Returns to Equity, Investment and Q: Evidence from the United Kingdom

Abstract

Conventional wisdom has it that Tobin's Q cannot help explain aggregate investment. This is puzzling, as recent evidence suggests the closely related user cost approach can do so. We do not attempt to explain this puzzle. Instead, we take an entirely different approach, not using the first-order conditions from the firm's maximisation problem but instead exploiting the present-value expression for the firm's value. The standard linearised present-value asset price decomposition suggests that Q should be able to predict other variables, such as stock returns. Using UK data we find that it has strong long-horizon predictive power for debt accumulation, stock returns and UK business investment. The correctly signed results on both returns and investment appear to be robust, and are supported by the commonly used and bootstrapped standard error corrections, as well as recently developed asymptotic corrections.

Keywords: Investment, Tobin's Q, long-horizon forecasting

JEL Classification: E22, E27