Rational Asset Prices

The mean, co-variability, and predictability of the return of different classes of financial assets challenge the rational economic model for an explanation. The unconditional mean aggregate equity premium is almost seven percent per year and remains high after adjusting downwards the sample mean premium by introducing prior beliefs about the stationarity of the price-dividend ratio and the (non) forecastability of the long-term dividend growth and price-dividend ratio. Recognition that idiosyncratic income shocks are uninsurable and concentrated in recessions contributes toward an explanation. Also borrowing constraints over the investors’ life cycle that shift the stock market risk to the saving middle-aged consumers contribute toward an explanation. ∗ University of Chicago and NBER. I thank John Campbell, Gene Fama, Chris Geczy, Lars Hansen, John Heaton, Rajnish Mehra, Ľuboš Pástor, Dick Thaler, and particularly Alon Brav and John Cochrane, for their insightful comments and constructive criticism. Finally, I thank Lior Menzly, for his excellent research assistance and insightful comments throughout this project. Naturally, I remain responsible for errors. A central theme in finance and economics is the pursuit of a unified theory of the rate of return across different classes of financial assets. In particular, we are interested in the mean, co-variability, and predictability of the return of financial assets. At the macro level, we study the short-term risk-free rate, the term premium of long-term bonds over the risk-free rate, and the aggregate equity premium of the stock market over the risk-free rate. At the micro level, we study the premium of individual stock returns and of classes of stocks, such as the small-capitalization versus large-capitalization stocks, the “value” versus “growth” stocks, and the past losing versus winning stocks. The neoclassical rational economic model is a unified model that views these premia as the reward to risk-averse investors that process information rationally and have unambiguously defined preferences over consumption that typically (but not necessarily) belong to the von Neumann-Morgenstern class. Naturally, the theory allows for market incompleteness, market imperfections, informational asymmetries, and learning. The theory also allows for differences among assets for liquidity, transaction costs, tax status, and other institutional factors. The cause of much anxiety over the last quarter of a century is evidence interpreted as failure of the rational economic paradigm to explain the price level and the rate of return of financial assets both at the macro and micro levels. A celebrated example of such evidence, although by no means the only one, is the failure of the representativeagent rational economic paradigm to account for the large average premium of the aggregate return of stocks over short-term bonds and the small average return of shortterm bonds from the last quarter of the nineteenth century to the present. Dubbed the “Equity Premium Puzzle” by Mehra and Prescott (1985), it has generated a cottage industry of rational and behavioral explanations of the level of asset prices and their rate of return. Another example is the large increase in stock prices in the early and middle 1990s, which Federal Reserve Chairman Alan Greenspan decried as “Irrational Exuberance” even before the unprecedented further increase in stock prices and pricedividend ratios in the late 1990s.