Estimating and Testing Fundamental Stock Prices: Evidence from Simulated Economies

Abstract

A core tenet of financial economics states that, in a market populated by rational investors, the fundamental price of an asset equals the expected discounted present value of its cash flows. This implies, for example, that in a rational and efficient bubble-free market, stock price movements are driven by forecasted changes in dividends and interest rates and not by the irrational exuberance of traders. Interest in the extent to which market prices are fundamentally driven has been particularly high of late, with everyone from academics to the Chairman of the U.S. Federal Reserve Board, Alan Greenspan, offering opinions. Research into fundamental valuation is therefore particularly timely. Several different procedures have been developed to estimate fundamental prices and to test whether market prices deviate in a significant way from the estimated fundamental price. Many studies have reported what appear to be important deviations from fundamentals, especially around market crashes such as those which occurred in 1929 and 1987. Unfortunately, the accuracy of various fundamental price estimating procedures has not been measured by anyone in the literature, to the best of our knowledge. The issues of the magnitude of the bias, of the reliability of fundamental prices, and effects these have on excess volatility tests are analytically tractable under only very restrictive special circumstances. However, they can be evaluated more generally through Monte Carlo simulation. The Monte Carlo is accomplished by generating economies calibrated to the real world, with a known fundamental price, and investigating the properties of the fundamental prices - the bias and the average deviation. The simulations show that popular fundamental price estimation methods are unreliable, and also provide strong support for the absense of bubbles in the S&P 500 index.