Created: 99-04-15
Last Modified: 99-04-15
Go to
Brad DeLong's Home Page

Teaching | Writing | Career | Politics | Book Reviews | Information Economy | Economists | Multimedia | Students | Fine Print | Other | My Jobs

Noise Risk and Financial Markets

J. Bradford DeLong

A talk at the Fourth Annual J.P. Morgan Client Equity Derivatives Conference, Stein Eriksen Lodge, Deer Valley, Utah, April 8-11 1999

"If the reader interjects that there must surely be large profits to be gained... in the long run by a skilled individual who... purchase[s] investments on the best genuine long-term expectation he can frame, he must be answered... that there are such serious-minded individuals and that it makes a vast difference to an investment market whether or not they predominate... But we must also add that there are several factors which jeopardise the predominance of such individuals in modern investment markets. Investment based on genuine long-term expectation is so difficult... as to be scarcely practicable. He who attempts it must surely... run greater risks than he who tries to guess better than the crowd how the crowd will behave." (John Maynard Keynes (1936), p. 157.)


Almost any professional economists will tell any non-professional investor--and most professional investors--that their best investment strategy is buy and hold: figure out what you risk tolerance is, invest the appropriate fraction of your wealth in stocks, and invest in stocks by buying the market portfolio--a broad diversified array of equities. After all, all non-professional and most professional investors do not truly have an informational edge over the market. Thus when they buy something they are running a real risk of buying it from someone who genuinely knows more than they do--in which why are the more-informed so eager to sell? Unless you have a real informational advantage you cannot expect to make money by trading, and so you should diversify in order to reduce risk.

Yet--index funds aside--non-professional investors and many professional investors whose macroeconomic and fundamental analyses are by no means deep enough to support any credible claim of an informational advantage fail to diversify. They log onto The Street.com. When individual investors do diversify they entrust their money to mutual funds the majority of which charge relatively high fees while failing to beat the market. Moreover, individual investors churn their mutual fund holdings at a furious rate.

Even though investors don't behave as economists say that they should, nevertheless economists feel safe telling everybody else that asset prices still behave as economists say they should. Economists preach the gospel of the efficient market, even though investors don't behave as the efficient market hypothesis said that they should.

The argument that "noise traders" do not affect asset prices--at least not in the long run--is, like much else in economics, Milton Friedman's, and is more than fifty years old. Milton Friedman pointed out that people who trade more-or-less at random are eventually going to run up against people who have a sounder grasp of values and of where values are going. As Friedman said, "noise traders" are inevitably going to sell low, and buy high. Thus they will find themselves losing money, and as they lose money they will lose influence in the market and lose their ability to affect asset prices.

Yet when we look at asset markets--especially equity markets--today or any day in long-term historical perspective, it is kind of hard to think that the efficient market hypothesis--that whatever is generally known or even semi-widely known is already in the price--is even approximately correct. Dividend yields on broad market indices like the S&P composite today are less than half what they had been at previous troughs in dividend yields, whether in the early 1970s, the late 1890s, or 1929. Some of this you can explain by reference to the fact that dividends are a lousy way to get cash out of the corporation into the hands of shareholders, and that the true mystery is why dividend yields have been so high for so long.


But corporations in the aggregate appear to have realized sub-market rates of return on their retained earnings. And earnings yields on broad market indices are also at record lows. You can explain some of that by asserting that for a long time the gap between average rates of return on equities and average rates of return on other assets has been too high--but in the past every time someone has claimed that old bases of valuation have changed and no longer apply, they have been wrong. Over the course of the past century and a half, a high market-wide price-earnings or price-dividend ratio has been a sign not that future earnings and dividend growth is going to be high, or that required expected ex ante rates of return on equities have fallen, but that market-wide price-earnings and price-dividend ratios are going to come down.

So these considerations lead a bunch of people--Richard Thaler and Rob Vishny at the University of Chicago, Andrei Shleifer at Harvard, Larry Summers before he became a bureaucrat--to want to look back at exactly how air-tight was Milton Friedman's argument that we shouldn't have to worry about destabilizing or random speculators--that they couldn't affect prices much for long.

And the answer seems to be that one circumstance could undermine a bunch of Milton Friedman's argument. If the people who seriously collect information about fundamental values or about macroeconomic trends have to show results relatively quickly--if they have to prove to their bosses, their clients, their customers within less than five years that they know what they are doing--then it is very likely that people who trade essentially at random will not only affect prices, but also survive and flourish in the market--and may well see their returns exceed those of better-informed investors, albeit at the price of being exposed to truly extravagant levels of risk.

For there is an important source of risk borne by anyone knowledgeable about fundamentals who bets against what appear to be noise-driven beliefs about asset values: there is the risk that noise trader beliefs might become even more extreme in the near future, and that their price pressure will not disappear until after the fundamental investor has to show results--to the boss, or the client.

Moreover, the mere fact that people who know little about fundamentals have opinions about a security, and trade it, makes it more risky. If this risk is priced, then noise traders create room for themselves by thus raising the expected rates of return on those assets in which they take positions. Noise traders can earn higher expected returns from their own destabilizing infiuence, not because they perform the useful social function of bearing fundamental risk.

Now Milton Friedman's buy high-sell low effect is still there. But in our world the fact that fundamental investors do not have multi-generation horizons limits the size of the positions they seek to take in opposition to noise-driven price movements, so the buy high-sell low effect is weaker than Friedman imagined. And there are these other considerations: the fact that noise-driven price movements increase risk, and in a risk-averse world drive up average rates of return on the securities in which those who listen to The Street.com concentrate their portfolios.

And when we look at the world it is very hard to understand the movements either of market averages or of individual stocks--or even of securities like orange juice futures--in which the big source of fundamental news is how many days below freezing you have in California and Florida--as due to news either about future profitability or about future discount rates. For in the long run measures of scale like price-dividend ratios do have predictive power for asset returns independent of interest rates or economic growth rates: when prices are high relative to historical average multiple of dividends, prices are likely to fall. In the bond market the expectations hypothesis of the term structure-or, indeed, any model with relatively constant term risk premia in which the long rate is related to an average of expected future short rates--would predict that when the term structure is especially steep that long-term interest rates should be on the rise. Instead, when the term structure is especially steep long-term interest rates are likely to fall.

In a world with noise, economists' advice that relatively uninformed investors pursue buy-and-hold strategies is simply wrong because generally available information is not all already included in theprice. Instead, the preferred investment strategy is some form of contrarian: bet on your knowledge of macroeconomic trends, or on your informed assessments of fundamentals.

But such strategies have two problems. First, how can you be sure that you are in the more rather than the less informed part of the market? Second, successful pursuit of such contrarian investment strategies requires a very long time horizon, very patient bosses and clients, and a willingness to bear a lot of risk. Anomalies such as the high dollar of the mid 1980s and the extraordinary price-earnings ratios found in Japanese stocks in 1987-89 persisted for years even though many investors recognized these anomalies because betting against such perceived mispricings requires bearing a lot of risk.

Let me close with one final thought. Many market practitioners share a view of the relative social merits of "speculation" and "investment" that has found little sympathy among academics. Participants in financial markets often argue that the presence of traders looking only for short term profits and not long-term values is socially destructive.

The standard economist's refutation relies on recursion: If one seeks to buy a stock now to sell in an hour, one must calculate its price in an hour. But its price in an hour depends on what those who will purchase it think its price will be a further hour down the road. Anyone who buys an asset--no matter how short the holding period--must perform the same present value calculation as someone who intends to hold the asset for fifty years. Thus a linked chain of short term "traders" performs the same assessment of values as a single "investor," and short-term trading cannot be worse than investing. Prices are unaffected by the holding period as long as rates of time discount and willingness to bear risk are unchanged.

But once you drop strong belief in the efficient market hypothesis, then the holding period does matter. And in general the more patient is capital, the higher is investment and the better-performing is the economy.


Professor of Economics J. Bradford DeLong, 601 Evans Hall, #3880
University of California at Berkeley
Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax

This document: http://www.j-bradford-delong.net/Econ_Articles/JPMorgan/JPM_Noise_Risk.html

Search This Website