The award of the Nobel prize in economics to Daniel Kahneman of Princeton University, a pioneer in the relatively new field of behavioral finance, has turned a spotlight on an important branch of financial economics. Yet this branch has been the subject of some disagreement among those who support efficient market theory and those who believe that inefficiencies exist and that insights from behavioral finance can help explain many of them and enable sophisticated investors to exploit them. In physics, quantum theory and relativity have yet to be reconciled, but physicists know they must use both to understand the world. The theories work on different physical scales - quantum theory on the tiny, relativity on the vast.
So, too, behavioral finance and efficient market theory appear inconsistent, yet both are essential to understanding the behaviors of investors and the performance of securities. They, too, work at different scales, in this case different time scales. In the short run - and the short run can last for many years - psychological propensities may drive both market valuations and the prices of individual securities beyond reasonable levels. Over the long run, however, efficient markets will prevail. Stock prices will be dragged back to reasonable valuations. There is good reason why Benjamin Graham is well remembered and Warren Buffett is the second richest American.
Behavioral finance melds psychology into finance and economics to explain how individuals perceive and react to uncertainties, and how that affects their investment decisions - which, in turn, determine market prices. To illustrate, both laboratory experiments and data on actual investors show that individuals are reluctant to realize financial losses. Thus, otherwise risk-averse investors show a lack of rationality by gambling at unfavorable odds once they are in the loss column.
As this year's Nobel economics prize attests, our understanding of human decision processes and their real world consequences has progressed substantially in recent years. The discoveries of severe biases in financial decisions and securities prices are particularly arresting, given that investors have straightforward objectives, much advice and ample data for making decisions.
Pensions, the major investment for most Americans, provide a fertile area for identifying behavioral propensities. Most individuals take little time in making highly important decisions about retirement plans. They often follow the path of least resistance. For example, U.S. firms typically allow each employee to decide whether or not to participate in a 401(k) plan, but usually set a default election. When enrollment is the default, virtually all employees participate. When the default election is non-enrollment, only one-third of employees sign up, despite generous employer matching contributions.
Moreover, three-quarters of employees who enroll by default passively accept the default savings rate (typically around 2% or 3%) and the default asset allocation (typically 100% in a money market fund). Workers who choose for themselves rarely make such conservative choices. Given N available options, be they equity or bond funds, their selection is best described by the 1/N rule: pick randomly.
Given that investors are often non-rational, it is no surprise that markets swing wildly. Individuals tend to extrapolate recent past behaviors to the future. Thus, a major driver of the three-year runup of the S&P 500 starting in 1997 was a 50% rise in price/earnings ratio, to above 30. Individuals simply assumed that rapid price gains would continue. The belief was self-reinforcing, until even wishful thinking could not make it so.
Finally, behavioral finance helps to explain situations where some participants in the market mislead others. Consider the manipulation of corporate earnings. Investors process financial information relative to performance thresholds, such as whether profits meet analysts' expectations. Recognizing this, corporations manipulate earnings so as "not to disappoint," possibly at great future cost. Empirical studies bear out such manipulation: There is a dramatic disproportion of earnings meeting or slightly beating thresholds, with very few just below.
The insights generated by Daniel Kahneman and others, if properly applied, can be used to produce better investment decision making at all levels. At the 401(k) plan level, for example, employers can frame the options in ways that lead employees to make better decisions. At the portfolio level, portfolio managers who are aware of their own biases and proclivities, and those of other investors, may be able to make better investment decisions, and exploit the poor decisions of others.
Richard Zeckhauser is Frank Ramsey Professor of Political Economy, Harvard University, Cambridge, Mass. He runs an executive program for finance professionals on investment decision and behavioral finance, to be held next at Harvard Nov. 18 and 19; Daniel Kahneman is among the faculty.
