Institutional Ownership and Conservatism

In our paper, Institutional Ownership and Conservatism, forthcoming in the Journal of Accounting and Economics as published by Elsevier, we examine the relation between institutional ownership and accounting conservatism. Ball (2001) and Watts (2003) propose that equity investors are an important source of demand for conservatism as a governance device. Recent empirical evidence supports this proposition. Consistent with equity investors creating demand for conservatism, LaFond and Roychowdhury (2008) show that conservatism is greater when the separation of ownership and control is more pronounced, and LaFond and Watts (2008) find that higher information asymmetry between managers and shareholders leads to more conservative reporting. These findings raise an important question: Which equity investors demand conservatism?

A large body of research documents that individual investors are generally small unsophisticated investors who trade primarily for reasons unrelated to information (e.g., liquidity or rank speculation). Accordingly, individuals are unlikely to be sophisticated enough to gauge whether firms consistently use conservative financial reporting. In contrast, institutional investors are both more sophisticated and more important price setters in capital markets. Thus, if conservative financial reporting provides governance benefits, institutional investors are more likely to understand and value such benefits, and as a result, demand conservative accounting from managers. On the other hand, as institutional investors likely have privileged access to management and inside information (Carleton et al. 1998), they may rely more on direct monitoring and less on monitoring through accounting numbers. Thus, whether institutions demand conservative financial reports is an empirical question.

Among institutions, institutional characteristics such as investment horizons, concentration of share holdings, and independence from firm management induce higher monitoring incentives among some institutions relative to others. Further, within the set of monitoring institutions, the investee firm characteristics that influence the extent of direct monitoring possible are likely to drive the demand for conservatism as part of institutions’ monitoring efforts. More specifically, prior research suggests that institutions that have long investment horizons, concentrated share holdings, and independence from management (hereafter, monitoring institutions) are most likely to monitor managers. Institutional investors’ demand for conservatism is thus more likely to emanate from monitoring institutions. In our first set of tests we examine whether monitoring institutions’ ownership is positively associated with conservatism.

Within the set of monitoring institutions, direct monitoring is more difficult in investee firms with more growth options (Smith and Watts 1992) and higher information asymmetry (Prendergast 2002). Yet, for such firms, the potential governance benefits of conservatism in disciplining managers’ investment decisions (Ball 2001) and reducing managers’ ability to overstate earnings (Watts 2003) are likely to be greater. Thus, in our second set of tests we examine whether the association between monitoring institutions’ ownership and conservatism is more positive among firms with more growth options and higher information asymmetry.

We define monitoring institutions as those that are both (1) dedicated institutions (i.e., institutions with long investment horizons and concentrated holdings), following Bushee (2001), and (2) independent from management, following Brickley et al. (1988). We use Basu’s (1997) earnings-return model to gauge conservatism. To mitigate concern over the endogeneity of institutional ownership, we perform our analyses using a measure of residual ownership, defined as the residual from a regression of ownership on its economic determinants. Based on a sample of 16,911 firm-years over 1995-2006, we find that higher residual ownership by monitoring institutions is associated with greater conservatism in firms’ financial reporting, and that this positive association is more pronounced among firms with more growth options and higher information asymmetry. These results are consistent with monitoring institutions demanding conservatism, especially when the costs of direct monitoring are higher and the potential governance benefits of conservatism are greater. Our results are robust to using an alternative measure of conservatism that does not rely on stock returns to measure economic news.

While the above results are consistent with monitoring institutions demanding conservatism, it could also be the case that firms with more conservative financial reporting attract investment by monitoring institutions (“reverse causality” explanation). We conduct two tests to examine the direction of causality. First, we examine the relation between conservatism and monitoring institutions’ lagged, current, and lead residual ownership. We find that conservatism is positively related to the lagged residual ownership, but unrelated to the current or lead residual ownership. Second, following LaFond and Watts (2008), we sort firms into quintiles based on the change in monitoring institutions’ residual ownership in year t, and then for each quintile we estimate the level of conservatism in years t-2, t-1, t and t+1. This approach allows us to examine the change in conservatism in the pre-ownership-change period (from t-2 to t-1), during ownership-change period (from t-1 to t), and post-ownership-change period (from t to t+1). We find that a large increase in ownership by monitoring institutions is associated with an increase in future (but not past or concurrent) conservatism. These results are consistent with monitoring institutions demanding conservatism, but not conservative reporting attracting investment by monitoring institutions. However, to the extent that we cannot perfectly correct for reverse causality, our results could be affected.

Overall, our results suggest that monitoring institutions demand conservatism in investee firms’ reporting practices, and that this demand is more pronounced when both direct monitoring is more difficult and conservatism provides more governance benefits. This study contributes to our understanding of the nature of the economic forces that generate demand for conservatism (Watts 2003). In particular, building on prior studies (LaFond and Roychowdhury 2008; LaFond and Watts 2008), we provide direct evidence suggesting that monitoring institutions are an important class of investors that demands conservatism as a governance device.

The full paper is available for download here.