Listen to the podcast:
Financial market observers may have suffered a bit of cognitive whiplash with this week’s announcement of the joint winners of the Nobel Prize in economics. Two of the winners appear, on first glance, to be polar opposites. Eugene Fama is the father of the iconic efficient markets theory, and Robert Shiller suggests that markets are anything but efficient – often greatly overreacting or underreacting to new information. But Wharton finance professor Amir Yaron says that the ideas of the two winners are ultimately complementary. In this Knowledge@Wharton podcast, he explains why.
An edited transcript of the conversation follows.
Knowledge@Wharton: We’re meeting today with Amir Yaron, a Wharton finance professor, about the recent awarding of the Nobel Prize in economics to Eugene Fama and Lars Peter Hansen of the University of Chicago, and Robert Shiller, from Yale. All won for their work covering trends in asset prices. So, thank you for joining us at Knowledge@Wharton today, Amir.
Amir Yaron: Thank you.
Knowledge@Wharton: There are a couple of interesting things about this year’s winners. Professor Fama is famous for his ideas that markets are efficient. This has long been textbook theory, since he introduced these ideas in the late 1960s, early 1970s — that they reflect all the information – at least, publicly available information — that’s out there.
One idea that flows from this is that you can’t really predict what markets will do in the future, because prices have already taken into consideration all the possibilities. Robert Shiller, you might say, comes along and says [in effect], wait, not so fast. Markets are not that rational. In fact, they can be irrational. They can be too exuberant, too depressed. They overreact, and underreact, in an emotional way.
So, at first glance, it seems a bit odd for these two finance minds to share the Nobel Prize, because in some ways, they seem to be polar opposites. Could you give your views about that?
Yaron: Sure. There’s an element of correctness in that polar view idea that you are stating. But I think there’s a broader message about trying to analyze financial markets. And the way I view the connection is that Gene’s early work focused mostly on short-run predictability. By and large, they’ve shown, using daily or specific announcements, or even weekly or monthly [ones], that there’s very little room for predictability. And I think that evidence, by and large, stands. In fact, that led to practical issues such as [the creation of] index funds — things that have really helped the common person, in the sense of the construction of a low-fee index funds. So, there’s clear contribution there.
The work of Shiller showed – what he’s really cited for is less for the elements of underreaction or overreaction that are due to psychology — his excess volatility paper, which just shows that prices have moved too much, relative to purely constant discount dividends. And then, later on [he] also showed that there is [price] predictability over a multi-year horizon.
One has to understand that this is an important finding. But that finding can be interpreted by a rational model that either attributes [cause and effect in price changes longer term] to different risk aversions, or to different market conditions, which would justify, let’s say, low prices in a recession, that can therefore predict future expected return, and rationalize it. Or, it could be interpreted as [as result of] some behavioral elements that may change expectation. And part of what we are still doing research on is trying to decipher those two phenomena. But the two [views] are complementary, in my view, rather than polar opposites — at least in terms of the findings that the Nobel committee has posed.
Knowledge@Wharton: As a common denominator, would it be fair to say that both [Fama and Shiller] are saying that prices are difficult to predict in the short term, but in the medium and longer term, there’s some ability to make some reasonably accurate predictions?
Yaron: There are certainly certain variables. Let’s say if we’re talking about the aggregate market portfolio, such as the price-dividend ratio. We know that when that’s low, it tends to predict higher return down the road. And the question is, is that due to some behavioral traits? Or, as I mentioned, it could be interpreted that we are in a recession. People are highly risk-averse. We are seeing low cash flows, and therefore people demand higher expected return down the road. And that’s the form of predictability. One could be interpreted [as a result of] … a rational world. And another [interpretation] could be as a more behavioral finance interpretation, which is sort of the path that professor Shiller had taken subsequently.
Knowledge@Wharton: The idea of polar opposites could be seen in that some people have argued that Fama’s ideas and theories didn’t see the financial crisis that began in 2008 coming, whereas Shiller’s view of the world kind of predicted it. And therefore, in that sense, they did have different ways of looking at the world.
Yaron: Well, obviously, Shiller is partly known in the popular press for calling the dot-com and the housing market [bubbles]. As to Gene, I think he would just say – you know, markets, the volatility in the market, it was expected upon such a crisis. That’s what you expect in an efficient market, where it’s hard to aggregate, and there’s a lot of uncertainty, and the difficulty of calling it is inherent in what is happening in such markets.
Knowledge@Wharton: The third prize-winner, who hasn’t gotten quite as much attention as the other two famous names, is Lars Peter Hansen, who you actually have a personal relationship with. Could you talk about that briefly, and then tell us about his ideas?
Yaron: Lars was my main advisor at Chicago. I know him very well. I’ve actually written a paper with him. Lars’ contribution is in developing statistical models, for testing many of the theories that we’ve discussed. The very theory basically asserts that if you save a dollar today, you have an expected return on what that dollar will give you tomorrow. He laid out the foundational ground for doing the testing, and that’s called the general method of moments.
And in the original test, what a dollar means to you today in utility and in the future, the specifications were very simple, and the basic model was rejected. And that was consistent with the notion of rejecting the consumption CAPM [the capital asset pricing model], which is consistent with the Shiller finding that prices were too volatile to be reconciled by a simple [calculation of the] present value of dividends.
Knowledge@Wharton: When you say the model was rejected, what do you mean, exactly?
Yaron: What it means is when you compare in the data, the cost side of saving a dollar today, versus the benefits of getting tomorrow the dollar and the expected return, those didn’t seem to be lined up with reasonable risk attitudes that we think ought to [be able to] reconcile them with. And so, the profession has moved on, consistent with these two background topics that you’ve mentioned: the lack of predictability in the short run, and some predictability in the longer term by Shiller. And this evidence has moved on, in a couple facets. One is to change the preferences, but still stay in a rational world. So, when I say change preferences, I mean … for example, [recognizing] that people are very risk-averse in a recession. And when they see a lot of uncertainty or low expected cash flows, price-to-dividend ratios are low. But nonetheless, expected returns are high. And that is a completely rational story.
Another approach is to go somewhat to the behavioral route, and basically claim that people have certain behavioral biases, and that changes their expectations. And that could [lead to] some changes. A third approach would be to essentially talk more about market frictions. That cost and benefit side that I mentioned – saving today, borrowing today to invest in the market, is not so easily done…. And in this mix, there’s also the issue of how you measure the risks — the present value of dividends. Obviously [that carries] uncertainty. We see the Vix [the Chicago Board Options Exchange Market Volatility Index, often called the fear index] now, and other measures of uncertainty. They become very important. And so, just the measured risks have been challenged, to some extent. And I’ll put in a plug for my own work, which is actually related to that. I like to think it actually influenced some of Lars’ more recent work, which has to do with whether we are measuring the riskiness in dividends and in uncertainty appropriately.
So, if you look at U.S. dividends, they look very much like what we call white noise, which would not rationalize a lot of risk.
Knowledge@Wharton: Why do they look like white noise?
Yaron: They go up and down. So, if you just look at them in a plain vanilla sense, you would think they shouldn’t be too risky. But if you looked at them more carefully, there is some signal there. And part of the debate is whether that signal is there or not — and those are statistical issues. But if it’s there, and with appropriate preferences, again, with a rational story, you can go a pretty long way towards reconciling some of the price movement that we’ve observed. And so, Lars’ contribution has been in developing these methods, pushing them, and also developing methods about what we call alternative preferences, where people are afraid of not knowing the environment they are living in.
So, in that sense, he’s also filling up the gap, to some extent, you could say, a little bit between Shiller and Fama, in the sense that he’s not strictly viewed as a behavioral. But some of his very recent research on robust control, and with another Nobel Laureate –Tom Sergeant — has pushed the agenda of robust control, which takes very seriously the idea that agents are not completely confident of what environment they are in. And consequently, they are behaving [as if they were] in a more fearful environment. And that affects prices in a particular way.
Knowledge@Wharton: So, when an individual investor is going to make a decision about what they’re going to invest in, how do those ideas fit into what might be an unconscious decision for them?
Yaron: One way that we examine the data is on the aggregate economy. And we often look at aggregate data. And that view is in the context of the consumption [based] CAPM [or CCAPM] — that somebody there on the margin is saving a dollar today, let’s say, and investing it. And they’ve got to be compensated the right way. And they are very cognizant when they do that. Of course, there are a lot of people who are – partly because of the lack of predictability — sort of happy, and should be happy, putting their money in index funds, because it’s going to be quite difficult to beat the market by timing it.
Knowledge@Wharton: And this is what professor Fama’s theory had led to, which is you can’t really predict the market. Therefore, an index fund … just tracks the general market.
Yaron: The general markets. Right. And it saves you costs. And so, that’s in terms of the aggregate. Gene also later developed issues about, actually, the failure of the CAPM, in the cross-section. The CAPM basically tells you that stocks that have high-exposure to the market return should have the highest return. But he also showed that other factors – their exposure to size, and their exposure to book-to-market – is very important for understanding the universe of returns. And that has also translated into very practical notions. If you go to many investment houses today, you will see that the universe of stock advice is: Do you want to invest in high growth? Do you want to invest in value? Big [large cap], small? That is, in some sense, a testament of how that research got translated very fast into practical implementation in the real world.
Knowledge@Wharton: Could you just make that connection? How that led to this segregation of the theories?
Yaron: Well, it suggests that it’s not just the market that is the sort of univariate risk that’s out there.… But rather, there are these other two risk factors: Book-to-market, which you can think of as value stocks versus growth stocks, [and size]. …When you go and first put your money [down], [analysts will] divide up the universe of stocks in that dimension. And I think that’s a testament [to the theories], that that view has transcended to the practitioners’ side. Today, when people propose a new model, and whether a particular model is supposed to generate a trading strategy or a better return, it is often benchmarked against what’s called the Fama-French three-factor models. So, it needs to show that it gets a better return not against the CAPM but against this three-factor model.
Knowledge@Wharton: So many modern investment ideas have their foundation on Fama’s ideas. Why do you think it took so long for him to get the recognition that he got this week?
Yaron: I don’t have a great answer to that. His name was out there for a while. Maybe they thought, this is the balance [awarding Fama the Nobel Prize jointly with Shiller]. This balanced view is the right view. I’m not sure. This is probably something the committee should answer.