Investment Waves under Cross Learning∗

We investigate how firms’ cross learning amplifies industry-wide investment waves. Firms’ technologies are subject to idiosyncratic shocks and a common shock, and their asset prices aggregate speculators’ private information about the two types of shocks. In investing, each firm learns from other firms’ prices (in addition to its own) to make better inference about the common shock, leading to higher investment sensitivity to the common shock. To respond, speculators put a higher weight on the common shock in trading, making prices even more informative about the common shock. This spiral results in higher investment and price comovements in investment waves. Moreover, cross learning imposes a new pecuniary externality on other firms, because it makes their prices less informative about their idiosyncratic shocks thanks to speculators’ endogenous overweighting on the common shock. This externality increases in the number of firms, suggesting that more competitive industries may exhibit more inefficient investment waves.