Information, Analysts, And Stock Return Comovement
National University of Singapore (NUS) – Department of Finance
University of Alberta – Department of Finance and Statistical Analysis; National Bureau of Economic Research (NBER)
UNSW Australia Business School, School of Banking and Finance
National University of Singapore – Business School
July 30, 2015
Analysts follow disproportionally firms whose fundamentals correlate more with those of their industry peers. This coverage pattern supports models of profit-maximizing information intermediaries producing preferentially information valuable in pricing more stocks. We designate highly followed firms whose fundamentals best predict those of peer firms as bellwether firms. When analysts revise a bellwether firm’s earning forecast, it changes the prices of other firms significantly; however, revisions for firms that are less intensely followed do not change the prices of heavily followed firms. Unidirectional information spillovers explain how the more accurately priced stocks might exhibit more comovement.
Information, Analysts, And Stock Return Comovement – Introduction
Stocks followed by more analysts appear to be priced more accurately (Brennan, Jegadeesh, and Swaminathan 1993; Walther 1997), yet their returns are also more prone to comove (Piotroski and Roulstone 2004; Chan and Hameed 2006), which seems anomalous because a firm’s stock price moves idiosyncratically as it incorporates new firm-specific information (French and Roll 1986; Roll 1988). Moreover, higher firm-specific return volatility (i.e., lower comovement) is linked to more intense information incorporation into stock prices (e.g., Morck, Yeung, and Yu 2000, 2013; Wurgler 2000; Durnev et al. 2003, 2004; Jin and Myers 2006).
This paper resolves the seeming anomaly by showing that stocks that are covered widely by analysts exhibit more comovement precisely because they are priced more accurately and, therefore, provide signals with which to update the prices of more opaque stocks. Thus, higher return comovement associated with more analysts following need not imply stock pricing that is less informative. Rather, it reflects spillovers as granular firm-specific information changes the prices of related stocks. Specifically, information about well-covered firms affects prices of thinly covered firms (contemporaneously and with a lag), but no opposite effect is evident.
To examine the analysts’ role as information intermediaries, we derive testable propositions consistent with recent models of information intermediaries (Veldkamp 2006a). First, because information is a nonrival good, profit-maximizing analysts, incurring a fixed cost of information production, produce the information that fetches the highest price from investors. Rationally, investors value information useful for predicting many stocks more highly than information useful for predicting only one stock. Analysts therefore ought to disproportionately follow stocks whose fundamentals correlate more with those of other firms. Following financial practitioners, we dub these “bellwether firms.” Second, new information about a bellwether firm should commensurately change the stock prices of other firms whose fundamentals correlate highly with those of the bellwether firm because investors use it to infer changes in their fundamentals too. This spillover should be stronger to stocks that are less followed.
Consistent with the first proposition, we find more analysts following stocks whose fundamentals are more correlated with the fundamentals of many other firms. Our findings hold after controlling for other firm characteristics found to attract analysts following, including market capitalizations, trading volumes, volatility, and institutional ownership (Bhushan 1989; Brennan and Hughes 1991; Alford and Berger 1999).
Consistent with the second proposition, we find strong evidence of information spillovers from high-analyst firms to other fundamentally related firms. From the highest tertile of analyst coverage in each industry each year, we select the firm whose fundamentals correlate most strongly with those of all other firms in the industry, and label these “industry bellwether firms.” When analysts revise their earnings forecasts for bellwether firms, we observe significant effects on the current and future stock prices of their industry peers. Moreover, this effect is higher for peers with lower analyst coverage. Importantly, these information spillovers are unidirectional: earnings forecast revisions for firms that are less intensely followed do not predict the prices of heavily followed firms. Our estimates show an analyst’s revision of a bellwether firm’s forecasted earnings having significant cross-firm spillover for up to 1 month (in event time).
These findings complement those in Kelly and Ljungqvist (2012), who show elevated asymmetric information in the prices of other firms with correlated fundamentals following a firm’s coverage terminations, as implied by Admati (1985) and Veldkamp (2006a). Following Kelly and Ljungqvist (2012) in taking as exogenous the analyst coverage terminations because of brokerage firms’ closures of research departments, we show further that coverage terminations in a firm appear to cause its investors to rely more heavily on the information about bellwether firms in the industry. Specifically, we again find a one-way spillover from bellwether firms to industry peers when coverage exogenously declines for the thinly covered firms.
See full PDF below.