The following is an excerpt from Snap Judgment: When to Trust Your Instincts, When to Ignore Them, and How to Avoid Making Big Mistakes with Your Money
How do we pick a particular mutual fund to invest in? The answer is very simple: Its performance last year, last quarter, or even last month. Of course, advertising helps, too, but "returns-chasing behavior" as it is technically known, pretty much captures our actions when it comes to mutual fund choices. The #1 fund in the country last year in terms of performance will be near or at the top of mutual funds attracting the most new investment dollars.
And performance here means absolute raw performance, not adjusted for risks, fees, taxes, or compared to a benchmark such as the S&P 500. If it makes more money than we paid for it, we see it as a winner, even if the S&P did much better. Alternatively, if a mutual fund is below what we paid for it, but every other investment is doing significantly worse, we aren't impressed.
But after we make our performance-related investment choice, our one moment of action, we as investors appear to go to sleep. Our inbuilt inertia and apparent passivity when it comes to investing take over. We open an account and then sit on our hands. The fund may stumble, but we don't seem to react. We leave our investment where it is. Most investors don't reallocate from low-performing to high-performing funds, even within their portfolio. Returns-chasing behavior, therefore, only goes in one direction: We give new assets to high-performing funds, but leave old assets with the underper-forming managers. An unattractive analogy is to roach motels: We check in, but don't check out.
This returns-chasing behavior is seemingly hardwired in humans, from the most naive individual investors to the most sophisticated institutional investors. Everyone does it. It is why hedge funds, which may have had their highest returns in the 1980s and 1990s (although no one knows for sure), became such a popular investment a decade later, even though their best days were probably long behind them as a strategy. Financial planners, in surveys, say they don't chase returns, but an analysis of their investment choices by Daniel Bergstresser of Harvard Business School found plenty of evidence of it. And we know that returns chasing is what drives individual investors in their choice of mutual funds.
Returns chasing isn't automatically a bad strategy. In most circumstances, learning about the track record and history of what you are investing in should be valuable information. Even so, as every investment prospectus warns us, in a rare moment of clarity, "past returns are no guarantee of future returns." This is true, but do past returns tell us anything about future returns when it comes to mutual funds? They do, but not in ways we might assume.
Performance and Persistence
Alpha is the catchphrase bandied about by hedge fund managers and other finance types as shorthand for their market-beating performance (in another words, their skill). Sometimes called "Jensen's alpha," it is a technical measure of the "excess return" on an investment compared to what the market would give you, adjusted for risk. For hedge funds, alpha is their mantra, their obsession, maybe even their mojo. It's everywhere in their speeches and publications and the names of their conferences. They call each other "alpha" males. Mutual funds managers keep a bit quieter about alpha, but nonetheless mention it, too.
The word alpha, as used in a financial context, was coined in the 1960s by Michael Jensen, then a graduate student at the University of Chicago. He was interested in mutual fund performance—was it based on a manager's skill from picking stocks or just the market going up or down on its own? He had to figure out a way to measure this, which is when he invented the concept of "alpha." He defined this as "a risk-adjusted measure of portfolio performance that estimates how much a manager's forecasting ability contributes to the fund's returns." Alpha was indeed an analytical breakthrough—the word and concept stuck and is now part of financial culture.
But that is only part one of Jensen's story. There is a second part that is completely forgotten, at least by hedge fund guys. And that is what happened when Jensen went looking for evidence of alpha in his study. He couldn't find any, at least over an extended period of time. The mutual funds he studied were not able to beat the market for any long time period. Examining the performance of mutual funds for the period 1945 to 1964, Jensen concludes: "The evidence on mutual fund performance indicates that these 115 mutual funds were not on average able to predict security prices well enough to outperform a buy-the-market-and-hold policy. Also there is very little evidence that any individual fund was able to do significantly better than that which we expected from mere random chance.... On average the funds apparently were not quite successful enough in their trading activities to recoup even their brokerage expenses."
The ability to beat the market for a specified period is known, at least by academics, as persistence. Though Jensen didn't use the word persistence when writing about alpha, it is implicit in his measure and another of his discoveries. As he modestly wrote to me when I asked him about this research: "I do not recall the idea of persistence being in existence before my study, but the study was a long time ago."
Everyone still talks about alpha, but no one talks about persistence. An investment manager might get lucky once or twice, but whether he or she can persist in generating excess returns over the market for more than a few months is really the question—and the question you should ask when selecting a fund that claims to be able to beat the market. If you ever meet a fund manager who brags about his alpha, you can ask it of him directly: "What about your persistence. How long can you keep it up?" (There might be a less aggressive way of asking this, but the point remains).
If mutual funds can't persist in beating the market, then returns chasing makes no sense as a strategy. Doing better than the market in the past is no guide to the future. The question of persistence in mutual fund performance is an unsettled one. It has been studied extensively since Jensen's day. Most academics are intrinsically skeptical that a mutual fund would be able to beat the market for very long. They use seemingly noncommittal but in fact savage language, claiming that any real evidence of persistence of market outperformance by mutual fund managers is "elusive."
Wall Street has a different view, of course. There is always the example of Peter Lynch, who ran Fidelity's Magellan Fund and had returns almost double the S&P for many years. Part of the discussion of persistence has to do with the time period you are talking about. There are managers such as Bill Miller of Legg Mason who had great runs for a while. Miller had a 15-year winning streak, only to be followed by a 10-year losing streak. This culminated in his disastrous 2008 presentation at an investment conference in New York. The topic: "The Credit Cycle—What's Next?" Miller recommended financial stocks as a great buy. He singled out Bear Stearns, in which he was one of the biggest investors, having committed $200 million of his fund's money. An audience member raised his hand and asked Miller a question: Was he aware that Bear Stearns was in crisis, in fact was cratering that very morning? Miller seemed shaken. He quickly left the conference. His $200-million dollar investment was soon worth only $15 million dollars.
Miller's downfall is exceptional, but it is also clear that many mutual fund managers have exceptional skills at picking stocks, as Miller himself did for a while. Mutual funds have some great years, extraordinary years, that can't be explained by dumb luck. So, what's really going on here in terms of persistence and performance?
This is a superb entry!
Most mutual funds have past performances that go back years, but they are rarely managed by the same fund managers. That complicates this in terms of figuring out a fund manager's ability to generate alpha (and eventually, persistence), much less whether they can persist with similar performances into the future.
Plus, yes, a lot of celebrity fund managers have had a hard time. Miller is a perfect example. So is Ken Heebner. Actually Peter Lynch had a hard time towards the end of his career as well, but he also complained about having his hands tied by regulations.
I remember the issue of persistence being brought up over at diehards, but basically, by the time you have enough data to figure out persistence, they're old enough to be retiring. :D
I fully agree with the point on persistence. Unfortunately, I don't think it's something that can be implemented easily.
That said, Warren Buffett is always an interesting case study. Here's a link to a white paper on whether his career's returns are based on luck or skill. (The conclusion was skill.)
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=806246
Again, great entry. Thanks for sharing!
Posted by: Eugene Krabs | June 25, 2009 at 11:19 AM
Article said: "How do we pick a particular mutual fund to invest in? The answer is very simple: Its performance last year, last quarter, or even last month. Of course, advertising helps, too, but "returns-chasing behavior" as it is technically known, pretty much captures our actions when it comes to mutual fund choices. The #1 fund in the country last year in terms of performance will be near or at the top of mutual funds attracting the most new investment dollars."
This is probably one of the dumbest statements I've ever seen written. What a mutual fund did in the past has absolutely nothing to do with what it will do in the future. That goes for any investment or anything in life for that matter. Whoever wrote this needs to do us all a favor and find a new career. Finance isn't this person's specialty.
Posted by: Jim | July 08, 2009 at 01:54 AM