What’s it all about, Alpha?

Who does what with which to whom, or, how efficient is the market?

Few pursuits are as evenly divided between theory and practice as finance. The investment industry sells not only mutual funds, hedge funds and variable annuities, but ideas.

Peter Bernstein has long been the epitome of the thinking practitioner. From 1951 to 1973 he was chief executive of Bernstein-Macaulay, where he personally managed billions in individual and institutional accounts. In 1973 he founded Peter L. Bernstein Inc., a consulting firm, and the following year he became the first editor of the Journal of Portfolio Management, a scholarly journal published by Institutional Investor.

In recent years he has brought high-end finance to broader audiences with his bestselling books: Capital Ideas: The Improbable Origins of Modern Wall Street(1992); Against the Gods: The Remarkable Story of Risk(1996); and The Power of Gold: The History of An Obsession (2000). Bernstein’s books contain the clearest plain-English explanation of the ideas that underpin the investment world -- from Black-Scholes option pricing models to the capital asset pricing model -- and they are a vital synthesis of the disparate fields of modern finance.

Bernstein’s always thought-provoking. Last year he startled the worlds of finance theory and portfolio management practice by arguing, in publications and speeches, that institutional investors should discard “policy portfolios,” which set down in stone their long-term asset allocations, and take a more opportunistic, market-timing approach to investments. So important is Bernstein, and so radical was his suggestion, that perturbed Vanguard founder John Bogle dedicated a speech to attacking the proposition.

Bernstein and others have since qualified, and expounded, his original remarks, but they continue to stir up debate about the extent to which the practice of money management has become too rigid. “The single most profound change in investment thinking since [Harry] Markowitz’s work on portfolio theory is upon us,” wrote Barclays Global Investors senior investment strategist Kevin Kneafsey in an August 2003 research piece. “Not surprisingly, Peter Bernstein is in the middle of it.”

The following is adapted from a recent issue of Bernstein’s “Economics & Portfolio Strategy,” his semimonthly letter to institutional investors.

We all know the market is hard to beat. But what does “beat the market” mean, anyway? We talk a lot about alpha-this and alpha-that, we port alpha and extend alpha and engage in other delightful activities with alpha, but how difficult is it to find alpha in the first place? And as alpha is an inherently volatile concept, how accurate are the measurements on which we lean to prove its existence? How palpable is alpha? How active can active risk be before we call a halt to the whole effort?

In what follows, I suggest answers to these questions. But the answers you will read are not the answers you might expect. There is a lot more to this matter than just numbers and equations. The issues go to the very heart of the money management process.

The most familiar answers to these questions conclude that the market is generally efficient -- very hard to beat. The long track record of mutual fund performance is the most powerful demonstration of the difficulty of beating the market. Study after study has arrived at this sad conclusion, beginning with Michael Jensen’s seminal analysis, “The Performance of Mutual Funds in the Period 1945'64,” which appeared in December 1965. Steve Ross of the Massachusetts Institute of Technology recently summed up the whole matter in these words: “There should be no debate, then, over whether the efficiency glass is half full or half empty; it is simply quite full.”

The story is too familiar to require any elaboration. No one can argue with what these studies show. But it is fair to ask whether these kinds of studies are sufficient to prove the point they purport to prove. The whole business is a lot more complicated than we have been led to believe.

STEIN’S CASE

A recent National Bureau of Economic Research working paper, “Why Are Most Funds Open-End? Competition and the Limits of Arbitrage,” by Harvard University economics professor Jeremy Stein, is must-reading. (You can obtain it at www.nber.org by ordering working paper No. 10259 and charging your credit card with $10.)

The basic message of this fascinating study appears in the very first paragraph:

“This paper is motivated by two stylized facts. First, the vast majority of professionally managed investment vehicles (i.e., mutual funds, hedge funds) are structured on an open-end, as opposed to closed-end, basis, making it possible for their clients to liquidate shares on demand. Second, both theory and evidence suggest that the open-end form imposes serious constraints on would-be arbitrageurs. In particular, being open-end exposes arbitrageurs to the risk of large withdrawals if they perform poorly in the short run. This risk in turn makes it dangerous for them to put on trades that are attractive in a long-run sense but where the convergence to fundamentals is unlikely to be either smooth or rapid. . . . To take one leading example, open-end funds are unlikely to want to bet heavily against something like the Internet bubble of the late 1990s.” (Stein provides a footnote to soften his use of hedge funds as an example.)

In light of this argument, Stein goes on to ask why the open-end format is so dominant. The open-end format extends far beyond the mutual fund concept to include most investment management arrangements: Even when clients cannot withdraw on demand, they can usually quit a manager at the end of a yearly contract. Stein then offers a possible explanation for the popularity of the open-end format: Clients or shareholders are concerned that the managers they select will turn out to be incompetent or dishonest in some fashion. They do not want to commit without an exit strategy. When we look at the open-end format from this standpoint, it represents a socially efficient outcome.

But Stein has a different take on what is going on. He argues, with a nice turn of phrase, that “the end result may be a degree of open-ending that is socially excessive [his italics].” He sets forth the hypothesis that “the gains from being able to undertake longer-horizon trades in the closed-end form outweigh the potential losses that come from being unable to control wayward managers.” This time the italics are ours.

LOOKING DEEPER

There is little doubt that the closed-end form does give managers greater leeway in making longer-term or less-liquid bets -- the kind Jack Treynor, a former editor of Financial Analysts Journal and a leading financial theorist, characterized as “slow ideas.” Indeed, although Stein -- erroneously, in my view -- includes hedge funds in the open-end class, most managers who function with lockup restrictions (which prevent clients from withdrawing on demand) justify the lockup because it provides the managers with more leeway in the kinds of positions they can take on. A manager with a lockup arrangement, for example, would be much more willing to take the risk of selling or shorting stocks in the midst of a roaring bull market like the 1990s than if clients could withdraw funds at the end of the contract year.

Stein poses a question worth considering. If the closed-end format can give managers a better opportunity to generate alpha, why is the investment management fraternity unable to persuade more people to sign up and let them manage money that way? Closed-end funds control a tiny fraction of the trillions invested in mutual funds, and lockups in the institutional investment management arena are strictly limited to hedge fund, private equity, venture capital and other types of relatively illiquid fields of investment.

Stein’s answer to his question is entirely different from what we might expect. It is not the clients who are blocking this outcome, but the managers! The appetite for inflow is the villain, not the risk of outflow.

Consider for a moment an imaginary world where all investment management firms operate in the closed-end framework. Nobody has ever heard of anything else. Now suppose one of those organizations has a fantastic track record -- year after year it beats the market. On the tenth anniversary of their first year of statistically significant alpha, the senior executives of this happy group sit down to lunch together to celebrate a decade of extraordinary success. The head man pipes up: “Hey, guys, we are passing up a lot of goodies for ourselves. That’s not all. We are also depriving the world out there of our talents. Why should we restrict ourselves to just this one little pot of money? Let’s go open-end. We will gather in piles of assets, swelling our fee revenue, and make countless clients rich in the process.”

The proposal would be so irresistible, it would be set before luncheon tables in many closed-end outfits with exemplary track records. Once one took the plunge, most of the superior managers would be unable to resist the temptation to follow suit. Open-end would be a mark of talent, and the clients, like lemmings, would rush into the open-end funds. Any organization still clinging to the closed-end format would be identified as a denizen of the bottom quintile. So even these backsliders would hasten to open-end in order to avoid the stigma.

Welcome to the world of today.

MARKET IMPLICATIONS

All we know about market efficiency is what we see around us: a world in which most of the dominant players -- the managers of institutional funds, occasionally supplemented by the more active members of the 401(k) fraternity -- are operating on a short-term tether. In this world, where so many investors are wrapped into a short time horizon, beating the market is extremely difficult. Alphas are scarce indeed. The flood of publicly available information every minute of the day and night provokes intense competition among managers who do not want to wander too far away from consensus views and must continually adjust their portfolios to this plethora of earnings forecasts, earnings adjustments, earnings announcements, merger rumors, analyst ratings changes and goodness only knows what else that bombards the Bloomberg, CNBC, Dow Jones and Reuters screens. Nor, if small-cap is hot or technology is cold, do these managers want to have to face clients with an underweight in small-cap or a fistful of tech stocks.

These patterns influence clients as well as managers. Chief investment officers at pension funds, charitable foundations and educational endowments have to face investment committees. They, too, are reluctant to see their portfolios turn into outliers, even if briefly. If Old Siwash is down 10 percent in a year when the endowment average is up 10 percent, well . . .

Cuddling up to the consensus is the better part of valor. As Mark Kritzman and George Chow pointed out in Financial Analysts Journal in the spring of 1995, even if a manager seeking a high alpha is blessed with perfect knowledge of expected returns and the distributions around them, “there is a substantial likelihood that the actual returns will be significantly different from the expected returns. Thus, even though you may be right on average over the long run, you will almost certainly be wrong (due to high standard deviations) and alone (due to an uncommon allocation) over some shorter measurement periods. Most of us, owing either to career or psychological considerations, are unwilling to risk the chance of such an unpleasant outcome. The regret might be unbearable.” Or take my own contribution to these matters, “Where, Oh Where, Are the .400 Hitters of Yesteryear?” (Financial Analysts Journal, November/December 1998), in which I concluded that “increasing fear over time of being ‘wrong and alone’ is a convincing explanation of the narrowing spread between top managers and the S&P 500.”

Not much has changed over the past eight years. In an article in the latest issue of the Financial Analysts Journal about the source of the higher returns of value stocks, Louis Chan and Josef Lakonishok suggest that the value premium is not a reward for taking higher risk, as Eugene Fama and others have contended. Rather, they write that a more convincing explanation “rests on characteristics of investor behavior and on the agency costs of delegated investment management,” which is what Stein and Kritzman are talking about. Contrary investors may win out in the long run, but in the short run they often find themselves wrong and alone. Chan and Lakonishok’s analysis is well worth reading in full because the authors provide a rich display of evidence to reveal that Kritzman’s aphorism about fear of being wrong and alone is alive and well.

But what makes managers so timid at the prospect of being wrong and alone? Because they are operating in the open-end format. It is too easy for assets under management to melt away as a result of one or two years of poor performance, even if over a longer period a nice fat alpha is being baked into the cake of security selection.

In case you think this might be an exaggeration, consider Bill Miller’s miraculous management of the Legg Mason Value Trust. According to Morningstar data, Miller beat his benchmark and was in the top quartile every year but one from 1992 to 1999. Miller sports a Sharpe ratio of .08; his peer group’s Sharpe ratio is .48. His average turnover rate since 1992 is 22 percent; his peer group’s turnover rate averages 76 percent, with the low year at 65 percent! We do not lightly use the word “miraculous” here.

From a starting value of $842 million at the end of 1992, Miller scored total compound returns of 28 percent a year over the seven years to 1999. As a closed-end fund, the value at the end of 1999 would have been $4.8 billion. But the actual value at the end of 1999 was $12.5 billion, for a total of return plus cash inflow of 47 percent a year over those seven years. Nearly $8 billion of new money had arrived at Miller’s doorstep by the end of 1999 -- ten times his starting value. During the down markets from 1999 to 2002, Miller continued to beat the index and remained in the top quartile. As a closed-end fund, the assets would have fallen from $12.5 billion to $8.5 billion. In fact, they fell to $7.3 billion. Despite the continued outperformance, more than a billion dollars exited by way of the open end.

Miller lost 42 percent of his assets between 1999 and 2002. While this did not appear to cramp his style, and while he still had more than $7 billion under management at the nadir, think of how much better he might have done if his itchy shareholders had left him alone -- or had not had the option of pulling out any time they wished to do so. Miller’s outperformance in those three down years was much narrower than it has been in the up years. And without all that cash inundating him in the good years, his upside returns might also have been even better.

Stein cites managers of corporations as a different example of a closed-end format. With the advantage of a closed-end environment, managers of overpriced dot-com companies had no difficulty betting against the Internet bubble in 1999 because the market gobbled up their issues of new shares of stock (the equivalent of going short) as though there would never be a day of reckoning. If these executives were wrong -- if the market continued to rise -- they could just sit on the cash with “only the subtle opportunity cost of having not timed the market even better,” Stein says. Existing shareholders, whose interests had been diluted, had to sit and take it -- there was no easy way for them to raid the company’s assets and take the cash for themselves.

Meanwhile, a hedge fund manager executing a short sale of the Nasdaq index in early 1999 would have had a most difficult year with the limited partners. If you do not believe that, ask the committed value managers how much business they lost during 1999. Or imagine how Bill Miller felt as his shareholders pulled his fund’s assets away from him.

Clincher: The most famously successful investor of them all, Warren Buffett, sits happily behind the impregnable wall of a closed-end format. You can trade the shares of Berkshire Hathaway all you want, but you cannot invade the company’s assets.

Here is just one more example. If you are managing a portfolio with a 20 percent or so annual turnover rate, you are holding your investments for an average of five years. If your turnover rate is 75 percent, your holding period is only 16 months. Which portfolio management organization worries more about the skill of the traders? On which portfolio does an eighth or a quarter on a trade make more difference? The answer is obvious. The increasing focus on trading costs in recent years is not a symptom of growing investor sophistication but rather growing investor focus on the short term -- a direct consequence of the open-end format.

THE FULL MEANING OF THE STORY

If fear of being wrong and alone under the open-end convention gives most active investment managers little choice but to herd and focus on only short-term -- and short-lived -- pricing aberrations in the market, we should not be surprised when the empirical evidence confirms that alphas are few and far between, thin where they may exist and with high volatility around their means. Steve Ross unquestionably has the best of that argument.

Ross’s argument is only part of the story. The more that managers converge on the near term, the greater the odds of opportunities lurking in the longer term. This view is Warren Buffett’s whole philosophy. This view is Jack Bogle’s refrain. But except for a few brave souls, the gold in them thar hills is not being mined. If this is the case, the market is not so efficient after all. Mispricings do exist and do persist.

THE PRACTICAL SIDE OF IT ALL

The trend toward so-called alternative investments in recent years -- private equity, timber, real estate and venture capital -- suggests that institutional investors are willing to accept closed-end and longer-horizon types of risk-taking with lockups. But the lockup exists because of liquidity constraints in the nature of the assets, not because it permits the manager to take a longer-term horizon in making investment choices.

The hedge fund fad is more interesting, because hedge funds invest primarily in marketable securities. The peculiar risks of going short do require some protection from unscheduled client withdrawals, but in any case, the popularity of hedge funds indicates that perhaps investors would consider being weaned, at least in part, from the viciousness of the open-end format. Now closed-end, rather than open-end, is a mark of talent.

When I feel like fantasizing, I try to draw an investment management contract with a five-year, instead of a one-year, horizon. What would it look like? The goal would be to give the manager maximum leeway and still protect the investor from incompetence or malfeasance. Although we all know that track records do not predict future performance, they should indicate something about degrees of competence. Even the open-end format has not protected clients from malfeasance when the managers were determined to steal. Perhaps there could be an option to quit at the end of three years, with the payment of a premium in advance for that option. Or there could be quit options maturing each year, with premiums declining with the life of the option. Welcome to the world of tomorrow.

Some institutional investors and mutual fund companies are passing up an opportunity to earn palpable alphas. All they need is the right framework. This is a fantasy worth bringing down to practical applications.

Related