Lessons from a crisis

5 investment experts debate how the credit crisis unfolded and whether anyone will be learning from it

120808 roundtable group
Participants in the round table (from left): Leslie Rahl, Cliff Asness, Bennett Golub, Peter L. Bernstein and Harrison Hong.

Fannie Mae. Freddie Mac. AIG. Citigroup. Household names for sure, but also companies among the biggest recipients of the government's largess as officials try to grease the wheels of finance to get the economy and capital markets back on track after a truly awful fall.

The government is pumping $150 billion into AIG alone. Much of the insurance giant's hemorrhaging comes from writing thousands of credit default swaps on esoteric investment vehicles, such as synthetic collateralized debt obligations. When AIG officials wrote those swaps, did they have any idea of how much risk the company was taking on? If they didn't know, how could their investors know? How could they manage their risk? No matter how fancy a risk measurement tool investors used, how could they understand the risks if the data were wrong?

On Nov. 11, Pensions & Investments brought together five leading investment experts and thinkers to discuss how U.S. capital markets first, and then global markets, succumbed to the credit crisis that's already in its second year and has pulled economies around the world into recession. This high-level group not only talked about what lessons history already had to offer before the crisis hit, but also what role leverage has played and what new lessons are being taught — and whether investors are in a learning mood.

Participating in the “Picking up the Pieces” round table were:

Peter L. Bernstein, president, Peter L. Bernstein Inc., New York.

Cliff Asness, managing principal, AQR Capital Management, Greenwich, Conn.

Bennett Golub, vice chairman, BlackRock (BLK) Inc. (BLK), New York.

Leslie Rahl, president, Capital Market Risk Advisors, New York.

Harrison Hong, John H. Scully '66 professor in finance, Princeton University, Princeton, N.J.

Joel Chernoff, executive editor, moderated the discussion, which was held at P&I's New York offices.

120808 golub
Bennett Golub: “You can make lots of money riding bubbles, you just need to know when to get out of them.”

Joel Chernoff: Ben, clearly BlackRock (BLK) was early in picking this out, but should investors as a group have realized that the real estate bubble was going to burst? And if so, what could they have done differently to manage their risk?

Bennett Golub: Well, sort of by construction, when you call something a bubble, then its bursting is, you know, a given. So I think that the really important question is not whether the bubble is going to burst, but rather when. That's, of course, the hard one. Because, in fact, you can make lots of money riding bubbles, you just need to know when to get out of them.

It was — you knew something had to do it, and we kept on thinking, what was the catalyst going to be. So I'm not sure that there was a particularly obvious way to predict the time of the bubble bursting, but that it had to burst I think was pretty straight forward.

Mr. Chernoff: Leslie, was there a good way to reduce one's risk that investors didn't do?

Leslie Rahl: Well, I'm not sure if it's a matter of reducing risk or at least understanding the risk that you were taking. What troubles me is I don't know how to make money without taking risk, so risk unto itself isn't bad.

I think there were a lot of people, though, who thought they were taking low risk who, in reality, were taking high risk, and that's a real failure of the system.

Now part of it, I think there is no such thing as a free lunch, and if someone is getting an extra yield, there's a reason. I also think that asking questions about why is this a better investment, although it is highly complex, highly structured and very, you know, very highly leveraged, people — and I have many clients who, unfortunately, when you try to explain it — they don't necessarily, in the good times, want to hear what the issues are.

So ... they might have done things differently, they might not have done things differently. But I think there was a lack of awareness of the amount of risk that people were taking that is quite concerning.

Peter Bernstein: When you look around the world and see how much of the rest of the world was engaged in this, it is amazing.

Harrison Hong: It's hard, because I think as we're beginning to see, a lot of smart people basically, people who have had terrific track records have gotten killed in these most recent times. So it's a little hard I think with hindsight to say ... that it was so obvious.

But I do think that one of the lessons you draw from the history of these crises is that housing bubbles are particularly nasty. I think if you look at most of the real serious stuff that had major contagion into lots of the real (estate) sector, it's typically been always about housing.

I think, in particular, the Fed really kind of made a mistake of not calling or at least recognizing that if it was something having to do with housing, we should be a little careful, because I think they drew the wrong lesson from let's say the tech bubble. We came out of the tech bubble, things were kind of fine, right, and in some ways you would have thought the technology was pretty great coming out of the tech bubble. It's not even clear it was a bad thing. But now at housing I think, if you look at the history, at least the Fed should have been more aware of the systematic risk associated with housing.

Mr. Chernoff: And investors, should they have done anything differently?

Mr. Hong: I think it's hard to put that onus on regular investors when you have pretty sophisticated investors who couldn't see through all this stuff. Again, this is the case where I really doubt if anybody could have seen just how much — how destructive this has been. I think you're beginning to see how interconnected markets are, the importance of leverage. I don't think that was so obviously forecastable upfront.

Cliff Asness: It's not just knowing you're in a bubble, it's what do you do about that as an active investor, someone who's trying to add value. You have to try to think about when that's going to burst. But that's art, not science. Let's not kid ourselves.

Mr. Bernstein: The most important thing I learned writing “Against the Gods” is Pascal's Wager. What are the consequences if I'm wrong? I have two choices. If I do this and I'm wrong, what are the consequences? The people who were taking out subprime mortgages could not be wrong. Housing prices had to keep going up. So, OK, they were going up. But by 2005, they had just about quadrupled. So that's a moment, I think, where you can say what happens if I'm wrong and housing prices shift and begin to go down from this point? I mean, I'm ruined. I'm ruined.

So that question wasn't asked. People couldn't conceive of the possibility they might be wrong. And you can be wrong any minute. We can be wrong one minute from now. You have to keep asking that question.

Mr. Asness: I think you're dead right, but the world where the person taking out the subprime mortgage asked that question and gets it right ... I'm cynical we'll ever get to that world. Even though you're dead right, that question should have been asked.

120808 hong
Harrison Hong: “Individually, it may be optimal to take a lot of leverage to max out, but from a society perspective, if everybody does this, it can be very suboptimal.”

Mr. Chernoff: I'd like to move on to alternatives and leverage, which we've discussed briefly. Harrison ... what are the lessons, if any, from use of increased leverage across alternative asset classes? And I know these are quite distinct, but I'll just throw out private equities, hedge funds and real estate.

Mr. Hong: I think this is probably my favorite question, because the two big lessons I would draw that we didn't know beforehand is that — individually, it may be optimal to take a lot of leverage to max out, but from a society perspective, if everybody does this, it can be very suboptimal.

And I think the mechanism that you see now playing (out) is that guys took a lot of leverage in lots of these different asset classes, but this leverage has basically now increased a lot of the correlation among these classes. There are guys who just can't hang onto positions, and therefore, they have to delever, which leads them to what might be a pretty good asset class that's not being affected by some fundamental shock, being now basically affected by the delevering process. And I think to some extent there's a lack of coordination at a societal level of leverage. I think it's maybe the biggest factor that's creating the systematic meltdown that you're seeing, and I think that was one of the things that we probably didn't see upfront, even though I think as economists, we kind of understood this point that there's a leverage externality. We don't incorporate the leverage that we take for the rest of society or for the system as a whole.

Mr. Chernoff: So you're saying that increased leverage increased correlations among these alternative asset classes?

Mr. Hong: Absolutely.

Mr. Chernoff: Peter?

Mr. Bernstein: That's important. My concern about alternatives in institutional portfolios really is a matter of liquidity. Except for pension funds, most institutions are spenders rather than lenders, and so they need cash to fund their budgets. And it's OK when asset prices are high, this is very easy to do, but you're in trouble when asset prices are down.

And I think the answer to the question you put to me is really a risk management problem. I mean, how do you deal with — you know, you have the problem — you're going to have to pay out cash from time to time. You also know that if the stock market takes a header, it would be nice to be able to get in there and not be locked into real estate and so on and so forth.

So this is a risk management problem. It has to be thought through. But liquidity, liquidity is a risk management problem.

Ms. Rahl: I've been espousing a concept for several years now that has gotten zero traction, but that I still believe makes a lot of sense, and that is a term structure of liquidity. That people who are willing to invest for a longer period of time, should get either higher returns or lower —

Mr. Asness: Lower fees.

Ms. Rahl: Lower fees. Keep more of the upside, and that there be a term structure of liquidity. And, again, I've tried that on many endowments and foundations and pension funds, as well as on hedge funds and others, and people look at me like I've got two heads. And maybe now they'll start thinking about it, because I really think it makes sense.

Mr. Asness: Leslie, I've spoken to clients and at conferences and been quoted as saying I would cut fees in half for a longer lockup. And it's a great way for a long-term institution to flat out monetize the fact that they have a long horizon. Now if they don't really have a long-term horizon because they're overcommitted and need the cash, they can't do it and they've got to rethink that. But it's a way to turn a long-term horizon not into expected value, but into literal — the difference between expected value and actual value is variance — to a zero variance expected value to certain cash. And I have one investor who's taken me up on that in about six years of talking about it.

Ms. Rahl: And that would also imply that funds of funds should pay higher fees than other types of investors.

Mr. Asness: Let's not even talk about them. The four-letter words will start flying.

Mr. Golub: What I take away as my first lesson is that the only source of liquidity is cash flow, and anything else is a conjecture, and it's a conjecture that assumes that there will be an orderly market — that there'll be buyers and sellers.

So you have to go to the fundamentals of the type of institutions and type of assets, and I think the liquidity mismatch has created tremendous havoc. So if you are an endowment and you have some amount of your assets, which are probably on call for that new building, that tells you that you can have a certain percentage of your assets that are illiquid, because you actually try to line up the demands for your liquidity vs. your ability to supply it.

And there we have all sorts of havoc going on. So, you know ... SIVs did not collapse because of derivatives. SIVs collapsed because of the most fundamental mismatch of liquidity. In fact, there was an S&L crisis and probably even an earlier one where you had a situation (of) long assets vs. short liability. It doesn't work.

So I think that is, you know, a critical thing. I think as you combine that with leverage, we should also be very cognizant of the fact that there are different types of leverage. So if you have an illiquid asset and you fund it with a long liability, that might actually work. If you have a structured finance vehicle and it's a cash flow asset with a cash flow contingent liability, that actually will work.

The same asset funded with a market-value-driven entity can blow up because you have mismatches.

120808 asness
Cliff Asness: “There are two disasters in a complex portfolio: not having a risk system or believing the one you have.”

Mr. Chernoff: Let's move ahead to value at risk, and Leslie, let me ask you, value at risk focuses on volatility as the key measure of evaluating risk. Are there better ways of measuring risk, in the context of institutional investors like pension funds, endowments and foundations?

Ms. Rahl: Well, I go back to a paper that Ben and I both participated in preparing back in February before this latest part of the crisis, and we and the other people we worked with concluded, as I have believed for as long as I can remember, that there is no one measure that you can rely on. So we can debate whether VaR is better than stress testing is better than the Sharpe ratio. It doesn't matter. You need at least four or five different measures to get any real understanding of a complex portfolio.

So I think the debate is not can VaR be one of the measures that someone looks at? The issue is no one measure should be the only measure that someone looks at.

And I think for institutional investors, there's an additional part of that question, and that is, what is your horizon? So even if you decide that VaR is one of the four or five you want to look at, I think if you're truly going to be a long-term investor, you have to have a slightly longer concept or there's going to be a real mismatch between your risk, the way of looking at risk and the way you look at your overall portfolio.

Mr. Asness: The world is much more fat-tailed at short horizons than it is in long horizons.

Ms. Rahl: Absolutely. But I'd always remind you that after the (Long Term Capital Management) crisis in 1998, there were a lot of regulators and other studious groups who said just that. And everyone quickly forgot that you need to complement VaR with other metrics.

Mr. Bernstein: Everyone forgot?

Ms. Rahl: That you need to complement VaR with other metrics. That stress testing and other tools are essential, and perhaps even more important than VaR, although VaR does have its place.

Mr. Asness: Well, I'm a fan of saying there are two disasters in a complex portfolio: not having a risk system or believing the one you have. And it comes down to the same thing. Some common sense has to be used. Multiple measures have to be used. You're still not going to always get it right. It's a best guess.

Running a modern portfolio is somewhat about estimating what the worst cases would look like, because there's always a worst case that can destroy everyone. If you run from that — Leslie said it earlier — you have to take some risk. You can't assume worst cases are infinite.

Ms. Rahl: There's also the issue of a portfolio manager, a trader, anyone who's actually running a day-to-day business has to think about the most likely case. A risk manager has to think about the tails and what could go wrong, and therefore, VaR is a more useful metric in thinking about what's likely to happen and not really as useful in fearing what could happen.

Mr. Asness: And even then, that's art, not science.

Ms. Rahl: Absolutely.

Mr. Asness: I'm a fan of the rather prosaic phrase “anything that has happened before, can happen triple,” which is not a great risk management system necessarily, but something to keep in mind.

Mr. Chernoff: Peter, I'd like to turn to you. In markets, swings often go too high on the upside and too low on the downside. Will investors' risk aversion follow the same pattern? When the dust settles, what's the new normal for institutional investors going to look like?

Mr. Bernstein: Actually, I have two scenarios, and I don't know which one is more likely, but — and they're kind of extremes, but I think they describe something.

The bailout business, which nobody talks about, I watch it in astonishment, because people used to complain about moral hazard and bailouts, and here we have jumbo bailouts —

Mr. Asness: Particularly Republicans used to complain about it.

Mr. Bernstein: Exactly. The conservative upright people in the community would complain about this, and this is astonishing what's happening. And how will it change, our sense of risk going forward: It really doesn't matter what the hell we do, because the government is going to make it come out all right anyhow. So the devil may care.

The other is what I suggested before, that the nightmare memories linger on as they did after the Depression, and there's a long period of intense risk aversion, which makes it very nice for people who are prepared to take risk. I mean great opportunities.

My sense is that that's the more likely outcome because that's what history tells us. On the other hand, history has been a very failed guide for the last six to nine months. This bailout thing is something nobody could have imagined, but it will resonate for a long time, and moral hazard problems I think are for real.