Bubbles, fundamentals, and investment: A multiple equation testing strategy

Abstract

Dramatic fluctuations in the stock market raise questions about whether actual asset prices correspond to the expected present value of future cash flow and whether deviations from this fundamental price can affect real investment spending. Even if there are deviations from fundamental price, they may not distort real behavior if firms ignore these deviations in making their investment decisions. On the other hand, overvaluation of equities could provide firms with a relatively cheap source of finance and might therefore influence investment. To evaluate these issues, this paper introduces a new econometric testing strategy, and assesses the existence of stock market bubbles and the sensitivity of investment spending to bubbles. We estimate the Q and Euler equations in a simultaneous equations model and exploit the idea that these equations reflect different information about the stock market and investment spending. Based on U.S. data for 1911–1987, our formal statistical tests indicate that bubbles exist but real investment decisions are based on fundamentals. A variety of robustness checks and three types of collateral evidence corroborate this interpretation.