Introduction

Despite a strong theoretical basis for the random-walk model of stock returns (Bachelier, 1900), evidence for predictable patterns in stock returns has mounted. Two particularly well documented patterns are the momentum and reversal effects.1 These effects document that stocks that exhibit high (low) relative past returns tend to exhibit high (low) abnormal returns over ‘‘intermediate’’ horizons of 3–۱۲ months (i.e., momentum) and low (high) abnormal returns over ‘‘long’’ horizons of 24–۳۶ months (i.e., reversal). The central interesting feature of momentum and reversal is the persistent time delay in how prices evolve. If investors can anticipate momentum or reversal patterns, researchers question why competitive trading pressure does not eliminate these intermediate or long horizon effects by adjusting security prices immediately to reflect the anticipated returns. Why does the market delay in the realization of these anticipated returns? Identifying the source of this delay in equity price return adjustment remains a compelling research question. The literature has proposed several hypotheses to explain the delay in price adjustment. Barberis et al. (1998) and Daniel et al. (1998) propose that investors maintain behavioral biases in updating their beliefs. Hong and Stein (1999) provide a model based on the interactions of investors that respond to different information to explain the price delay. Vayanos and Woolley (2013) propose a delay caused by the inertia associated with investment-fund flows. In this paper, we propose a new explanation for the time delay in returns observed in momentum and reversal. Our proposal emphasizes the role of firm investment in stock price behavior, and includes as a key feature the impact of time delay associated with realizing firm investment. We suggest that since real firm investment cannot be instantaneously executed, return patterns are delayed in time because investment realization is inherently delayed in time. Using a panel of U.S. CRSP-Compustat stocks, we show that momentum and reversal patterns in stock returns occur exclusively in concert with the predicted patterns in firm investment. For example, winners continue to win only when there is subsequent investment, and losers continue to lose only when there is subsequent disinvestment. We suggest that the momentum and reversal regularities in returns do not occur in isolation, but are dependent on systematic patterns in firm investment such that there is no residual momentum or reversal effect in stock returns independent of that associated with firm investment. While we do not empirically show causation, our results are consistent with a unique linkage between investment and both the momentum and reversal return regularities. We motivate our proposal with two hypotheses based on simple motivating assumptions of the term-structure in returns and based on established relations. Our hypotheses result from three assertions. First, firm investment takes time to implement such that investment execution lags with some delay the manifestation of the investment opportunity.