Asset pricing.

• Author: Campbell, John Y.

The NBER established its Program in Asset Pricing in the fall of 1991. This program, together with those in monetary economics and corporate finance, grew out of the earlier financial markets and monetary economics program, and has close links with the Program in International Finance and Macroeconomics. Asset pricing is the study of markets for financial assets, including stocks, bonds, foreign currencies, and derivative securities, such as futures and options. It is a highly technical field of economics, but also one in which new ideas are applied rapidly by practitioners, who take a keen interest in academic research. NBER economists have been studying a variety of topics within the field, including general equilibrium asset pricing models, international financial integration, derivative securities, and some intriguing microeconomic puzzles about asset price behavior.

Asset Pricing in General Equilibrium

One fundamental insight of modem financial theory is that a "stochastic discount factor" exists that can be used to calculate the expected return and price of any asset, given information about the pattern of its cash flows. Without this factor, investors could make riskless profits through arbitrage operations. Different asset pricing models imply different stochastic discount factors, and some models include many stochastic discount factors that will price the assets that are traded. But any model that rules out arbitrage opportunities has at least one stochastic discount factor.

Several NBER economists have been trying to characterize the stochastic discount factor that prices the assets traded in today's world economy. John H. Cochrane and Lars P. Hansen have provided a useful general overview, showing that the stochastic discount factor must be highly volatile if it is to price U.S. equities and fixed-income securities.(1) This poses a problem, because the standard macroeconomic asset pricing model, which aggregates investors into a single "representative agent" who consumes aggregate consumption, implies that the stochastic discount factor is not very volatile when the representative agent has standard preferences and attitudes toward risk.

One response to this problem is to explore different models of investors' preferences. Phillip A. Braun, George M. Constantinides, and Wayne E. Ferson have argued that past consumption may increase the marginal utility of today's consumption; this "habit-formation" effect greatly increases investors' risk aversion whenever consumption is close to the habit level determined by the recent past history of consumption.(2) Geert Bekaert, Robert J. Hodrick, and David A. Marshall have explored a model of "first-order risk aversion" in which investors are more concerned about small risks than they are in the standard model.(3) Shlomo...