Institutional Trading and Corporate Earnings and Returns
1460 STANFORD UNIVERSITY WORKING PAPER SERIES
Posted: 18 May 1998
Date Written: October 1997
This paper provides evidence on the association between quarterly changes in institutional holdings and firm performance. Specifically, this study examines the association between institutional trades and firm performance, to better understand institutional behavior and its potential effect on share prices and corporate management.
Our analysis contributes to the literature in three ways. Most importantly, we examine the relation between changes in institutional holdings and earnings, whereas previous research has focused on returns. The link between earnings and changes in institutional holdings is important because an association between institutional trade and returns does not necessarily provide an incentive to manage earnings. Prior research suggests, for example, that the market ?sees through?; the expensing of research and development and does not penalize market price for the reduction in earnings. Our results suggest earnings are significant in explaining changes in institutional holdings, and that earnings are incrementally significant after controlling for returns. We also provide evidence of a positive association between changes in institutional holdings and lagged values of returns and earnings after controlling for the concurrent relations, suggesting that the relation between changes in holdings and performance is not limited to current performance. These findings suggest that institutional trade is responsive to earnings, and that such trade may provide management with additional incentives to focus on earnings results.
Second, we consider the trading behavior of a different and broader set of institutional investors, which includes banks, colleges and universities, insurance companies, investment advisors, private foundations and private and public pensions. Jacobs , O'Barr and Conley  and Black , among others, suggest that the trading behavior of pension funds differs from other institutions because of the relatively longer-term horizon of the pension beneficiaries. Differences in trading patterns between pensions and other institutions may also stem from tax incentives for trading on returns performance which is limited for pension funds relative to those for other types of institutions. Our findings indicate that the relation between trading and returns for institutions as a whole is positive, stronger and more pervasive than that documented by Lakonishok, Shleifer, Thaler and Vishny (1991) and Lakonishok, Shleifer and Vishny (1992) for their sample of externally-managed pension funds.
Third, we provide evidence of differences in trading behavior across the major categories of institutions (banks, colleges and universities, insurance companies, investment advisers, investment companies, private foundations, and private and public pensions). In addition to examining each class of institutions individually, we classify institutions into two primary groups: (1) those which we argue have potentially stronger incentives to trade on performance (banks, insurance companies, investment advisers and investment companies) and (2) those with potentially weaker incentives (colleges and universities, private foundation and private and public pension funds). Our results suggest that there are important differences in trading behavior across types of institutions. There is a significant positive relation between returns and changes in holdings for institutions with potentially stronger incentives to trade on performance and a significant negative relation for institutions with weaker incentives. The correlation between changes in holdings and earnings, on the other hand, is generally positive for both groups of institutions.
JEL Classification: M41, M43, G29
Suggested Citation: Suggested Citation