Essays on the investment effect

Abstract

The thesis consists of two essays on the investment effect. The first essay is an international examination of the investment effect, and the second essay explores the rationale behind the investment effect based on U.S. data.

A growing body of finance literature has documented an investment effect that firms making larger investments earn lower future stock returns. While the negative relation between corporate investments and future stock returns is well accepted, it is in a great debate for why an investment effect exists. Two major hypotheses are proposed to account for the investment effect, mispricing and rational pricing. The mispricing hypothesis focuses on market participants’ irrational behavior, including managerial overinvestment and investors’ extrapolation of past firm performance. The rational pricing hypothesis, however, centers on examinations of the q theory. Under the q theory, the net present value of potential projects can be high if either the future marginal productivity is high or the future discount rate is low. As a result, for a given level of future profitability, firms making large (small) investments are likely to be those with low (high) discount rates. This predicts low (high) stock returns following large (small) capital investments.

The first essay tests the q theory explanation for the investment effect using international data. I show that the investment effect exists across international markets and differs substantially across countries. I find a stronger investment effect in countries with better corporate governance, lower limits to arbitrage, and more developed equity markets. I construct a composite Q index, based on corporate governance, limits to arbitrage and market development, to separate countries with a strong investment effect from the rest. The empirical results are consistent with the q theory explanation for the investment effect.

The second essay investigates the relation between the investment effect and intangible returns, as well as external financing, in the U.S. market. Extending the model in Daniel and Titman (2006), I operationalize an empirical design of log-linear decomposition of the book-to-market ratio. Using a three-period empirical model, I examine the relations among intangible returns in period one, real investments and external financing in period two, and stock returns in period three. The empirical evidence suggests no additional explaining power of investments for future stock returns, when contemporaneous external financing and prior intangible returns are controlled for. The abnormal return patterns associated with real investments documented in prior studies are consistent with, and part of, the broader return pattern that characterizes the value/growth anomaly. I show that these findings are consistent with the q theory, but inconsistent with the mispricing story.