Abstract: Financial data vendors intermediate the flow of information from firms to investors. I study frictions that arise in the context of this intermediation by focusing on one of the most prominent data vendors in the finance industry -- Standard & Poor’s ('S&P') Compustat database. Compustat provides subscribers with decades of 10-K and 10-Q data; however, it does not cover every public firm in every period. I show that institutional investment is over 36% below its unconditional mean for firms not covered in Compustat. A quasi-natural experiment confirms a plausibly causal connection between Compustat's data coverage and investor demand: a technology shock at S&P in the 1990s causes a discrete reduction in missing data. This change in data coverage is followed by a significant increase in institutional investment for treated firms relative to control firms. I then show that missing Compustat data is associated with lower informational efficiency of equity prices. These results highlight the role that data vendors play in facilitating the flow of information within financial markets.
Dmitriy Muravyev, University of Illinois-Urbana-Champaign
Abstract: The SEC’s EDGAR introduction slashed the costs of acquiring and trading on accounting information, especially for smaller investors. We both causally identify and assess how these information costs affect stock anomalies. Using the staggered EDGAR introduction, we show that average alphas for 125 accounting anomalies decline substantially and that the decline explains most of the pre-EDGAR alphas. By contrast, alphas for 80 non-accounting anomalies do not change significantly. Information costs are as substantial as the other limits to arbitrage.
Abstract: We use abnormal undercutting activity (QIDRes) to measure informed trading risk, reflecting liquidity-providing algorithms competing less to fill marketable orders when adverse selection exposure rises. Despite its simple construction, when examined around information events, QIDRes behaves similarly to existing measures of informed trading intensity/probability whose constructions are complex. QIDRes predicts arrivals and magnitudes of imminent information events. Moreover, episodes of high QIDRes coincide with weaker subsequent price reversals, increased accumulation/covering of short interest, and increased informed institutional trades. QIDRes from prior quarters positively predicts monthly stock returns, especially among stocks with tighter short sale constraints. Since QIDRes is orthogonal to stock liquidity and is not a persistent stock characteristic, we attribute its return predictability to limits to arbitrage.
Discussant: Charles Martineau, University of Toronto
