Lessons of Past
May Offer Clues
To Market's Fate
By E.S. BROWNING
August 20, 2007; Page A1
As investors around the globe try to gauge whether this month's financial-market turmoil is a passing storm or a more-lasting disturbance, they are looking at two past periods of turbulence for signs of what could come next.
Those two periods -- the stock-market crash of 1987 and the downdraft of 1998 -- bear striking similarities to the present. They also provide insight into the role of the Federal Reserve, which bolstered markets Friday, sparking a rally.
In both 1987 and 1998, stocks fell sharply starting in July or August and, although markets seemed to stabilize by September, they abruptly plunged again, and didn't come out of their tailspins fully until October. Whether stocks will suffer a similar fate now, or escape it due at least in part to timely Fed action is the big question on investors' minds.
In the previous two cases and again this time around, market downturns turned into routs as computer-based stock-trading models blew up in the faces of the investors who used them.
The crash of 1987 took place during the original buyout boom, which bullish investors said would keep stock prices high for years. And 1998 was the year a big hedge fund called Long-Term Capital Management almost imploded, forcing the Fed to step in to calm credit markets.
Big corporate buyouts, such as Texas utility TXU Corp.'s agreement to be acquired for $32 billion by a group of private investors and student lender SLM Corp.'s deal to be acquired for $25 billion, have formed part of the backdrop to this year's trading. So, too, have battered hedge funds: Two funds at Bear Stearns Cos., for example, lost almost all their value after investing in securities linked to low-quality mortgages.
ln 1998, the Fed wound up intervening three times, because its first attempt proved insufficient. In 1987, the Fed didn't intervene until after the crash, although when it did step in, it succeeded in stanching the bleeding.
This time, as in 1998, the Fed has tried to intervene before things get worse. But unlike Fed Chairman Alan Greenspan in 1998, his successor Ben Bernanke has taken a gradualist approach. He has cut the rate the Fed charges to provide cash to the banking system. But he has avoided taking the action Mr. Greenspan took in 1998 -- cutting the more broadly influential federal-funds rate, which governs banks' prime lending rates. A cut in the federal-funds rate would more directly help small borrowers, but the Fed fears it also could rekindle inflation.
Yale economist Robert Shiller, whose book "Irrational Exuberance" appeared just in time to predict the bear market that began in 2000, sees unsettling parallels with 1987. Amid the stock declines earlier this month, he moved some of his savings out of stocks. He beefed up his safer investments in things such as Treasury bills, leaving about one-third of his money in the stock market. But he worries that he may have been hasty, and says he certainly wouldn't advise others to imitate him now.
The market's direction is much harder to read than it was in 1999 or 2000, Prof. Shiller says: "I'd say the odds are about 55 to 45" that the market's declines have farther to go.
As in 1987 and 1998, one of the most unsettling aspects of the past weeks' selloff is that stocks are falling and people don't fully understand why. A big reason in each case was the role of computers programmed by people who were supposed to be market geniuses.
This time it was hedge funds using mathematical models, whose forced selling contributed to huge market swings and massive trading volumes over the past few days. The hedge funds, many of whose models were strikingly similar, had to unwind unsuccessful trades involving millions of shares after troubles in the mortgage markets bled into the stock market. The sight of unknown sellers using computers to sell millions of shares of many different stocks, while buying millions of shares of other stocks, sowed panic among other investors.
"It caused so much mass buying and selling that the market couldn't easily handle it. It was almost like trying to get a stampede of elephants through a small door," says Gordon Fowler, chief investment officer at Philadelphia money-management firm Glenmede Trust, who witnessed similar breakdowns in computer models in 1987 and 1998.
The models on which investors relied in 1987 were known as portfolio insurance, and the idea seems surprisingly suspect in the light of history. The portfolio-insurance models called for investors to protect themselves from losses by making sales in stock-futures markets if their actual stock holdings fell a predetermined amount.
The models, which were based on detailed analyses of market history, didn't take into account what would happen if everyone using these models all tried to do it at the same time. Markets couldn't absorb all the sales demands, and the selling pressure helped cause the 1987 crash.
"The models du jour failed in both 1987 and 1998," says William Hackney, a managing partner at Atlanta Capital Management. He remembers being amazed at the stock gains earlier in 1987, despite rising interest rates and inflation. Only later did he realize that people had felt free to behave recklessly because they had "portfolio insurance."
Indeed, in all three years, market turmoil was made worse by overconfident investors using borrowed money. The unwinding of all of that leverage, or borrowing, can be brutal because as stock prices fall, investors are forced to sell stocks to pay back their loans, creating a downward spiral.
In 1998, the blowup that forced the Fed to act came at a multibillion-dollar hedge fund called Long-Term Capital Management, which had become the dominant player in the Treasury-bond market. In the summer of 1998, LTCM was making highly leveraged bets against Treasury bonds and in favor of other bonds, including those of Russia. Its models showed the risk of losing money that way to be minuscule. When Russia nonetheless defaulted on its debt payments, LTCM faced bankruptcy until the Fed helped persuade a group of banks and brokerage firms to rescue it.
For a while, the incident soured many investors on computer models and borrowed money, but with time, new models emerged and began showing large gains. In response, investors pulled billions in investments away from traditional "long-only" money managers, who simply tried to pick which stocks would do best. The latest vogue was for so-called quantitative, market-neutral hedge funds, which were supposed to avoid market gyrations by betting on gains in one large group of stocks, bonds or currencies, while simultaneously hedging their risks by betting on declines in other large groups.
"At dinners with other managers, I would hear people talk about how they have 'quant' models, or that they were long-short managers or market-neutral managers. They would look at me like I was a dinosaur," says Hersh Cohen, co-chief investment officer at ClearBridge Advisors in New York, which manages $100 billion in the traditional way, by actively choosing which stocks to buy. "Now I realize that what sounded impressive was not much more than a thing we saw played out in 1987 and 1998," says Mr. Cohen, who managed money during both of those crisis years.
Another big lesson of all three years is that it is the debt markets that often pose the biggest threats to stability in the stock market. In 1987, it was rising interest rates that eventually sent stocks plunging. In 1998, it was the risk that the collapse of LTCM could roil the bond markets. Lately, it has been securities backed by high-risk home mortgages that had found their way into a wide variety of investment portfolios. The mortgage securities were so widely held that, when they went bad, they caused credit markets to freeze up, hurting other, higher-grade bonds.
As in 1987, huge investment funds used junk bonds in recent years to take over companies once thought too big to be acquired. In 1987 and again this year, the market suffered once buyouts started to face trouble.
"My view of booms is that they generate laxity in standards for loans because there is a general sense of optimism. That is what we saw in the late 80s," says Yale's Prof. Shiller.
Also supporting stocks in 1987, as well as in recent months, were a strong world economy and healthy corporate profits, which investors expected to last for years. Both times, the sudden stock declines shook those hopes.
In all three cases, markets turned increasingly volatile. Sharp drops earlier on proved to be temporary affairs, helping create a false sense of optimism.
To be sure, there are also some big differences. One is what analysts call valuation -- stock prices compared to companies' underlying value. Valuation looked excessive in 1987, with stocks trading at more than 20 times corporate profits. Price/earnings ratios were similarly high in 1998. This time around, the ratios remain in the teens, close to the post-1945 average of about 16.
The real excess this time has been in lending markets, where investors bid up mortgage-related securities and junk bonds to unheard-of levels, and where investment banks invented novel bond-like securities.
One reassuring difference is that, in the previous crises, problems developed in the bedrock Treasury-bond market. This time, Treasury bonds remain an island of relative tranquility.
Another hopeful sign: In both 1987 and 1998, the market's woes were severe but brief. Despite the sharp market drops (more than 36% in 1987, and nearly 20% at one point in 1998), the Dow Jones Industrial Average finished both years with gains. It was up 2.26% in 1987 and 16.1% in 1998, and the Standard & Poor's 500-stock index was up both years as well.
Although memories of the 1987 crash still make some investors shiver, that day -- Oct. 19 -- actually marked the bottom of the bear market. The next day, stocks began to recover, beginning a new bull market. One reason for the quick rebounds was prompt action by the Fed, and another was that the economy avoided recession in both 1987 and 1998.
But the fast rebounds led to more excess, and more abuse of borrowed money. The buyout boom resumed in 1988 but went bust in 1989, when a proposed buyout of United Airlines fell through and the junk-bond market plunged. The Dow industrials fell into a bear market in 1990, as recession arrived.
The consequences of the post-1998 tech-stock excess were worse: the 2000-2002 bear market.
Besides valuation, Fed intervention and the economy, investors need to watch for other corporate and hedge-fund blowups to get an idea what might lie ahead. The longer it takes to work these problems out, the more likely credit and stock-market problems will be to spill over into overall economy, causing a slowdown, or worse, a recession.
"When the market does recover, it should recover quickly," more like 1987 and 1998 than 2000, says Burton Malkiel, a Princeton University economics professor. But Prof. Malkiel doesn't think the hedge-fund blowups are over yet.
"It is undoubtedly the case that we will be reading about more now," he says. "It is not just the Bear Stearns one and the Goldman Sachs one. Undoubtedly, a lot of those funds were more highly leveraged in some of those investments that now are no longer profitable. A number probably bought mortgage-backed securities that they didn't completely understand."
Write to E.S. Browning at jim.browning@wsj.com
|
