Random Talks with Eugene Fama

An Interview with One of the Investment Community’s Brightest Thinkers

It was only fitting that Eugene Fama speak at our User Conference. Roger Ibbotson considers Fama his mentor. After all, he was on Roger’s PhD thesis committee at the University of Chicago. It’s easy to see why Roger looked up to Fama.

Gene Fama helped write the book on modern finance. He was a tenured professor at the University of Chicago before he was 30. He began teaching modern portfolio theory in the 60s at the University of Chicago before modern finance was an established field.

His PhD thesis, “The Behavior of Stock Market Prices,” took up an entire issue of the University’s esteemed Journal of Business. A simplified version of this paper titled “Random Walks in Stock Market Prices” would eventually be published in Institutional Investor and its significance rippled throughout the investment industry far beyond the halls of academia.

Widely known for his work on efficient markets theory, Fama coined the term with his 1970 paper on the topic. “Efficient Capital Markets” argues that on average, it is nearly impossible for an individual to consistently beat the stock market as a whole because of the broad availability of public information. A common analogy of efficient markets theory is that an investor who throws darts at a newspaper’s stock listings has as much a chance at beating the market as any professional investor. It states that stock price behavior is random and not predictable by so-called market forecasters.

From the moment that Fama stepped into the academic scene, his work raised the eyebrows of finance professors and investment professionals alike and continues to do so thirty years later.

Currently, Fama is Professor of Finance at the Graduate School of Business at the University of Chicago, where he is also chairman of its Center for Research in Security Prices (CRSP). In addition to his academic post, Fama is Director of Research at Dimensional Fund Advisors, who have based several of their investment products on his findings.

What follows is part one of our interview with Gene Fama.

Interface: How do you pick a topic for a research paper or does it pick you?
Gene Fama: I don’t know. I never really thought about it. It’s just something that you do. It happens. Talking to people about things and something comes up. What looks like a small project turns into a big project. What looks like a big project turns into a small project. So there’s not a lot of planning involved, I don’t think. At least not in my case, there hasn’t been.

How do you go about collaborating with other people and how does that work? Do they start off as casual conversations with somebody?
More or less. Usually it starts with people that are around here [University of Chicago]. French [Ken French, Professor at Yale University] isn’t here anymore. Sometimes you do only one paper but if it works well, you might do more. So it depends on how the work experience goes, whether you keep going or not.

Do you prefer to work on your own as opposed to working with others or do they both have their benefits?
Yeah. My work is probably about half and half. Recently it’s been done mostly with French.

What provokes your insights? Where do your ideas come from?
I don’t have the slightest clue. You just keep working hard at it and hope that they pop out at you. [Laughs]

I remember at our User Conference you said that you never saw a number you didn’t like.
Well, I was the original computer user among economists here with work on my PhD thesis. It was the first computer that had arrived, the [IBM] 709. I do think I have a skill for looking at, for interpreting data. It’s not something easily taught. Some people are good at theoretical things; some people are good at looking at data.

How do you compare those days with your computer work now? I see you have a nice PC on your desk.
Yeah. This is a thousand times more powerful. That machine [IBM 709] probably cost in those days three or four million dollars, which now would be thirty million dollars. [Today] you can get something a thousand times more powerful for a couple thousand.

That brings me to another topic. In those days, even with Markowitz’s work [mean-variance optimization], it was harder to implement those ideas and harder to see the data. Now that computer technology is so much more readily available, enabling users to crunch the numbers, do you think we’ll see more research? Do you think we’ll see better ideas come out of that?
We’ll definitely see more research. When I started, the journals of finance were just the Journal of Finance. Now there are four or five journals. There are many more people and many more places doing research. And I think it’s because of data availability and the development of the field. People are trained in it; so they’re researching it.

Finance is really—I bill it as the most successful area of economics. But back in the early 60s when I started, it wasn’t really a, quote unquote, scientific area. It was just starting to be. The development of the theory, empirical work and computers all joined together to make an explosion of work in this area. We’re advantaged relative to other areas of economics because data on markets are much easier to get than like, macroeconomic data, other data of higher quality for markets, things that economists tend to work with. So I think we had an advantage there and we took advantage of it.

At our User Conference you mentioned the CRSP tapes [Nine-track magnetic tapes used with mainframe computers] and how finance professors all over the world have access to them and are “spinning the tapes,” looking for something in the data. Now that people have begun to spot these anomalies in the market, do you think these anomalies are a recent phenomenon or do you think because so many people are watching the market now and have access to the data and are looking at it, that these anomalies were always there, that we are just now beginning to see them?
Well, I think it is probably the second. The way science goes, I think, is that if they’re just anomalies, they’re not very interesting—because you may get the result of dredging through the data looking for anomalies. Until you find something that can replace the theory of efficient markets with a systematic alternative theory, you don’t have anything. I think efficient markets [theory] is going to stand as the paradigm until there’s an alternative paradigm that’s rejectable. Scientific hypothesis is either true or false. The problem with anomalies is that you don’t see one you don’t like. Under-reaction is as good as over-reaction. But you have to develop a scientific theory that would explain this phenomenon.

Are you fairly involved with DFA (Dimensional Fund Advisors) at this point in time?
Quite a bit. All their fixed-income products are based on research that I did ten or twenty years ago. And now they have all these value products that are based on the Fama/French research. And I’ve been on their board of directors since they started.

Does working with DFA affect your attitudes? Do you see yourself getting more involved with the application-side?
I keep them in the loop. I send them all the papers that I think are relevant to an investment manager. I try to keep them educated with what’s going on in academics so they can be on the front line. They’ve taken up on some of the research that I did, on some of the research that I’ve generated along with Ken French. They’ve got a product based on that. As a result, I might do some client contacting. Speaking engagements. I do a lot of talking in general. But they do all the business side. I don’t do any of that.

A lot of your work would state to an investor, do not worry about the active managers, focus more on the indexes.
Focus more on . . . the choice of stocks versus bonds is much more important than the choice of active versus passive [management]. They use the term sectors, I guess they call it. I guess style, the choice of style is more important. These words change every year. We don’t use them in academics. [Laughs] It’s hard to latch on to the current—

lingo. I think it’s style at the moment. But style is much more important than active versus passive. The stock-bond mix, and within the stock mix to what extent you go to small stocks versus big stocks, whether you go value versus growth, and then you have the whole international business to get to.

A lot of your work has shown evidence that it’s nearly impossible for an individual to beat the market. All the evidence seems to suggest that if you can’t beat the market, then you might as well join it by investing in an index. And a lot of what Ibbotson Associates preaches to clients and the investment community is asset allocation. Given all this, wouldn’t you agree that for an investment professional the single most important decision to make would be the asset allocation policy decision or the appropriate mix of assets?
Absolutely. That’s the only decision in my view. That’s what I meant that style is everything that active versus passive is—active management, I think, is kind of like investing in pornography. Some people like it but they’re not really getting better than real sex. If you’re willing to pay for it, that’s fine. But don’t pay too much. Pay more attention to your style [asset mix].

How do you invest?
I’m a passive investor. All stocks. I believe in efficient markets. I know nothing about stock picking. I don’t trust anyone else to interpret the data better than myself so I don’t believe the opportunities are there to beat the market.
Interface: In your 1992 paper “The Cross-Section of Expected Stock Returns,” co-written with Ken French in The Journal of Finance, you state that efficient market hypothesis has become “more thornier” in the twenty years since your previous work. In what way has it become more thornier?
Gene Fama: The initial tests that were done were kind of naive. For example, they assumed that the level of expected return would be pretty constant through time. There’s no reason in economic theory, or in efficient markets, for that to be true. So as people get more sophisticated about models of markets of equilibrium and the interplay of macroeconomics and markets, then the notion that expected return varies through time has to be faced.

For example, expected return on stocks over bills is not a constant. Sometimes it’s narrower; sometimes it’s bigger. And you have good economic reasons for why that might be true.

The early tests of efficient markets didn’t allow for such things. The theory of efficient markets has to be looked at on a context of a model of market equilibrium that tells you what prices should look like. Then efficient markets theory asks whether they do look like that model of market equilibrium. A naive model of market equilibrium is going to produce bad tests of market efficiency. A bad model of a market equilibrium of any sort, to be exact, will result in bad tests. So you always have this problem of not being able to test whether the market’s efficient outside of the model of market equilibrium. And all these things are just models so they’re false. They’re not exact depictions of the world. So the world gets more sophisticated.

Do you feel it’s possible to beat the market using information other than price such as earnings reports, economic information, financial statements; or do you think that price already reflects all that?
It surely doesn’t fully reflect all that. But the relevant question is, ‘how many people can actually take advantage of that?’ And I think that number is very low. I think that most plan sponsors pay for active management and don’t get the benefits. If you think of all the active management—if you really were a market beater why would you pass those returns on to your customers? Why wouldn’t you internalize it and keep it into a smaller pool of participants. The whole agency relation between the active manager and the customer is kind of suspicious to begin with. Plus there’s very little evidence that active management as a whole does anything but subtract value to cover the costs that they incur.

Rex Sinquefield [of Dimensional Fund Advisors], who’s Roger’s co-author on Stocks, Bonds, Bills, and Inflation, says there are three classes of people who don’t think markets work: the Cubans, the North Koreans and active managers.

In terms of the efficient markets theory, do you feel that large institutions are the only ones that have the resources to do the research and exploit any inefficiencies? That this is the reason why the rest of us—the individual investors, the majority—can’t beat the market?
I don’t think there’s a lot of evidence that institutions do it [beat the market].

Like a full time research house, you don’t think that they can exploit inefficiencies?
No. If you look at those full time research houses, they turn out to be full time marketing houses. They’re trying to sell things. They’re not doing it [beating the market]. There’s not much evidence that they do it.

The things people turn to, like Warren Buffet or Peter Lynch, people that have these incredible track records—whether or not they can or not [beat the market] is another issue. But if you say that their records are real, well, these turn out to be individuals. They’re on their own. They’re not the research houses. In Buffet’s case he’s not even following the market. He’s just following a few things. So I think all the research houses are just marketing hype, not real research.

So you feel if anyone is exploiting the market’s inefficiencies, that they’re probably not telling anybody about it, that they’re just raking it in for themselves.
Right, but I think that’s part of it, to the extent that there is any of that. Take for example studies on inside trading—you’d think that if anybody had good information about a company that wasn’t publicly available it would show up with huge insider returns. So the evidence is: is there any difference? Insiders make money trading on their own stocks. But they don’t make a lot of money. The average returns that they get in terms of buying their own stocks are about 1 percent, or a little more maybe, or a little less maybe. Not a lot.

The risk is probably greater than the return.
Yeah, you’re right. They’re taking a big chance with insider information. And it doesn’t seem to be worth a lot. So the ones not trading on insider information must have even less. If insiders don’t have a lot, what are the outsiders going to have?

In your presentation at the User Conference you presented evidence that small value companies have a proven track record for beating the market. How do you reconcile that with your past work on efficient markets?
We [Fama/French] think that there are basically three risk factors that you use to explain stock returns. Market factor, to distinguish stocks from bills. Another factor, which for one reason or another we don’t quite understand, tends to be related to size. So small stocks versus big stocks. The third one’s related to value versus growth.

We’ve written a series of papers which indicate that there are at least three risk factors in the returns, and sorting stocks on these dimensions seems to capture them. When we say that value stocks earn higher average returns than growth stocks we think that’s a compensation for real risk, not an under-pricing or some over-reaction. That’s the tack we’ve taken in our research, that we tried to document.

Do you believe that this compensation results from the existence of so many small companies? Perhaps there just aren’t enough analysts or people to follow them, allowing room for investors to exploit these opportunities?
Well these are returns that are earned by passive investments. Studies like the small stock premium just take all small stocks. They’re not trying to select the ones that are under-priced. They’re just buying them all. You’d have to explain why the whole population is undervalued to do it.

Can you talk a little about some of your other work outside of your famous papers, “Random Walks” and “Efficient Markets.”
Oh, I’ve worked in lots of different areas—statistics, economics, finance—through time. At the time that I’m doing it, that work always seems the most interesting. At the moment, I’m working more on corporate issues. I don’t know how long it’ll last but at the moment that seems very interesting to me.

In the past I’ve worked on organizational economics with Mike Jensen [finance professor at Harvard Business School]. Miller [Merton Miller, finance professor at University of Chicago and 1990 Nobel Laureate in Economic Sciences] and I wrote The Theory Of Finance 25 years ago. Which I think was the first real attempt to put together all the theoretical seedings of portfolio theory: capital markets, capital market equilibrium, and integrate it into corporate finance. It didn’t sell much though.

Why do you teach?
[Laughs] It’s hard to imagine an easier job. It doesn’t interfere with my athletic life very much.

I hear that you’re quite active with tennis.
I play tennis. I wind-surf. I bike a lot in California. Mostly it’s tennis and wind-surfing at this point.

What do you think you’ll be known for 20 or 30 years from now or what work of yours do you think will be most remembered?
Who knows. I’ll be happy if it’s anything. But I won’t care because I’ll probably be dead.

Copyright © 2000 Ibbotson Associates
1 800 758 3557