been the right thing to do, incidentally.

But they — for some obscure hedge fund that nobody in the world virtually had heard of, until that point, and had started threatening the U.S. — the stability of the U.S. financial system.

Charlie?

CHARLIE MUNGER: Well, I think you’re right to point to this creditworthiness of the counterparty risk.

My guess is that both Fannie and Freddie have been pretty intelligent at thinking through a whole lot of different scenarios where they’ll be OK, or close to OK, if all the counterparties pay.

And I would bet a lot of money that they weight the possibility that counterparties won’t pay a lot lower than we do.

I think a lot of the counterparties are behaving in a lot more dangerous way than Fannie and Freddie are, and that — and the counterparties can get in trouble because of that. And they can translate that trouble to people who assume that they’re hedged.

WARREN BUFFETT: And that’s true of the mortgage guarantee institutions that take part of the risk away.

Fannie and Freddie are very sophisticated institutions. Very, very sophisticated institutions.

But if you depend too much on other people, there can be periods in financial history where all the sophistication in the world may not save you. The best thing to do is to be able to count on your own resources.

And at Berkshire that’s basically the way we operate. And it may be safer than necessary, but, you know, Charlie and I are rich enough already. We do not need to stay up at night.