We identify a new channel – market makers' attention constraints – through which earnings announcements for one stock affect the liquidity of other stocks. When some stocks handled by a designated market maker have earnings announcements, liquidity is lower for non-announcement stocks handled by the same market maker, with the largest effects coming from earnings surprises and stocks with high earnings response coefficients. Half of the liquidity decline reflects attention constraints binding on the individual market maker, and the other half is explained by the market maker's inventory. We further find that a market design change that increases automation alleviates the liquidity effect of attention constraints, despite an increase in the number of stocks allocated to each market maker.

The trading of shares of the same firm in multiple markets has become common over the last 30 years, but there is little empirical evidence on the extent to which investors actively exploit multimarket environments. We introduce a volume-based measure of multimarket trading to address this question. Analyzing a large set of cross-listed firms, we find higher multimarket trading among markets with similar designs and strong enforcement of insider trading laws and for firms with higher institutional ownership. These findings are important for firms evaluating the benefits of cross listing and for markets competing for order flow.

We document strong weekly lead-lag return predictability across stocks from different industries with no customer-supplier linkages (economically unrelated stocks). Between 1980 and 2010, the industry-neutral long-short hedge portfolio earns an average of over 19 basis points per week. This predictability is related to common institutional ownership and is distinct from previously documented lead-lag effects. Common institutional ownership is a complementary rather than a substitute explanation for return predictability. Information linkages are enough to induce return predictability among stocks in the same industry, but economically unrelated stocks exhibit return predictability only when they have common institutional owners. Our findings suggest that institutional portfolio reallocations can induce return predictability among otherwise unrelated stocks.

We use a proprietary data set to test the implications of several asymmetric information models on how short-lived private information affects trading strategies and liquidity provision. Our identification rests on information acquisition before analyst recommendations are publicly announced. We provide the first empirical evidence supporting theoretical predictions that early-informed traders “sell the news” after “buying the rumor.” Further, we find distinct profit-taking patterns across different classes of institutions. Uninformed institutions, but not individuals, emerge as de facto liquidity providers to better-informed institutions. Placebo tests confirm that these trading patterns are unique to situations in which some investors have a short-lived informational advantage.

Using a database of daily institutional trades, we document that a majority of short-term institutional trades lose money. In aggregate, over 23% of round-trip trades are held for less than 3 months, and the returns on these trades average -3.91% (nonannualized). These losses are pervasive across all types of stocks, with the lowest returns occurring in small stocks, value stocks, and low-momentum stocks. Short-term trades lose more in more volatile markets. Across funds, the worst short-term returns accrue to funds that do the most trading, and there is no evidence of persistent skill or disposition effect in short-term institutional trades.