Fortune's Formula (21 page)

F
EW THEORETICAL FINDINGS
changed finance so greatly as the Black-Scholes formula. Texas Instruments soon offered a handheld calculator with the formula programmed in. The market in options, warrants, and convertible bonds became more efficient. This made it harder for people like Thorp to find arbitrage opportunities.

Of necessity, Thorp was constantly moving from one type of trade to another. In 1974 Thorp and Regan changed the name of their fund to Princeton-Newport Partners, a name steeped in the Ivy League and East Coast old money. The Newport was not the one in Rhode Island, of course, but Newport Beach, California. The Princeton was not the university but the town. Regan preferred the commute into Princeton to the more hectic one into Manhattan. Thorp and Regan also set up a firm called Oakley Sutton Management (after the partners’ middle names) to hire employees and create a brokerage subsidiary in order to save on some commission costs.

In 1972, 1973, and 1974, the fund posted net returns on investment of 12.08 percent, 6.46 percent, and 9.00 percent. This demonstrated the value of being market neutral. The stock market declined steeply in 1973 and 1974. By the end of 1974, the fund was just shy of having doubled its original investors’ money. Thorp and Regan were then managing $20 million in assets.

It was hard to keep that kind of success under wraps. On September 23, 1974,
The Wall Street Journal
ran a front-page profile of Thorp and Princeton-Newport. It began with the idiosyncratic poetry of
Journal
headlines:

The
Journal
writer was amazed at Thorp’s disregard of fundamental analysis and his reliance on computers. In 1974 the
Journal
’s average reader had as much hands-on experience with computers as with moon rockets. A computer was something you saw in a movie (often it went berserk and killed people).

The
Journal
linked Thorp’s operation to “an incipient but growing switch in money management to a quantitative, mechanistic approach.” It mentioned that the Black-Scholes formula was being used by at least two big Wall Street houses (Goldman Sachs and Donaldson, Lufkin & Jenrette). The latter’s Mike Gladstein offered the defensive comment that the brainy formula was “just one of many tools” they used.

“The whole computer-model bit is ridiculous because the real investment world is too complicated to be reduced to a model,” an unnamed mutual fund manager was quoted as saying. “You just can’t replace the money manager using security analysis and market feel with a machine.”

Yet the
Journal
reported that Thorp’s “machine” outperformed all but one of the 400-some mutual funds tracked by Standard & Poor’s. Said Thorp, “The better one was one of those crazy funds that invested in only gold stocks.”

Thorp computed that out of 200 hedged trades he completed for a pension fund client, 190 made a profit, 6 broke even, and 4 lost. The losses ranged up to 15 percent of the value of the long side of the deal. One of the worst things that can go wrong is for the company to go bankrupt. Thorp had a $250,000 convertible bond hedge involving U.S. Financial Corp. When the company filed for Chapter 11, Thorp’s fund lost $107,000.

Another problem was the one that sunk Merton, Black, and Scholes’s warrant experiment. Princeton-Newport took a proactive approach, phoning companies’ attorneys to try and get a line on whether they were planning to change the rules.

W
ILLIAM
S
HARPE WAS ONE
of the brightest and most militant of the Random Walk Cosa Nostra. He would go around asking money managers if they
really
beat the market. They would usually huff and say they did; then Sharpe would turn prosecuting attorney and grill them over the details. Sharpe subscribed to the view that successful portfolio managers are like successful astrologists—good at convincing the wealthy and gullible that their services are valuable.

For two years Sharpe was a professor at UC Irvine. He came to know Thorp, and they had a number of friendly parries over market efficiency. At Irvine, Sharpe was working on the theory that would make him famous, the Capital Asset Pricing Model.

Sharpe moved to Stanford. In 1975 Thorp invited him back to UC Irvine to lecture. During the visit, Thorp tried again to win Sharpe over to his position. Thorp had just been starting out as a market-beating(?) investor when Sharpe taught at UC Irvine. Now he had a track record.

Thorp described some of the trades he’d made to Sharpe. One was a 1974 trade in an American Motors Corporation (AMC) convertible bond maturing in 1988. Issued at $1,000, the bond had sunk to $600. That gave it a high return—it was a convertible junk bond. The bond could be exchanged for 100 shares of AMC stock. The stock was then selling for $6 a share. The bond therefore sold for exactly the same price as the stock you could get by converting it.

That was insane
, Thorp realized. The bond paid 5 percent interest. The stock paid no dividends. Owning the bond gave all the upside potential of owning the stock. If the stock went up, you could always convert the bond to the stock. But there was no rush! The bond-holder collected interest and was insulated from the downside potential of the stock. Someone patient enough to hold the bond until the 1988 maturity was guaranteed the $1,000 repayment of the original loan.

Thorp bought the convertible bond and sold short AMC stock. What could go wrong? The company could go under. Thorp would make
more
money if that happened. In case of bankruptcy, the company would be liquidated and the proceeds distributed to the bondholders. There probably wouldn’t be enough to pay off the bonds in full. That means there would be nothing left over for the stockholders. AMC stock would be worthless. Bankruptcy would therefore hurt the bondholders, but it would hurt the stockholders a lot more. This would be good for someone who owned the bonds and had sold short the stock.

The true worst-case scenario was for the stock to stay exactly where it was. In that case, Thorp still made a decent return. The AMC notes were paying 8.33 percent. Thorp had borrowed at 8 percent to buy them, earning a net 0.33 percent. But since Thorp had sold the AMC stock short, he already had the cash and was able to lend it out at 6 percent. He was making a net 6.33 percent return even if the stock did nothing.

Because this trade was a sure thing with no risk of ruin, the Kelly system permitted leverage. Thorp added borrowed capital to multiply his profits. “Situations that simple and clear are few and far between,” Thorp explained, “but we made a large part of our living off scenarios just like that.”

Sharpe was unconvinced. There are anomalies challenging every scientific theory ever propounded. It is a tough call knowing which to take seriously and which to shrug off.

Efficient market theorists claim that the market can act as if it were more rational than many of its participants. One mechanism of that is arbitrageurs—such as Thorp—who step in to make a profit whenever prices start to get out of line.

The efficient market people generally suppose that it is such a cinch to exploit arbitrage opportunities that prices never get significantly out of line for long. Thorp’s experience differed. He had learned that arbitrageurs were often constrained by trading costs, the supply of mispriced securities, the Kelly formula, and other factors. It took weeks, months, and more for mispricings to diminish, even with Thorp trying to profit from them at the very mathematical maximum rate.

Sharpe offered a counterargument. Divide the world into “active” investors and “passive” investors, Sharpe said. A passive investor is defined as anyone sensible enough to realize you can’t beat the market. The passive investor puts all his money into a market portfolio of every stock in existence (roughly, an “index fund”).

An active investor is anyone who suffers from the delusion that he
can
beat the market. The active investor puts his money into anything
except
a market portfolio.

By Sharpe’s terminology, an active investor need not trade “actively.” A retired teacher who has two shares of AT&T in the bottom of her dresser drawer counts as an active investor. She is operating on the assumption that AT&T is a better stock to own than a total market index fund. Active investors include anyone who tries to pick “good” stocks and shun “bad” ones, or who hires someone else to do that by putting money in an actively managed mutual fund or investment partnership.

Who does better, Sharpe asked: the active investors or the passive investors? Collectively, the world’s investors own 100 percent of all the world’s stock. (Nothing is owned by extraterrestrials!) That means that the average return of all the world’s investors—before you factor in management expenses, brokerage fees, and taxes—has to be identical to the average return of the stock market as a whole. It can’t be otherwise.

Even more clearly, the average return of just the passive investors is equal to the average stock market return. That’s because these investors keep their money in index funds or portfolios that match the return of the whole market.

Subtract the return of the passive investors from the total. This leaves the return of the active investors. Since the passive investors have exactly the same return as the whole, it follows that the active investors, as a group, must also have the same average return as the whole market. This leads to a surprising conclusion. Collectively, active investors must do no better or worse (before fees and taxes) than the passive investors.

Some active investors do better than others, as we all know. Every active investor
hopes
to do better than the others. One thing’s for sure. Everyone can’t do “better than average.”

Active investing is therefore a zero-sum game. The only way for one active investor to do better than average is for another active investor to do worse than average. You can’t squirm out of this conclusion by imagining that the active investors’ profits come at the expense of those wimpy passive investors who settle for average return. The average return of the passive investors is exactly the same as that of the active investors, for the reason just outlined.

Now factor in expenses. The passive investors have little or no brokerage fees, management fees, or capital gains taxes (they rarely have to sell). The expenses of the active traders vary. We’re using that term for everyone from day traders and hedge fund partners to people who buy and hold a few shares of stock. For the most part, active investors will be paying a percent or two in fees and more in commissions and taxes. (Hedge fund investors pay much more in fees when the fund does well.) This is something like 2 percent of capital, per year, and must be deducted from the return.

Two percent is no trifle. In the twentieth century, the average stock market return was something like 5 percent more than the risk-free rate. Yet an active investor has to earn about two percentage points more than average just to keep up with the passive investors.

Do some active investors do that? Absolutely. They’re the smart or lucky few who fall at the upper end of the spectrum of returns. The majority of active investors do not achieve that break-even point. Most people who think they can beat the market do worse than the market. This is an irrefutable conclusion, Sharpe said, and it is not based on fancy economic theorizing. It follows from the laws of arithmetic.