MITOCW | watch?v=tL7Lcl90Sc0 The following content is provided under a Creative Commons license. Your support will help MIT OpenCourseWare continue to offer high quality educational resources for free. To make a donation or to view additional materials from hundreds of MIT courses, visit MIT OpenCourseWare at ocw.mit.edu. PROFESSOR: What I want to do today is to continue where we left off last time in talking about the empirical properties of stocks and bonds. I want you to develop an intuition for how to think about markets. We've already done that over the course of the last few lectures by looking at market prices and understanding how to price them, but I'd like you to get some kind of a historical perspective now on specific asset classes. Because we're going to be relying on market prices to make inferences about other kinds of securities and other decisions you're going to make. As I told you at the very beginning of the course, we're going to rely on markets for information, because it's the wisdom of crowds that really gets us the information we need in order to make good financial decisions. So I want to begin that process of now giving you the intuition about the wisdom of crowds by looking at the historical performance of stocks and bonds. And then we're going to talk about how to quantify risk more analytically and put it all together in the very basics of modern portfolio theory. So I want to start by asking the question, first of all, what characterizes US equity returns? How do we get our arms around the behavior of that asset class? And the way I'm going to do that is to give you some performance statistics about the volatility, about the average return, about how predictable they are, and also patterns of returns across different kinds of stocks. So we're going to look at some empirical anomalies before actually turning to the analytical work of trying to figure out how to make sense of this from a more formal mathematical framework. Before I do that, let me ask you to think about the following question, which is, if you are designing a market for stocks, what properties would you want that market to have? And I'm going to argue that there are a few properties that all of us, I think, can recognize as being good properties for stock prices. So the first is that stock market prices are random and unpredictable. Now, that might seem a little counterintuitive, and certainly I think you would acknowledge that over the last several weeks markets have been supremely unpredictable. And that doesn't feel so good. It doesn't seem like that's a good thing. But in a minute, I'm going to try to make that a little bit more clear by looking at the alternative of predictable-- or unpredictable, which is predictable. So let me come back to that point.
The second property that I think you'll agree is a reasonable one for us to expect is that prices should react quickly to new information. It should adjust to new information really without any kind of delay. And finally, we'd like to see that investors shouldn't be able to earn abnormal returns after you adjust for risk. So in other words, once risk adjustment is taken into account, there shouldn't be any additional return left over. That's what we think of as a well-functioning market. Another way of putting it is that a market is highly competitive. It's hard to make money in those markets. Now, they may not be markets that you would enjoy trading in, but that's not the question. The question is, what would be a good market, an efficient market? So let me talk about predictability for a minute, because I said that it seems a little counterintuitive that a good market is one that's not predictable. So let's pretend that this is the stock market. This is the S&P 500. That looks nice-- a nice, regular curve. Anybody come up with a prediction for this? How would you go about predicting the behavior of this kind of a stock market? What's that? STUDENT: Cyclical. PROFESSOR: Cyclical. What kind of curve would you fit to this? STUDENT: Sine wave. PROFESSOR: Yeah, sine wave. In fact, that's how I generated this. I used a sine wave, and then I add a little noise. Now, why might this not be a good model for a market? If this were the stock market, what would you do? Yeah? STUDENT: Everybody would buy on the low and sell them high [INAUDIBLE] the other end. PROFESSOR: Exactly. After a few of these cycles, you sort of get the idea. And if you're down here, you're going to think, well, gee, I think it's likely to go back up so I'm going to buy a ton over here. And when you get right up there, you'll say, gee, you know, I think it's time for me to sell a ton. And you don't have to go through too many of these before you get richer than your wildest dreams. Yeah? STUDENT: I would think that [INAUDIBLE] regular market like that is not true, because as soon as you
want it to raise, it's going to collapse at 50, 10 points below. PROFESSOR: That's exactly right. So as soon as you start doing this, as soon as you try to do this, what happens to the pattern? The pattern disappears-- exactly. You see, this is one of the reasons why finance is a lot more challenging than physics. In physics, if you try to drop a ball in a gravitational field, it won't change its mind and say gee, now I'm going to change the gravitational constant on you just because you're testing me. But in financial markets, the moment you try to take advantage of this pattern, the pattern changes. In fact, the more you try to take advantage of it, the more quickly the pattern changes. In fact, if you do this a lot-- if there are a lot of people trying to predict patterns-- then you know what you get? You get no pattern. You get randomness. That's the idea behind an efficient market being random. If it were not random, then that means that there aren't enough people who are bothering to try to forecast the price and incorporate information into the price. Now, I said two things that at first seemed different, but in fact they're opposite sides of the same coin. When you are forecasting market prices, you know what you're doing? You're actually helping markets become more efficient by incorporating information into that price. How do you do that? Well, if, for example, you think that having a presidential election will cause volatility to decline, then if you know that there's a presidential election coming up, you will start trading in a way that will ultimately be betting on volatility declining. As you start that trading, you force that volatility index to go down. So the fact that you've got information and you think you can forecast prices, when you use the information, what does it mean to use the information when you buy or sell securities on the basis of that information? Then the price of the security ultimately reflects the information, right? So an efficient market is one where you don't have this. You don't have a very strong predictability. If it is strongly predictable, then most likely either the market is rigged, or there aren't enough people that are trading in order to make prices fully reflective of all available information. Now, this is the way markets really look. These are random walks with drift, drift meaning there's a positive trend or, in some cases, a negative trend. But otherwise, it's random around
that trend. So you can't really easily forecast it. And you can see that prices go up, they go down. There are long periods where they go up, but there are also, for other stocks, long periods where they go down. And you don't know what's going to happen next. This is a sign of a very efficient market. A while ago, there was an academic study that was done to try to test for efficiency, and one of the tests was that if the underlying price series was not very volatile, that was considered an efficient market. But it turned out that was roundly criticized because of the point that just because a market is not volatile, it doesn't mean that it's working well. And an example was, at the time-- this was, like, 20 or 30 years ago-- the Chinese stock market, the Shanghai Stock Exchange. It was a relatively young market, and at that time, there were only two stocks that traded on it. It was the National Railroad Company and the Bank of China. And at that time, which is, again, about 15 or 20 years ago, it was considered unpatriotic to sell the security if you had bought it. So you could buy it, but you weren't allowed to sell it. And so the price went way up and up and up, and that's not an example of an efficient market. It was not at all volatile. But as a result, there was no real information reflected in that price. Yeah? STUDENT: India is talking about getting out. It looks like somebody's doing one of those roller coaster rides that goes up and down and up and down. So is volatility a sign of [? inefficient ?] [INAUDIBLE]? PROFESSOR: Well, it's not volatility, per se, but rather the combination of the predictability per unit volatility. That's really what you want to focus on. We're going to come back to that when we talk about portfolio theory and look at this trade-off between risk and expected return. But no, I wouldn't say that the Indian market is inefficient. It's undergoing some pretty significant changes, as is the US and as is the world. But that's because the global economy is contracting as we know it because of this financial crisis. So I wouldn't characterize it as inefficient at this point, but that remains to be seen. Yeah? STUDENT: [INAUDIBLE] down the road [INAUDIBLE] market [INAUDIBLE]? PROFESSOR: There isn't any hard and fast rule, no. But if you take a look at the trade-off of risk to reward, in
Recommend
More recommend
