Mission Improbable: When Ken French says you cannot beat the market, he means YOU (probably)


It’s no surprise that Ken French1 has strong opinions about market efficiency. Still, who else could tell a room full of portfolio managers that it’s nearly impossible to find managers who can beat the market and get a positive response from his audience?

As the Carl E. and Catherine M. Heidt Professor of Finance at the Tuck School of Business at Dartmouth College, French is well known for his groundbreaking research (with co-author Eugene Fama) on the value effect and the three-factor model. But French isn’t an ivory-toweracademic. He splits his time between New Hampshire and Los Angeles, where he serves as director of investment strategy at Dimensional Fund Advisors, which has US$78 billion in assets under management.2

French is tired of talking about the tech bubble, but he was happy to discuss equilibrium, asset-class investing, and momentum — an anomaly that he views as “one of the biggest embarrassments for market efficiency.”

How efficient are markets?

I’m not sure there’s a standard metric, so I’ll say it’s about 87.32 percent.

Can you go into absolute versus relative prices?

First, I think it’s crazy to argue that all prices are exactly right all the time. Market efficiency is just a model, and like every other interesting model, it must be false. We’re just talking about how false. I mean, how bad is the market at setting prices? You can answer in a variety of ways. Let me personalize it. Do I think it’s a good use of my time to try to identify mistakes in the market? Absolutely not.

First, it’s not clear I have any skill at it. Second, even if I have some skill, I’m sure there are lots of people who are better than I am. Should all those smarter people try to beat the market? I think the answer for almost all of them is also no. Suppose I go to Las Vegas to play poker. If there are 10 people at the table, and I think I’m better than half of them, should I expect to make money? Probably not. In fact, I’m pretty sure the best player at the table is very happy I sat down. To him, I’m just one more sheep to fleece. And he’ll be even happier if I’ve fooled myself into thinking I will make money, because that will make it even easier for him to take mine. So, yes, there probably are mistakes out there, but I don’t think they are big enough or obvious enough to make it worth my while to look for them. And just about everyone else would also be better off not wasting his or her time trying to find alpha. Almost all of us should act as if prices are right.

If markets are as efficient as you believe, can beta managers
expect to have the last laugh?

In equilibrium, there is room for some people to have positive risk-adjusted expected returns. But even if a few people can beat the market — I mean in an expectational sense — the evidence is pretty clear that most investors can’t. And it seems to be at least as hard to identify the rare investment managers who can.

Think about what happens when investors discover a manager who has a strong track record. Whatever skill she has usually gets diluted as the investors chase returns and assets under management grow dramatically. And she often raises her fee. This makes sense. Standard economic principles say the scarce resource should capture the rents. Investors’ perfectly liquid capital flows to wherever the opportunities Of course, you might argue that you know how to pick winning managers, but that seems even harder than picking winning stocks, particularly since, in aggregate, active managers should — and do — lose to passive managers.are, driving down the expected returns for superior managers.

What is your opinion of the current emphasis on alphadriven,
absolute-return strategies?

I hate to keep coming back to this equilibrium view of markets, but it’s hard to believe there are enough opportunities out there to support either all the resources active managers spend trying to find undervalued stocks or all the payments investors make hiring managers to find those undervalued stocks. It doesn’t make any sense to me.

We know some people are going to get lucky. When you have more funds than you have stocks on the NYSE, some of them have to work out. And even on an ex ante basis, surely there are some people out there who have a positive risk-adjusted expected return. But it’s amazingly hard to identify them ex ante.

It takes about 35 years of returns to say with any statistical confidence that stocks have a higher expected return than the risk-free rate. Think about a hedge fund that has equity-like volatility. If the manager’s alpha was as large as the market risk premium — which would be huge — it would also take about 35 years to be confident the manager has any value added — and that’s before his fees of “2 and 20.” Even if that phenomenal manager is out there, is he likely to stick around long enough for us to be able to figure out he wasn’t just lucky?
Most of the time when you see people chasing returns, it’s all in vain. The numbers they’re chasing provide almost no information at all. Think about all those investors who leave a hedge fund after six months because their returns have been bad. That’s the dumbest thing in the world. Six months of returns tells you essentially nothing about the manager’s skill or future returns.

All the investors have learned is that their fees are lower than they used to be. Because the investors are now below their high water mark, their expected after-fee return is higher than it was when they started. If they liked the fund before, they should love it now. More constructively, if potential hedge fund investors think they would leave after six months of horrible returns, perhaps they should not invest in the first place.

Can you discuss Dimensional’s academic approach to money management?

We frame things from this equilibrium perspective. We think we can give our clients better portfolios if we don’t waste our time and energy competing with others to find the next Microsoft. Instead of distracting ourselves by trying to pick undervalued stocks, we focus on designing and delivering asset-class portfolios as efficiently and effectively as possible.

For example, our typical small-cap portfolio might hold 3,000 stocks. When we have money to invest, we pay attention to diversification, style drift, and market conditions, including the current liquidity of the eligible stocks. If there is lots of supply of some stocks and little of others, we will buy more of those that are easier to get.

Suppose instead that we were an active manager who just spent the last three weeks doing an incredibly involved analysis to figure out which stocks are undervalued and which are overvalued. Now that we know — or rather, now that we think we know — we have to buy the ones that are priced too low, and we have to sell the ones that are too high. We have lost the flexibility to take advantage of temporary imbalances between the short-run demand and supply think we do it better than anyone else, but there are plenty of other people who are doing it well.

Why aren’t more people doing this?

In fact, they are. More and more people are realizing that active management is not the best way to invest. And, as you might expect, as investors shift toward passive or asset-class investing, managers step up who are willing to provide the investment expertise. It’s not just Dimensional. We like to think we do it better than anyone else, but there are plenty of other people who are doing it well.

Are the terms active and passive still meaningful?

We struggle to define ourselves on that spectrum. We’re certainly not active. But we also don’t fit the standard definition of passive. When people say passive, they often mean index. We are hardly index managers. In fact, we think many indices have serious flaws when used as investment strategies. Indices are intended to be tools to measure performance, not rules for managing portfolios.

So what do you call Dimensional’s approach?

I often use asset-class investing, but that misses it when you’re buying an overall market portfolio. For example, we have several portfolios that buy most of the stocks in a broad universe, but we add size or style tilts by overweighting some stocks and underweighting others. Asset-class investing is definitely not a good description for these portfolios.

How does momentum relate to efficiency?

Momentum, from my perspective, is one of the big embarrassments for the efficient markets hypothesis. It may be that somebody will come up with a good story reconciling market efficiency and momentum, but I certainly haven’t heard it yet. I know lots of people have tried, and I’m sure I’m offending somebody who is convinced he or she has a rational model that reconciles the two, but I have not seen one that holds together.

Although momentum is a huge challenge, it’s hard for active managers to exploit it because it’s a fairly high turnover strategy, and the costs of trying to exploit it are high. But if they’re trading anyway, it’s not clear to me why everybody isn’t saying, “Let me incorporate momentum into the trading that I’m going to be doing anyway.”

There’s a great paper by Mark Carhart that shows how hard it is to actively trade on momentum. It’s in the 1997 Journal of Finance. Mark looks at the performance of mutual fund managers. He finds that there is some persistence in manager performance, but almost all of it is an artifact of momentum.

Because they don’t turn their portfolio over completely from one year to the next, the most successful managers from last year continue to hold some of the stocks that drove their portfolios to the top. Momentum says these stocks are likely to continue to do well this year, which gives last year’s best managers an edge over their competitors this year. And that’s what Carhart finds. Because of momentum, last year’s best performers start this year with a bit of a head start, and last year’s worst performers start with a handicap.

Most, if not all, of this persistence from momentum is probably inadvertent. When Carhart looks at managers who try to exploit momentum by actively trading on it, it looks like they aren’t able to capture the momentum returns. And that makes sense. Transaction costs probably consume the potential gain.

Why should the best managers have the highest cost of capital?

Good question. I don’t think they should. When I invest in a fund, what I perceive as my expected return, the manager perceives as his cost of capital. If he’s a great manager, I might be saying to myself, “This is awesome. I’m going to get a high expected return!”

Now think of it from his perspective. Do we really think he is saying, “Darn, I wish I weren’t a great manager; my cost of capital is too high.” Of course not. But if my expected return really were high, his cost of capital would also have to be high. This goes back to the argument I was making earlier. If he really is great, his strong returns will allow him to expand, spread his ability across more assets, and raise his fee. He captures the rent rather than the investor. That makes sense. So why doesn’t it happen? I believe it does. I mean, we see hedge funds now that are charging 3 percent of money under management and 30 percent of profits. That’s perfectly consistent with my story. Of course, I think investors are only willing to pay those fees because they are overreacting to prior performance, but that’s a different issue.

Am I correct in guessing that most money managers are less than thrilled with your pessimistic view of their market-beating abilities?

I haven’t presented this argument to many practitioner groups, but every time I have gotten a very positive response. My comments on equilibrium markets seem to be flexible enough to accommodate whatever views the participants have.

Every active manager that I’ve talked to afterward says something like, “Wow, those other people really are dumb, aren’t they?” Of course, maybe I’m suffering from selection bias. Perhaps the people who hated the presentation don’t come up afterward.

In my talk now, I point out that everyone in the audience can’t be above average. Some people in the room — in fact, probably most of the people in the room — are wasting their clients’ money trying to beat the market or trying to pick the active manager who can. I also remind them that overconfidence is almost certainly the most important bias in behavioral finance. But most people still think I’m not talking about them.

As both academic and practitioner, Kenneth French is an expert on the behavior of security prices and investment strategies. French is the Carl E. and Catherine M. Heidt Professor of Finance at the Tuck School of Business at Dartmouth College in Hanover, N.H. He also serves as director of investment strategy at Los Angeles-based Dimensional Fund Advisors, overseeing their research group.

He and co-author Eugene Fama are well known for their research into the value effect and the three-factor model. French’s most recent research centers on tests of asset pricing and the trade-off between risk and return in financial markets.

Prior to joining Dartmouth, he was the NTU Professor of Finance at MIT’s Sloan School of Management and managing director of the International Center for Finance at Yale University. French received his PhD in finance from the University of Rochester in 1983, having previously earned an MS and an MBA from the University of Rochester and a BS from Lehigh University.

1) As a consultant to Dimensional, Professor Kenneth R. French is director of investment strategy for Dimensional Fund Advisors Inc. He also oversees the research group. 2) Assets under management as of June 30, 2005.