The Effectiveness of Institutional Investors in Evaluating Analysts

The following post comes to us from Lily Fang of the Department of Finance at INSEAD and Ayako Yasuda of the Graduate School of Management at UC Davis.

In the paper, The Effectiveness of Institutional Investors in Evaluating Analysts, which was recently made publicly available on SSRN, we examine the effectiveness of institutional investors in evaluating analysts by comparing the performance of recommendations made by AAs—star analysts elected by institutional investors—with those made by other analysts.

We ask four related questions. First, does the AA status at least partially reflect analyst skill? That is, if AAs make more valuable recommendations, is it because of skill, or other factors such as luck, market influence, or access to management? If investors are effective in evaluating analysts, we expect the AA status to be indicative of skill. Second and closely related to the first question, does the AA status contain information about the analyst that is not entirely captured by other observable characteristics? The answer should be affirmative if institutional investors uncover unique information about analysts. Third, can institutional investors and the AA election process adapt to major changes brought to the industry by regulations and labor market movements? And finally, who benefits the most from the elected star analysts’ views?

We provide evidence that institutional investors are effective in evaluating analysts. First, the AA status at least partially reflects analyst skill. We find that AAs make significantly more valuable (in the order of 7% per year) recommendations than non-AAs, in both buys and sells. This difference exists both before and after the AAs are elected, is not explained by announcement effects, and is not significantly eroded by Reg-FD. Second, the AA status also predicts buy recommendation performance above and beyond other analyst characteristics. Third, institutional investors are able to respond to major shifts in the market by making promotions to and demotions from the AA ranks in a way that preserved the group’s overall performance. Finally, consistent with costly information production and the Grossman and Stiglitz (1980) notion of information efficiency, we find that most of the gains from identifying top analysts accrue to institutional investors who vote for star analysts and who also have advance access to analyst views.

Whether institutional investors are effective in evaluating analysts has important implications for the governance of sell-side research, an issue that regulators frequently grapple with. The difficulty is rooted in the fact that analysts are in the market of opinion provision. While research insight is a valuable public good, the cost of its production has to be borne privately. The funding of research in the sell-side through either banking or trading revenues creates conflicts of interest: What prevents the analysts from lying to investors in favor of either issuer firms (if research is funded through investment banking revenues) or simply trade-generation (if research is funded by trading commissions)? Fang and Yasuda (2009) show that one important enforcement mechanism is the reputational concern of sell-side analysts vis-à-vis institutional investors. For this mechanism to work, institutional investors must be effective in evaluating analysts in the first place. If institutional investors are effective in their evaluation of analysts, then ultimately it is them that play the key role of reinforcing the reputation mechanism.

The full post is available for download here.

Both comments and trackbacks are currently closed.