Print
E-MailThis post comes from Ohad Kadan of Washington University in St. Louis, Leonardo Madureira of Case Western Reserve University, Rong Wang of Singapore Management University; and Tzachi Zach of Ohio State University.
In our paper Conflicts of Interest and Stock Recommendations: The Effects of the Global Settlement and Related Regulations, which was recently accepted for publication in the Review of Financial Studies, we investigate the impact of regulatory changes, including the Global Settlement, NASD Rule 2711 and the amended NYSE Rule 472, on analysts’ recommendations. We address three main questions. First, have analysts’ recommendations become more informative following the regulations? Second, did the regulations mitigate the effects of conflicts of interest? Lastly, did the regulations affect the response of investors to analysts’ recommendations?
Following the regulations, most leading investment banks moved from the traditional five-tier rating system to a coarser three-tier rating system over a short period of time (typically one day). The adoption of new rating systems was accompanied by banks completely reshuffling their recommendations, obtaining a more balanced distribution. We examine the informativeness of recommendations, as proxied by investors’ reactions, and how it was affected by the regulations. We start by examining conditional informativeness measured as the abnormal price reactions to recommendations, conditional on their type (optimistic, neutral, and pessimistic). The results suggest that investors internalized the change in the distribution of recommendations in the period following the regulations. For instance, the price response to optimistic recommendations is more positive in the Post-Reg period, suggesting that optimistic recommendations are perceived to be more reliable. By contrast, price responses to neutral and pessimistic recommendations are less negative following the regulations, since more recommendations fall under these categories. In additional tests, we find that the overall informativeness of recommendations has significantly decreased following the regulations: The absolute price reactions to stock recommendations are significantly lower in the Post-Reg period. We further show that recommendations issued by brokers who use a three-tier rating system (before or after the regulations) provide less information to investors. Additionally, the decline in informativeness after the regulations is stronger for banks that were sanctioned in the Global Settlement, all of whom have switched to a three-tier system.
We use a difference-in-differences approach to gauge the impact of the regulations on conflicts of interest between investment banking and research. Our main proxy for the presence of conflicts of interest is past underwriting relationship between the brokerage house and the recommended firm (affiliation). We document a significant change in how conflicts of interest influence stock recommendations. We corroborate prior research and the concerns of regulators by showing that conflicts of interest were associated with excess optimism in the Pre-Reg period. We show that in the Post-Reg period, affiliated analysts are as likely to issue optimistic recommendations as unaffiliated analysts. Moreover, the difference-in-differences between affiliated and unaffiliated analysts across the two periods is significant, suggesting that analysts have changed their recommendation practices. In contrast, conflicts of interest might still be influencing pessimistic recommendations. In both the Pre-Reg and Post-Reg periods, affiliated analysts are more reluctant to issue pessimistic recommendations than unaffiliated analysts, and the difference-in-differences is not significant. Alternative measures of conflicts of interest, also suggest significant changes in analysts’ practices. Before the regulations, analysts were overly optimistic regarding firms that have recently issued equity, and with respect to firms that experience financing deficit. We show that after the regulations, this optimism has declined significantly.
Finally, we examine whether investors react differently to recommendations issued by potentially conflicted analysts before and after the regulations. We find that investors discount affiliated neutral recommendations to a lesser extent after the regulations. However, we do not find such evidence for optimistic and pessimistic recommendations.
Collectively, we view our findings as consistent with a limited achievement of the regulations’ objectives. Although the mix of recommendations has become more balanced, the overall informativeness of recommendations has declined in the Post-Reg period.
The full paper is available here.
