Structural incoherence and stock market activity

This paper argues that the efficiency of the price-setting process in the stock market is contingent on the coherence of a stock position in the industry-based classificatory structure that guides valuation. While this structure helps investors interpret ambiguous economic news, it is imperfect because stocks vary in the extent to which they are coherently classified, as revealed by the stocks' position in the network of coverage by securities analysts. The main hypotheses are that incoherent stocks are traded more often because such stocks are more likely to be subject to differences in the interpretive models used to understand material information; and that both volume and volatility are higher for incoherent stocks because convergence on a common interpretation relies more heavily on self-recursive market dynamics. These hypotheses are validated via analyses of market activity in the aftermath of first-quarter earnings announcements for US. based firms from 1995 to 2001. The results help reorient research away from debates as to whether financial-market activity is excessive and toward the mechanisms that underlie such activity. More generally, the approach advanced in this paper illustrates how structural sociology may illuminate the structural bounds on market efficiency.