The credit market turmoil of the summer of 2007 will go down in financial history as following the classic script written at least as far back as the Dutch tulip bulbs of 1637 and chronicled by Charles Kindleberger in Manias, Panics and Crashes (1978 and several reprints thereafter). Kindleberger is not the only one to explain how such things happen again and again, though he may be the best storyteller among those who cover these repeating histories. This column will not attempt to compete with Kindleberger; rather, it will try to remind us of lessons we seem to relearn each time—lessons best remembered before, not after, the storm clouds appear.
Most of the newspaper headlines, blog arguments and CNBC commentaries of August were devoted to demands that the Federal Reserve cut the Fed funds rate immediately. When the Fed trimmed the discount rate and left the funds rate untouched, the chorus of complaints grew even louder. The lessons behind the Fed's action lie in the difference between solvency and liquidity and a basic understanding of what banks do.
Banks borrow money by gathering demand deposits and lend money by making loans. Theoretically, demand deposits can be withdrawn any time while at least some loans cannot be called without warning or waiting. Banks traditionally are borrowing short-term money and using it to fund long-term loans. This is fine until all the depositors want their money at the same time and the bank doesn't have it because it made a lot of long-term longs. The result is a bank run. A bank is solvent if assets (loans) exceed liabilities (deposits). At the same time, if there is no cash in the till, the bank is not liquid and a bank run would ruin it. In normal times, no one cares about the difference between solvency and liquidity—good banks are both solvent and liquid. August was not a normal time. The Fed was concerned with liquidity and reminded banks that it would assure liquidity by lending cash to banks through the discount window. However, by requiring collateral, the Fed would only lend to solvent banks.
Modern banks do more than collect deposits and make loans. Some modern borrowers use commercial paper (CP) or other instruments instead of bank loans, and some of the clearest signs of liquidity problems were in the CP markets. The markets are all interconnected and a squeeze in one will trigger a run in another. Borrowers depending on commercial paper for short-term financing who find it harder or more expensive to sell paper immediately draw down their backup bank lines of credit, adding to concerns about liquidity. The Fed's decision to lend should not have been a surprise—a central bank's key responsibility is to protect the financial system by insuring that liquidity is available in a crunch. Writing about the Bank of England some 140 years ago, Walter Bagehot's prescription 1 was, "Lend freely."
Let's move from antique money and banking to 2007 behavioral finance. What happens when there are creeping signs of financial crisis? Near panic, and a desire by everyone to convert every asset into cash. There is a flight to quality, T-bill yields plunge and prices of U.S. Treasuries climb as virtually everyone sells something or everything to raise money and buy Treasuries. When the world seems to be in turmoil, Treasuries and cold cash are one of the few things that warm investors' hearts. In a looming crisis, there seem to be only two kinds of assets—cash/Treasuries and everything else—and no one wants the everything else.
Of course, index users are long-term investors who believe they can and should ride out these storms. Among long-term investors, the latest bit of turmoil provided some special surprises for those engaged in the increasingly popular quest to capture uncorrelated returns through hedge funds. Hedge funds have grown in popularity and notoriety in the last few years, and are distinguished by employing various investment strategies not widely used by either traditionally oriented pension funds or mutual funds. Most strategies include taking long and short positions in various securities so that the expected results depend largely on the manager's skill in finding alpha rather than on beta generated by broad market movements. One of the key ideas of modern portfolio theory is that positions in uncorrelated assets will protect an investor in the kind of markets we saw this summer. The surprise of the last few months as market volatility climbed was that many hedge funds weren't so uncorrelated—either with each other or with the broader equity and debt markets.
Analysts are searching for explanations for why so many hedge funds were stuck in the same boat as it took on water.
Despite efforts to keep investment strategies secret, many funds seem to understand what their peers are buying and selling. Further, since many of the portfolio managers read the same books, follow the same blogs, went to the same schools and study the same market signals, some correlation is expected. This, plus growth in the number of hedge funds, increases the chances that numerous funds almost mimic one another, raising correlations.
There is, however, an even larger factor at work in the markets when volatility jumps, as it has recently: fear. When fear strikes, cash is king and everyone sells everything else to raise cash. The markets then become increasingly correlated, as Treasury prices rise and all other prices fall. If all markets move together, uncorrelated assets suddenly become correlated. The problem is the market, and not hedge fund strategies designed with data from more peaceful times.
The lesson we should relearn from this summer is that in times of trouble, markets change—instead of each segment following its own fundamentals and sentiments, all markets follow one sentiment—fear—and have one fundamental need—cash. When that happens, even the best-designed strategies fail because the data have changed. Keynes commented 2, "When the data change, I change my mind. What do you do?" One right answer might be to reconsider one's investment strategy.