Nobody quite knows what "behavioral finance" means, and that's a good thing. Essentially, modern finance theory began with the notion of "efficient markets", risk-return tradeoffs, and complete markets. That theory remains the mainstream in academic circles. But at least since the 80s, finance researchers have been noticing "anomalies" in the predictions of efficient-market theories and mainstream risk models. Attempts to explain these anomalies, or to dig up more of them, loosely fall under the heading of "behavioral finance", probably because in the 80s, behavioral economics was just becoming popular. But behavioral finance isn't just behavioral econ applied to finance; it includes a whole big eclectic mix of ideas and observations about market institutions, overall market movements, and information processing, and at this point is really just "anything that isn't efficient-markets finance". The literature is really all over the place. But here are a few foundational papers to get you started.
"Speculative Investor Behavior in a Stock Market With Heterogeneous Expectations", by J.M. Harrison and David Kreps (1978)
This paper came out at a time when devotion to extreme rationality was de rigeur in most economic circles. Harrison and Kreps assume that investors have extreme irrationality, being so overconfident in their differing beliefs that they never reach agreement. As a result, financial assets carry a "speculative premium", because holding an asset gives you the option to resell it in the future to someone more optimistic than yourself. This paper really spawned the whole "overconfidence" literature, and the "heterogeneous beliefs" literature too. When Tom Sargent called himself a "Harrison-Kreps Keynesian", this is the paper he was talking about.
This paper came out at a time when devotion to extreme rationality was de rigeur in most economic circles. Harrison and Kreps assume that investors have extreme irrationality, being so overconfident in their differing beliefs that they never reach agreement. As a result, financial assets carry a "speculative premium", because holding an asset gives you the option to resell it in the future to someone more optimistic than yourself. This paper really spawned the whole "overconfidence" literature, and the "heterogeneous beliefs" literature too. When Tom Sargent called himself a "Harrison-Kreps Keynesian", this is the paper he was talking about.
"On the Impossibility of Informationally Efficient Markets", by Sanford Grossman and Joseph Stiglitz (1980)
This paper basically killed the "strong form" of the Efficient Markets Hypothesis. The intuition has become so common that you might even be surprised that someone wrote a paper on it. Basically, if all information is already incorporated in market prices, then there's no incentive for anyone to go gather information. In the model in this paper, prices are just "wrong" enough to justify the cost of going out and finding new information. But we all know that in the real world this is highly unlikely to be true. The idea of "information externalities" in financial markets comes in part from this paper.
"Do Stock Prices Move Too Much to be Justified By Subsequent Changes in Dividends?", by Robert Shiller (1981)
What if your weather forecaster kept switching his one-week-ahead forecast between 85 degrees Fahrenheit and 15 degrees Fahrenheit? You would fire that forecaster. But if you think stock markets are efficient, then stock markets should be the best possible forecaster of future dividends, since dividends are the "fundamental value" of stocks. In this famous paper, Shiller showed that stock prices fluctuate more than the eventual dividends that they were supposed to have forecast. Whatever is making stock prices gyrate, it doesn't look rational.
What if your weather forecaster kept switching his one-week-ahead forecast between 85 degrees Fahrenheit and 15 degrees Fahrenheit? You would fire that forecaster. But if you think stock markets are efficient, then stock markets should be the best possible forecaster of future dividends, since dividends are the "fundamental value" of stocks. In this famous paper, Shiller showed that stock prices fluctuate more than the eventual dividends that they were supposed to have forecast. Whatever is making stock prices gyrate, it doesn't look rational.
"Information, Trade, and Common Knowledge", by Paul Milgrom and Nancy Stokey (1982)
This paper might not strike you as behavioral, since it discusses hyper-rational behavior in an efficient market, and was written by the wife of Robert Lucas. But actually, this paper is about the fact that rational-expectations theory almost certainly can't explain real-world markets. The basic reason: if everyone were rational, and this was common knowledge, there wouldn't be much trading, because everyone would always be second-guessing their counterparty (e.g. "Why would you sell me this stock at $100 if you didn't think it was worth less than $100?", and so on). The fact that we see such huge trade volumes in real-world markets means that there is probably a lot of irrationality out there.
"Does the Stock Market Overreact?", by Werner DeBondt and Richard Thaler (1985)
Here's an easy way to beat the stock market: Buy the stocks that did the worst over the past 5 years, and hold onto them for another 5. Ta-da! DeBondt and Thaler show that this strategy makes more money than just holding an index fund. Under Efficient Markets theory, that should not be the case. This is known as the "long-term reversal" anomaly.
"Bubbles and Crashes in Experimental Asset Markets", by Vernon Smith, Gerry Suchanek, and Arlington Williams (1988)
The paper that started essentially all of experimental finance. Basically, the authors put some college kids in the simplest possible artificial market setup. It should have been a no-brainer, but what happened, again and again and again, was a giant bubble and crash.
Here's an easy way to beat the stock market: Buy the stocks that did the worst over the past 5 years, and hold onto them for another 5. Ta-da! DeBondt and Thaler show that this strategy makes more money than just holding an index fund. Under Efficient Markets theory, that should not be the case. This is known as the "long-term reversal" anomaly.
"Bubbles and Crashes in Experimental Asset Markets", by Vernon Smith, Gerry Suchanek, and Arlington Williams (1988)
The paper that started essentially all of experimental finance. Basically, the authors put some college kids in the simplest possible artificial market setup. It should have been a no-brainer, but what happened, again and again and again, was a giant bubble and crash.
"Noise Trader Risk in Financial Markets", by J. Bradford De Long, Andrei Shleifer, Lawrence H. Summers, Robert J. Waldmann (1990)
What would you do if a horde of zombies was running your way? If there were only a few you might stand and fight, but if there were enough zombies you might have no choice but to turn around and run in the same direction, essentially becoming a zombie yourself. This is the basic idea of "noise trader" models, which show how smart, rational traders can be overwhelmed by irrational hordes of "noise traders". In a bubble situation, the smart money finds it necessary to imitate the dumb money, and smart investors "ride the bubble", making the eventual crash even bigger.
What would you do if a horde of zombies was running your way? If there were only a few you might stand and fight, but if there were enough zombies you might have no choice but to turn around and run in the same direction, essentially becoming a zombie yourself. This is the basic idea of "noise trader" models, which show how smart, rational traders can be overwhelmed by irrational hordes of "noise traders". In a bubble situation, the smart money finds it necessary to imitate the dumb money, and smart investors "ride the bubble", making the eventual crash even bigger.
"Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency", by Narasimhan Jegadeesh and Sheridan Titman (1993)
This paper is kind of the mirror image of the DeBondt and Thaler paper above. Jegadeesh and Titman show that if you buy the stocks that performed best over the past three to twelve months and hold onto them for another few months, you will beat the market. This is the famous "momentum anomaly", and represents a severe violation of the Efficient Markets Hypothesis. Together with the long-term reversal anomaly, it implies a market where stock prices regularly "overshoot" and then return to a more reasonable level.
This paper is kind of the mirror image of the DeBondt and Thaler paper above. Jegadeesh and Titman show that if you buy the stocks that performed best over the past three to twelve months and hold onto them for another few months, you will beat the market. This is the famous "momentum anomaly", and represents a severe violation of the Efficient Markets Hypothesis. Together with the long-term reversal anomaly, it implies a market where stock prices regularly "overshoot" and then return to a more reasonable level.
"Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment", by Brad Barber and Terrence Odean (2001)
Guess what? You really suck at investing. And the reason is that you trade too often, ignoring the costs you pay every time you trade. Why do you do this? Well, maybe because you are overconfident, and you think (wrongly) that you know better than the market how to interpret the latest piece of news or data. If this is the reason, then you are more likely to over-trade if you are a man than if you are a woman, since psychologists have shown that men on average display more overconfidence than women. And guess what? The evidence says that this is true. Looking at data from a discount brokerage, Barber and Odean found that male individual investors trade more than female ones, and incur greater losses as a result.
Guess what? You really suck at investing. And the reason is that you trade too often, ignoring the costs you pay every time you trade. Why do you do this? Well, maybe because you are overconfident, and you think (wrongly) that you know better than the market how to interpret the latest piece of news or data. If this is the reason, then you are more likely to over-trade if you are a man than if you are a woman, since psychologists have shown that men on average display more overconfidence than women. And guess what? The evidence says that this is true. Looking at data from a discount brokerage, Barber and Odean found that male individual investors trade more than female ones, and incur greater losses as a result.
"Hedge Funds and the Technology Bubble", by Markus K. Brunnermeier and Stefan Nagel (2004)
If "noise trader" models are right, then "smart money" investors like hedge finds should ride bubbles instead of trying to pop them. Looking at data on hedge fund behavior during the tech bubble, Brunermeier and Nagel find that this is exactly what happened.
If "noise trader" models are right, then "smart money" investors like hedge finds should ride bubbles instead of trying to pop them. Looking at data on hedge fund behavior during the tech bubble, Brunermeier and Nagel find that this is exactly what happened.
"Subjective Expectations and Asset-Return Puzzles", by Martin Weitzman (2007)
One long-standing puzzle in financial economics was the fact that stocks tend to have much better returns than bonds, despite not having that much more risk. But as Martin Weitzman shows in this paper (with some fearsome math), the risk of stocks looks a lot larger if investors aren't sure about the underlying structure of the world. Which, of course, they aren't.
Anyway, these papers will get you started thinking about behavioral finance! Obviously there is much newer stuff, and topics other than these...it's a big and growing field. A very exciting one in which to work, if you ask me...
Update: People have been asking "Why no prospect theory?" Obviously, "Prospect Theory: An Analysis of Decision Under Risk" (Daniel Kahneman and Amos Tversky, 1979) is one of the most important, if not the most important, behavioral econ papers ever written. And people have certainly tried to apply that theory in finance.
One long-standing puzzle in financial economics was the fact that stocks tend to have much better returns than bonds, despite not having that much more risk. But as Martin Weitzman shows in this paper (with some fearsome math), the risk of stocks looks a lot larger if investors aren't sure about the underlying structure of the world. Which, of course, they aren't.
Anyway, these papers will get you started thinking about behavioral finance! Obviously there is much newer stuff, and topics other than these...it's a big and growing field. A very exciting one in which to work, if you ask me...
Update: People have been asking "Why no prospect theory?" Obviously, "Prospect Theory: An Analysis of Decision Under Risk" (Daniel Kahneman and Amos Tversky, 1979) is one of the most important, if not the most important, behavioral econ papers ever written. And people have certainly tried to apply that theory in finance.
No comments:
Post a Comment