The Hunt for Black October
From the Magazine: Monday, September 10, 2007
With the anniversary of the worst one-day decline in U.S. stock market history approaching, MATTHEW REES sets out to find its cause. And determine whether it can happen again.
On October 19, 1987, a bizarre and completely unprecedented set of events unfolded in America’s financial markets. In New York, the brokerage firm Donaldson, Lufkin & Jenrette posted uniformed security guards outside its office, fearing armed conflict with its clients. In San Francisco, a precious-metals and foreign-currency firm sold its entire stock of Krugerrands and American Eagle gold coins. At the New York Stock Exchange, executives of a Las Vegas company called Jackpot Enterprises tried to celebrate their new listing amid reports that floor traders were fainting.
For the preceding seven weeks, the stock market had been skidding. Now, on this sunny Monday, it was on the verge of total collapse. When the day was over, the Dow Jones Industrial Average had lost more than 500 points, or 22.6 percent of its value—the equivalent of a drop of about 3000 points today. A half-trillion dollars in wealth disappeared overnight—equivalent at the time to the entire gross domestic product of France. On the heels of the decline, a recession was considered a near certainty and a depression a distinct possibility. After all, on the worst previous day, October 28, 1929, the market had dropped just 13 percent. Now, 58 years later, a New York bar was serving a mixed drink of dark rum and black Sambuca called “Black Monday,” the same term applied to the previous crash.
The size and speed of the 1987 decline were breathtaking. So was the very fact that it happened: the U.S. economy was strong, and there had been no major destabilizing events. No terrorist attack, no presidential assassination, no failure of a major bank or brokerage firm. While there was apprehension, only a few analysts were predicting a major market downturn, and no one called a crash of this size.
Studies of Black Monday soon became a growth business, but none of them ever provided a convincing explanation of precisely why the market fell so far so quickly. Today, 20 years later, we don’t know much more about the causes of the crash than we did when it happened. Once again, we could be caught by surprise.
I decided to talk with some of the country’s leading market watchers, as well as with those who participated in the 1987 events, to see how their thinking has evolved over the past two decades. What was the explanation for Black Monday? But before we look at why it happened, let’s look at what happened.
Storm Clouds Gather
On August 12, 1982, with the last vestiges of inflation being wrung out of the economy, the Dow Jones Industrial Average, the popular index based on the stock prices of 30 large U.S. companies, bottomed out at 776.92. The following day would mark the start of a new era: the Federal Reserve reduced the discount rate on loans to member banks, and the Dow rose 11 points (the equivalent of nearly 200 points today) and then kept rising and rising. Over the next five years, it more than tripled, reaching 2722 on August 25, 1987.
Studies of Black Monday soon became a growth business, but none ever provided a convincing explanation of precisely why the market fell so far so quickly.
But starting around Labor Day, “the storm clouds were visibly gathering,” recalls Gerald Corrigan, then president of the New York Federal Reserve Bank. He had already met with the new chairman of the Fed, Alan Greenspan, who had taken office on August 11, to discuss technical changes in the markets—linkages among cash, futures, and options. The Dow was experiencing regular swings of 40 or 50 points—equivalent to more than 200 points today—and institutional investors were increasingly on edge.
In August, the big institutions—pension funds, investment firms, mutual funds—had been net buyers of stocks, by an average of 2.8 million shares a day on the New York Stock Exchange (NYSE ). In September, they became net sellers, dumping 300,000 shares a day. By the first half of October, daily net sales had increased to 4.4 million shares.
The selling was reflected in the Dow’s performance. During the week beginning October 5, the index fell nearly 159 points—one of its worst weeks ever in terms of percentage decline. On Monday, October 12, Columbus Day, the Dow fell again, nearly 11 points, on light trading. The next day, the market staged a brief rally, rising 37 points, and confident analysts at Kidder, Peabody & Co. advised their clients that “the long-term bull market remains intact. On an intermediate-term basis, early 1988 should still witness 2900 on the Dow Jones Industrial Average.” That forecast proved about 50 percent too high.
On Wednesday, U.S. trade deficit figures were announced as being $1.5 billion higher than expected, setting in motion a domino effect that raised the yield on 30-year bonds above 10 percent for the first time in two years. Higher interest rates tend to spook the stock market, since they depress corporate profits and draw investors from equities to bonds. The day ended with the Dow down 95—at the time, the largest one-day point loss ever. “The rise and fall in the Dow numbers,” said Democratic Representative Ed Markey, “is increasingly so great that it’s possible that their sheer magnitude, coupled with the market’s ever-increasing velocity, could affect investor response and create a near free-fall situation that feeds on itself.’’ It was a prescient thought.
Thursday began with surprising calm. Traders held their collective breath, and just 30 minutes before the market closed, the Dow was off only 4 points. Then, suddenly, investors began dumping shares. Volume tripled over the next half-hour, and the Dow finished down 57 in the fourth-busiest session in New York Stock Exchange history.
Some analysts at the time blamed a late-day announcement by Chemical Bank that it was increasing its prime lending rate; others believed the fall was triggered by Treasury Secretary James Baker, who said that the United States could “accommodate further adjustments” in the value of the dollar—code for a willingness to accept further declines in the U.S. currency, a process that could make dollar-denominated assets, like stocks, less attractive to foreign buyers.
Whatever the cause, the Dow had now declined in eight of nine consecutive trading sessions and there was growing nervousness that the market was entering truly dangerous territory. “For the first time in more than five years, many securities-industry professionals see a significant reversal in the stock market’s direction,” said a front-page article in Friday’s Wall Street Journal, headlined “Uneasy Street.”
Confident analysts at Kidder, Peabody & Co. predicted that 'early 1988 should still witness 2900 on the Dow.' That forecast proved about 50 percent too high.
Still, traders were buoyed early Friday morning because, half a world away, Japan’s Nikkei index had closed down only two-tenths of a percentage point. But at 8:00 a.m., New York-time, details began emerging of an Iranian attack on an American-owned oil tanker. An even more ominous portent came from across the Atlantic: the London Stock Exchange was closed by a freak hurricane, with wind speeds measured at 94 miles per hour, higher than Britain had ever recorded.
The Dow, nonetheless, remained stable in the first few hours of trading. But around noon it started slipping, and by 2:00 p.m. it was down 85 points. It bumped along until 3:30 p.m., and then in 20 minutes it dropped another 50 points—only to regain about half that in the final ten minutes.
It was a day of milestones. An unprecedented 338 million shares were traded on the NYSE, and the Dow’s 108-point plunge (another record) was the exclamation point to what was the biggest one-week decline in the blue-chip index—10 percent—since May 1940, when the French were overrun by the Germans.
Adding to the overheated climate was growing friction between the United States and West Germany over currency valuations. Baker said on “Meet the Press” on Sunday, October 18, “We will not sit back in this country and watch surplus countries jack up their interest rates and squeeze growth worldwide on the expectation that the United States somehow will follow by raising its interest rates.” Questioned moments after the show (and off-camera) by an influential economist, Henry Kaufman, about the wisdom of making such a strong statement, Baker replied, “Henry, some things need to be said.”
With temperatures rising and fortunes falling, market watchers began looking for culprits. A New York Times article that Saturday was headlined, “Market’s Slide Aided by New Trading Tools.” One of those tools was something called portfolio insurance, about which, more later. Yet the conventional wisdom, expressed by The Times, was still sanguine. The newspaper reported that investment analysts and portfolio managers believed “there are few signs to suggest that the stock market’s recent decline is the start of something calamitous. The economy remains healthy; corporate profits are strong; inflation, though rising somewhat, remains relatively modest.”
But unknown to The Times was that the nation’s largest savings and loan, American Savings, was about to implode. Senior officials from the Federal Reserve, the Treasury, the Federal Deposit Insurance Corporation, and the Federal Home Loan Bank Board worked furiously through the weekend to identify another institution that would purchase the S&L because the failure was scheduled to be announced that Monday, October 19. A senior official who was involved in the discussions said later that “we saw that the market was poised for a massive sell-off, and we were really worried that this announced failure was going to trigger an explosion.”
In fact, the impending S&L failure would prove to be a non-event, overrun by a roaring conflagration already underway. Having experienced nearly five years of bull markets, small investors were in a full-fledged panic over the weekend. Phone lines at mutual funds were overwhelmed by sellers. Eighty thousand had called Fidelity alone. Joseph Nocera later wrote, “The people who ran the likes of Fidelity and Schwab and Vanguard knew—absolutely knew—what was going to happen next.”
The Blackest Monday
Monday, October 19, was set to be a day of chaos. Before trading began, one additional dose of volatility was injected into the system: In response to Iran’s attack on an American oil tanker the previous week, the U.S. Navy destroyed two Iranian oil platforms that were being used as military staging sites. Traders knew that there would be devastation on Monday morning; they simply wanted to limit the damage.
An even more ominous portent came from across the Atlantic: the London Stock Exchange was closed by a freak hurricane, with wind speeds measured at 94 miles per hour.
The moment the market opened at 9:30, a whopping $500 million in sell orders were submitted, and another $475 million in the next half hour. The NYSE specialists, the floor traders who had the responsibility to make an orderly market by purchasing shares on their own account when they could not find other buyers, were immediately overwhelmed. Without buyers to match sellers, nearly 200 stocks—one-twelfth of the NYSE ’s listings—could not open for trading. It was a development without precedent in the Exchange’s modern history. By 10:00 a.m., the Dow was already down 68 points, and a half-hour later, 11 of the 30 companies making up the index were still closed. Eastman Kodak, which had a sell imbalance (that is, more shares put up for sale than for purchase) of 257,800 shares, did not open until 10:40 a.m., at which time its value promptly fell 15 percent. Exxon, with a sell imbalance of 361,200 shares, opened at 10:47 a.m.
By 11:00 a.m., the Dow was down more than 200 points, nearly double the record decline for a full trading day. But soon, the market rallied, and the Dow recovered about half its loss. The rebound encouraged some analysts to believe that a crisis had been averted. Greenspan, for example, decided to board a commercial flight to Dallas for dinner with British Prime Minister Margaret Thatcher and a speech to the American Bankers Association. The rally, however, soon died. By 2:05 p.m., the Dow was down 277 points, or more than 10 percent. Again, there was a brief upturn, but it did not last. In the final hour of trading, the index plunged another 200 points.
Because Greenspan had no access to a phone on his flight, he had no way to know what was unfolding in New York. He was met at the gate by a representative of the Fed’s Dallas branch and quickly asked how the market had fared. “It was down five oh eight,” he was told. Greenspan was relieved. He thought the local Fed man meant that the Dow had fallen 5.08 points.
Greenspan quickly learned that the Dow had fallen 508—a decline, in terms of points, that was nearly five times greater than the old record, as well as by far the biggest percentage loss in history. More than 604 million shares had traded hands on the NYSE, shattering the previous record of 338 million, which had been set just the previous trading day.
The stock market’s loss was broad and deep, afflicting America’s most recognized companies across all sectors. Here is a sampling of the one-day declines, rounded to the nearest point:
General Electric…………50 to 41
BankAmerica……………11 to 8
Digital Equipment………172 to 130
Procter & Gamble………..84 to 61
AT&T…………………….30 to 24
CBS………………………195 to 150
Exxon…………………….44 to 33
IBM………………………134 to 103
Eastman Kodak…………..89 to 63
International Paper………46 to 34
Goodyear………………...59 to 42
Westinghouse……………60 to 40
As John Phelan, president of the New York Stock Exchange, put it: “I wouldn’t want to be around for a day worse than this.” George Soros’s Quantum Fund lost 30 percent of its value, or $800 million. But the California savings and loan that was set to declare bankruptcy suddenly found new life, and the bailout was postponed indefinitely. Its assets rose as its bond portfolio’s value shot up; investors who were bailing out of stocks were moving into bonds, boosting prices and driving down interest rates.
In a single day, the Dow had dropped from 2247 to 1739. On Tuesday and Wednesday, stocks rallied, and the Dow jumped to 2028. The index began to lose steam again, then slowly climbed. It took two years for the Dow to get back to its high of August 1987. By July 1990, the index was over 3000, and in 1997 it breached 8000—a four-and-a-half-fold increase since Black Monday, not including dividends. An effective long-term strategy, as usual, was sitting tight.
Meanwhile, the expected recession never happened. The U.S. economy absorbed the stock-market shock and grew more than 3 percent after inflation in 1987 and more than 4 percent in 1988.
But what was it that caused the worst one-day decline in U.S. history? A post-crash report by the Securities and Exchange Commission concluded, “We may never know what precise combination of investor psychology, economic developments and trading technologies caused the events of October.”
But that hasn’t stopped people from trying. Here, then, are seven of the best explanations….
1) An instant correction
Testifying before the Senate Banking Committee in early 1988, Greenspan offered a simple explanation for the crash: “Something had to snap. If it didn’t happen in October, it would have happened soon thereafter. The immediate cause of the break was incidental. The market plunge was an accident waiting to happen.”
Greenspan is not alone in believing that the crash was simply a long-overdue market correction. The head of research at Goldman Sachs characterized the market as “the most overvalued it has ever been in the postwar period.” The editor of the Elliott Wave Theorist newsletter, Robert Prechter, a market technician who was very influential at the time, declared two weeks before the crash that stocks were overvalued and investors should sell—a forecast George Soros later cited as “the crack that started the avalanche.”
The New York Times reported, on the eve of the crash, that analysts believed 'there are few signs to suggest that the stock market's recent decline is the start of something calamitous. The economy remains healthy.'
Evidence of irrational exuberance could be found in price-to-earnings ratios (the number of dollars it takes to buy a dollar of a company’s annual profits). For all U.S. stocks, the P/E before the crash averaged 23—the highest level since World War II and particularly troubling at a time when interest rates on bonds were so high. The earnings yield on stocks (earnings divided by price) was only 4.3 percent, evidence of severe overvaluation when compared to a 10 percent yield for long-term Treasury bonds. By contrast, in early August 2007, the earnings yield on stocks was 5.8 percent, compared with a yield on 30-year bonds of 4.9 percent.
The Dow had risen 44 percent from the start of 1987 until its peak in August—part of the most persistent and powerful bull market in 50 years. Even the optimists at Fortune published a cover story in September headlined, “Are Stocks Too High?” The article quoted Soros saying, “Just because the market is overvalued does not mean it is not sustainable.”
Bears were starting to come out of hibernation, grumbling about a doubling in the price of crude oil over the previous year, coupled with emerging news of savings-and-loan failures and rising interest rates. Says James Grant, editor of Grant’s Interest Rate Observer, in retrospect: “You’d look at the rise in rate of inflation, the unstable currency situation, and the new personnel at the Fed, and there were some reasons to be bearish. The market was trading on vapors of what the brokers like to call ‘liquidity.’” Robert Glauber, who served as executive director of the Brady Commission, set up in the wake of the crash to probe its causes, agrees: “There was a lot of tinder waiting to be ignited by a match.”
Further proof of the crash-as-correction theory is the market’s performance in the period following Black Monday. Had the crash been driven by something other than economic fundamentals, says Jack Katz, who spent 20 years at the SEC, it would have recovered much sooner. “The market was repricing itself on October 19,” he says, “but instead of a 20 percent decline over a few weeks or a few months, it all happened in one day.”
Traders knew that there would be devastation on Monday morning; they simply wanted to limit the damage.
One of the more compelling arguments for this theory was published in the Harvard Business Review in 1988. The authors—Avner Arbel, Steven Carvell, and Erik Postnieks (all professors at Cornell)—wrote that the crash “was almost nothing of what so many analysts, investors, and observers believe it was. Instead of a panic, it was the restoration of sobriety and rationality.” The authors applied a pricing formula developed by Benjamin Graham, the Columbia University scholar and mentor to Warren Buffett, to 1,800 publicly traded companies. They looked at the companies’ valuations a year before the crash, two weeks before, and the day after. Their model indicated the overall market (not just the 30 stocks of the Dow) was overvalued by about 17 percent. The Wilshire 5000, an index that encompasses virtually every publicly traded stock, declined 18 percent. The title of their article: “The Smart Crash of October 19th.”
2) Washington did it
James Chanos, the New York investor who specializes in finding overvalued companies and shorting their shares (borrowing stock in hopes of returning it when it is worth less), was the first on Wall Street to identify and exploit the problems of a Houston energy company called Enron. When he looks back at the 1987 crash, he fingers Congress.
Late on October 13, Democrats on the tax-writing House Ways and Means Committee announced they had reached agreement on a measure that would limit interest deductions on debt used to finance corporate mergers and acquisitions. Chanos believes this little provision, approved on the Thursday before Black Monday, was a bombshell. “So much of the market growth was driven by takeovers,” says Chanos today, “that anything that would have burst that bubble could have been a culprit.”
Shares of companies that were targets of takeover activity—for example, Gillette, Tenneco, and Dayton Hudson—were devastated by the crash. A report by the SEC ’s Office of Economic Analysis later identified the bill, authored by Ways and Means chairman Dan Rostenkowski, as “a fundamental economic event” contributing to the market decline on October 14 and on the days that followed. The Wall Street Journal quoted a Washington-based merger specialist calling the measure “profoundly important…. [I]ts impact is extraordinary—I mean, unparalleled.” A month after the crash, Carl Icahn, the takeover specialist, described the bill as “the match that ignited the dynamite.”
3) Technology failed
During a market meltdown, the only thing worse than negative information is unreliable information—or no information at all. And on October 19, the mechanisms responsible for processing trades and transmitting information about market movements broke down, intensifying the scramble for the exits. The economist Frank Knight had pointed out in 1921 that, in conditions of severe uncertainty about what is happening in a game, the common response is to abandon the field of play. In the stock game, you quit by selling. And on October 19, everyone wanted to sell at once.
Since 1976, the New York Stock Exchange had used a trading system called Designated Order Turnaround, or DOT , which could be used for buying, or selling, up to 2,099 shares at the time. Here’s the SEC ’s description of how it worked:
“A card printer at the appropriate specialist post prints an incoming order on a machine-readable mark-sense card. The order is handled by a clerk, passed to the specialist who executes the order and returns it to the clerk; the clerk marks on the mark-sense card the turnaround code, price, number of shares, and the name of the broker on the other side of the trade. The card is then placed in a card reader connected to the CMS [‘Common Message Switch’—a tool for routing trading orders]. Using the turnaround number, the system retrieves the original order data, and generates and transmits an execution report to the member firm.
“The card printer can print out only a limited number of messages per second; thus it is a critical stress point in a busy market. At each of the 14 trading posts, there are eight active printers and two emergency printers…. Each printer, which generally supports two or three specialists, is capable of printing only 10 to 12 messages per minute, depending on the length and type of message.”
While this system sounds hopelessly antiquated, convoluted, and labor-intensive from today’s perspective, the DOT managed to process an average of 138,000 orders per day in the month before the crash. But on October 19, there were more than 471,000 orders, and the system was overwhelmed, acknowledges Catherine Kinney, who was in charge of it. “We didn’t have the capacity to keep up with demand during a meltdown.”
Barron’s, the Wall Street weekly, later reported that “the DOT was barely functioning for much of the day.” The result was long delays in processing orders for trades. By noon, the delays were anywhere from 45 to 75 minutes. The effect was toxic, said the Brady Commission report: “Market orders were executed at prices often very different from those in effect when” they were placed. If you put in a sell order at $25 a share, you might have received only $22. The shortfall sent a clear signal: get out now or prices will be even lower. Says Patrick Durkin, a senior staffer on the Brady Commission, “The market is dependent on information processing. If those systems fail you, you have no way to create any type of rational trading.”
The federal government’s watchdog agency, the General Accounting Office, later found that the DOT came to a complete halt four times on October 19 for intervals of three to 13 minutes, intensifying the decline by adding “uncertainty” and “confusion” to the market. And when the Brady Commission took inventory on the DOT ’s performance, it found that orders for 112 million shares were never executed—more than one-fourth of the total value of stock that had been routed through the system. The DOT was so strained that the day after the crash, the NYSE asked traders who employed computer-driven strategies not to use the system, to prevent a rerun of the previous day’s events.
4) Blame an ‘international panic’
When the market was closing on Black Monday in New York, it was already Black Tuesday in Tokyo. About six hours later, Japanese stocks started trading—and falling.
By the end of the day, the Nikkei index had dropped 15 percent—the biggest one-day decline since its creation in 1949. In Hong Kong, the index plunged 45 percent, causing the market there to be shut for nearly a week. The FTSE, London’s benchmark, was down 12 percent. Of 23 major markets, 19 lost at least one-fifth of their value during the terrible month of October.
None of these markets was handicapped by portfolio insurance, program trading, or poor-performing computers. Nor were all of them overvalued to the extent of the U.S. market. As a result, some analysts have concluded that the real cause of the crash was currency instability throughout the world.
Paul Volcker, whose tenure at the Fed ended two months before Black Monday, has written that the crash can be traced to “the feeling that the promising efforts toward coordination of economic and exchange rate policies were breaking down.” The editorial page of The Wall Street Journal was a vocal proponent of the same idea, citing the brewing feud between the Bundesbank and Baker, the Treasury secretary, as well as the longer-term problem of the dollar’s continued depreciation. “We think the threat the market saw was a dollar collapse, the severing of international cooperation and the victory of protectionism around the world,” said a Journal editorial on the first anniversary of Black Monday. Baker, in an interview with Time magazine soon after the crash, argued that the event was “triggered” by a front-page New York Times story on October 18 in which a “senior administration official” said the U.S. was allowing the dollar to decline against the West German mark. This, said Baker, “could not but contribute to market nervousness.” (Many believed the “senior administration official” to have been Baker, who had been blamed for talking down the dollar, but Baker strongly denied being the source for the Times story.)
Lawrence Kudlow, the CNBC host who, at the time, was chief economist at the investment firm Bear Stearns, says that “most of the traders on Wall Street believed the dollar was low enough. And they thought if it went lower, it would resurrect 1970s-style inflation. So they were rooting against Baker.” The currency war, he says, “was another nail in the market’s coffin.”
Alan Reynolds, now a senior fellow at the Cato Institute, has pointed out that the market decline was part of a pattern that started months earlier in other countries. While the U.S. stock market did not hit its high until late August, Germany peaked on April 18 and Britain on July 25. Reynolds also cites instability brought on by threats of protectionism, including the Reagan administration’s use of trade sanctions against Japan and the approval by the House of Representatives of an amendment by Representative Richard Gephardt to impose severe sanctions on countries running “excessive and unwarranted” trade surpluses with the United States. (Protectionism has also been blamed for the 1929 crash.) Connecting the problems abroad with the scrum between Baker and the Bundesbank, Reynolds concluded the crash was part of an “international financial panic.” In this view, global forces overwhelmed the good news at home (solid economic growth, a lower-than-expected budget deficit, and moderate interest rates).
If the catalyst for the crash was international finance, the debate over whom specifically to blame continues. Earlier this year, Helmut Schlesinger, who was vice president of Germany’s central bank in October 1987 (and later its president), wrote a letter to The Financial Times disputing a columnist’s claim that his comments started the tailspin. In fact, wrote Schlesinger, “the crash was the consequence of an attack on the Bundesbank by the very respected James Baker.”
5) The insurance that wasn’t
Among all the possible explanations of the crash, the most popular focuses on the meteoric rise of a financial strategy that at the time was little known and is now largely defunct: portfolio insurance.
Its inventor, Hayne Leland, a professor of finance and management at Berkeley, said he first conceived of the notion during a sleepless night in 1976. The underlying idea was the limit order, which calls for a stock to be bought or sold once it reaches a price predetermined by an investor. Thus, if you bought a stock at 40 and have watched happily as it rises to 60, you might place a limit order with your broker to sell the stock automatically if it drops to 55. But portfolio insurance was for large institutional investors, and Leland’s concept was not to sell the actual stocks, but rather to trade index futures contracts, which employed an enormous amount of leverage, so investors, for small sums, could control thousands of dollars of market exposure. The contracts were often sold to Chicago Mercantile Exchange floor traders acting as index arbitrageurs who, in turn, offset their long futures contracts’ position by selling bundles of S&P 500 stocks. The trader-arbitrageurs hoped to profit from the price difference between the futures contract price and the price of the stock bundle. In this manner, the portfolio insurance S&P 500 futures contract sale on the CME was transmitted to the sale of S&P 500 stocks on the NYSE.
As a result, institutions—primarily pension funds that managed billions of dollars in assets for workers and retirees—could limit their downside risk through automatic selling. They were, in effect, taking out an insurance policy. Peter Bernstein, who has written extensively about risk, dedicated an entire chapter to portfolio insurance in his book Capital Ideas and titled it, “The Ultimate Invention.”
Bernstein was hardly alone in his enthusiasm. Leland and his partners, John O’Brien and Mark Rubinstein, signed up their first client only in 1982. But by 1987, they were insuring $5 billion in assets, their licensees were insuring another $45 billion, and copycats were insuring an estimated $50 billion more. In those days, $100 billion was real money—more than 2 percent of total U.S. economic output.
The buyers of portfolio insurance included many of the country’s most important companies and their pension funds, including Boeing, Caterpillar, Chrysler, and TWA. Firms that sold the insurance were also distinguished: Aetna, Bankers Trust, JPM organ, Morgan Stanley, and Wells Fargo.
Portfolio insurance was not, however, without its critics. In 1986, John Phelan worried that the innovation could lead to a “financial meltdown.” Richard Ketchum, who in October 1987 was the SEC ’s chief of market regulation, has described portfolio insurance as “a classic example of a strategy that should have been viewed as art but instead was turned into science.”
Ketchum says that portfolio insurance convinced investors “they could buy protection and risk controls at a fraction of what they were paying before, so they could leverage themselves much more than in the past.” But this increased leverage itself brought more risk into the marketplace. With leverage, small price movements were amplified, and no one knew whether the futures contracts themselves could become unmoored from the stocks to which they were supposed to be tethered.
As the crash approached, the strategy had been employed for five years, but it had never had much of a stress test. Still, even as the market declined from late August into October, portfolio insurance programs were holding up. But, exactly a week before Black Monday, a Wall Street Journal article painted a frightening picture of how the new programs could provoke a market slide:
“The stock market goes into a tailspin for a fundamental reason such as a plunge in the dollar, the Federal Reserve pushes up short-term rates, oil prices skyrocket. Portfolio insurers sell stock-index futures contracts to reduce their exposure to the equity market. Money managers who pattern their portfolios on the S&P 500 stock index decide to buy the now cheaper stock-index futures contracts and sell the stocks that make up the index. Their selling further depresses stock prices, which causes the portfolio insurers to sell more futures, and the cycle can keep repeating.”
A few days later, that’s almost exactly what happened. As markets tumbled between Wednesday and Friday, portfolio insurance programs began kicking in, leading to what the SEC would later identify as “significant” amounts of selling of S&P futures. The price of futures contracts on the S&P 500 managed to stay closely aligned with the actual value of the shares in the S&P 500 until late Friday, but as more and more portfolio insurers began trying to sell their contracts in the futures market, they weren’t finding buyers.
In practice, portfolio insurance turned out not to be true insurance. If your house burns down, for example, the insurance company has the cash to pay you. For portfolio insurance to protect you, however, it required buyers of futures contracts to provide the cash. When buyers could not be found, insurers started to dump individual stocks in equity markets. This was a red flag to insiders that there was trouble ahead.
The system broke down. The firms managing insurance programs were able to sell less than one-third of the futures contracts they needed to fulfill their program mandates. In the measured words of the SEC, the significant overhang of uncompleted sales, coupled with investor knowledge of the problem, “may have contributed to the rapid decline on the afternoon of the 16th and undoubtedly added to the selling pressure on the 19th.”
Futures prices dropped dramatically. This drop, in turn, seemed to be telling investors that underlying equity prices would fall, too. And with portfolio insurers trading in the same direction as the market—selling as the market falls—all signals were negative. Here is how a Morgan Stanley banker, involved with portfolio insurance in 1987, describes the market psychology on Black Monday:
“A price drop is normally the dinner bell for buyers. So a precipitous drop should have had traders licking their chops. But it doesn’t work that way; if prices drop too far and too fast, it sends the wrong signal…. [Traders] wondered if it could be the result of some new market information. They viewed themselves at a disadvantage and elected to stay out of the market.
“Replay this portfolio manager’s reaction over many times and you basically have the fault line of the crash of 1987. Selling demand increased as prices dropped because of the pre-wired hedging rules of the portfolio insurance programs. Supply dried up because of the difference in time frames between the demanders and suppliers. By the time equity investors could have reacted to the prices and done some bargain hunting the specialists had moved prices so precipitously that these potential liquidity suppliers were scared away.”
But Leland’s partner John O’Brien says the problem wasn’t with portfolio insurance, but rather with what is called “front-running” among hedge funds, money-managers, and investment banks. O’Brien says that these investors could see the pressures facing the portfolio insurers and that they began selling before the insurers did. The front-runners’ selling produced even more aggressive selling by the portfolio insurers—which, in turn, contributed to the huge trading volume early on Black Monday. “If you’re looking for a trigger,” O’Brien said in a May interview, “this was it.”
But the verdict that portfolio insurance was the guilty party was far from unanimous. Among the luminaries who disagreed were Peter Bernstein and the late Nobel laureate Merton Miller. They wrote that the volume of portfolio insurance–driven sales was dwarfed by the volume of sales by other investors, including small individual shareholders and mutual funds acting on their behalf.
Bernstein, for example, noted that the volume of block trading by institutions (trades averaging about 25,000 shares) was only two percentage points higher on Black Monday than the mean for the preceding 50 days. The bigger culprit, says Bernstein, was the volume of smaller trades—triple the average of the previous year. This trading torrent, he says, was what “overwhelmed the whole communications network and explains…the darkened computer screens that intensified the chaos and fed investor panic.”
6) Loose lips sink markets
Some analysts maintain that one individual, with a single poorly timed and ill-advised comment, injected an element of extreme uncertainty which, added to the other fears that day, sent stocks into a tailspin.
On August 7, 1987, David Ruder, the low-key dean of Northwestern University’s law school, was sworn in as chairman of the Securities and Exchange Commission. On the morning of October 19, Ruder delivered a speech at the Mayflower Hotel in Washington, to a conference sponsored by the American Stock Exchange. No one remembers the speech.
When it was over, at 11:15 a.m., Ruder was besieged by reporters hungry for a comment about the Dow, which was down 200 points. Rather than issuing a simple “no comment,” Ruder went into professorial mode, and in an ill-advised burst of candor he told the journalists, “I’m not afraid to say that there is some point—but I don’t know what that point is—that I would be quite anxious to talk to the New York Stock Exchange about a temporary, very temporary, halt of trading. It may or may not be the time to do it. It may not be the policy which is effective. And our tentative proposals have not talked about any kind of long-term trading policy—only the very smallest, perhaps one half an hour or less, trading halt, in order to give the markets a chance to sort things out.”
For panicky investors, the prospect of a halt in trading—at a time when there were no circuit breakers, or breaks that had been set in advance if certain market milestones were reached—was another incentive to sell shares. Even if Ruder was only talking about a half hour, the halt could easily be prolonged. It might be hours or days before selling would be possible again. And Ruder’s extemporaneous, almost casual commentary could hardly inspire confidence. Indeed, it raised the question of whether anyone was in charge.
Ruder’s comment was not immediately reported—probably because the Dow was then on an upswing, so reporters and editors may have assumed there was no threat of a halt in trading anyway. But once stocks took another dive, Ruder’s off-the-cuff remarks, which he later claimed were misunderstood, were relevant again, and they hit the news wires at 1:04 p.m.
The SEC, recognizing the looming damage, issued a statement 21 minutes later, saying that the commission “is not discussing closing the nation’s securities markets.” But this trade had been made. The Wall Street Journal later quoted a floor trader from the Chicago Mercantile Exchange saying that, upon hearing what Ruder had said, he immediately began selling Standard & Poor’s index contracts, believing that “all hell’s going to break out.” And it did. Coincidence or not, the market never recovered that day after Ruder’s comment became public.
7) Sheer fear
The simplest explanation of the crash has little, if anything, to do with economic fundamentals, investment strategies, or gaffes. According to this school of thought, the cause was something more primal: sheer fear.
On October 19, and a few days after, Robert Shiller, a Yale economist who specializes in behavioral finance, sent questionnaires to a total of 3,000 institutional investors and wealthy individuals with large stakes in the stock market. He asked respondents to rank ten recent news stories according to how much the events reported affected their “evaluation of stock market prospects.”
The number-one factor was early news of the decline on October 19. Close behind was news of steep declines the previous week. “Black Monday,” wrote Shiller, “is best explained as a vicious circle—price declines feeding on previous price declines. There appear to be no other forcing events, other than previous price declines themselves, that
The best explanation
So what’s the answer? Despite Shiller’s claim, the crash almost certainly had multiple causes, several of them working in concert. But the most convincing explanation was given short shrift by many of the studies that followed Black Monday. The main cause, in my view, was the failure of technology (explanation number 3)—that is, a thoroughly inadequate market infrastructure.
Markets require prices to operate efficiently. But on the day of the crash, the trading systems became overwhelmed, and investors and speculators—particularly those engaged in index arbitrage—were unable to determine the price at which their trades would be executed. So, quite rationally, they stopped buying, particularly in the futures market, and a large gap developed between the low price of those futures and the higher price of the individual stocks that made up the baskets that comprised the futures. “As futures kept plunging,” said The Wall Street Journal two months after the crash, “they exerted a suction-like force on stock prices.”
How much did the market fall because of inaccurate pricing information? We’ll never know, but the good news is that trading systems have vastly improved over the past two decades. “The infrastructure is much more robust than it was in 1987,” says Corrigan, who headed the New York Fed from 1985 to 1993. There is greater transparency, better architecture, and improved supervision. New technologies can keep up with a vastly greater volume of trading than what swamped the markets 20 years ago. (It’s not unusual for 3 billion shares to trade each day on the NYSE—more than ten times the level of trading in 1987.) Investors understand their exposure better and have improved capital structures. And there have been enormous improvements in risk management. Last year’s collapse of Amaranth, a $6 billion hedge fund, registered barely a hiccup in the financial markets. And the Dow dropped only 7 percent on its first full day of trading after exchanges were closed for a full week after the 9/11 attacks.
In the aftermath of the crash, the New York Stock Exchange instituted circuit breakers—a temporary trading halt if the market drops 10 percent, a halt that is a bit longer if the market drops 20 percent, and a closure of the market if the decline is 30 percent. Trading halts, of course, can add fuel to a market conflagration, and, at best, they are widely recognized as an imperfect solution. But they seem to work, and no one in the past two decades has come up with anything better.
Still, technological risks remain. On the afternoon of February 27, 2007, the computer systems used to compute the Dow Jones Industrial Average were overwhelmed by heavy trading. As a result, the price of the Dow lagged behind the actual prices of its 30 components for over an hour. When the problem was fixed, the market index dropped 178 points, in one minute. The same day, electronic trading was halted at the NYSE, when buy and sell orders became backlogged. The total loss for the Dow that day was 416 points, and it was the fifth consecutive trading day on which the index declined. But the markets bounced back quickly. Fear did not beget fear—at least not in this case. Five months later, on July 27, the NYSE handled a trading volume of more than 4 billion shares. The Dow fell 208 points, but trading was orderly.
Manley Johnson, vice chairman of the Federal Reserve at the time of the crash, points out, “We’re more vulnerable to international turmoil because we’re more integrated with the global economy. Given the speed at which information and capital can now move, it doesn’t take as much to make markets nervous.” He adds that “the best policy weapon central banks have today is public credibility.” True enough, but David Ruder’s comments 20 years ago are a reminder that it doesn’t take much to rattle the markets.
Much of the worry about a cataclysm today focuses on the credit markets, where complex derivatives attempt to reduce risk but actually may heighten it. Still, catastrophe often strikes where and when it is least expected. Could today’s infrastructure handle a massive global stock market panic? The frightening truth is that no one really knows.
Matthew Rees is currently a senior director at The White House Writers Group, in Washington. He was formerly chief speechwriter for the chairman of the SEC and has written for The Weekly Standard, New York Times, and New Republic. His pieces on Mitt Romney and “Washington’s Five Greatest Hits” have appeared in previous issues of The American.
Image credits: Photo by flickr user Chris&AmyCate. Chart illustration by Peter Hoey.
Image credits: Photo by flickr user Chris&AmyCate. Chart illustration by Peter Hoey.