آیا بانک های سرمایه گذاری به تحلیلگران خود گوش می دهند ؟

To what extent conflicts of interest affect the investment value of sell-side analyst research is an ongoing debate. We approach this issue from a new direction by investigating how asset-management divisions of investment banks use stock recommendations issued by their own analysts. Based on holdings changes around initiations, upgrades, and downgrades from 1993 to 2003, we find that these bank-affiliated investors follow recommendations from sell-side analysts in general, increasing (decreasing) their relative holdings following positive (negative) recommendations. More importantly, these investors respond more strongly to recommendations issued by their own analysts than to those issued by analysts affiliated with other banks, especially for recommendations on small and low-analyst-coverage firms. Thus, we find that investment banks “eat their own cooking,” showing that these presumably sophisticated institutional investors view sell-side recommendations as having investment value, particularly when the recommendations come from their own analysts.

To what extent conflicts of interest affect the investment value of sell-side analyst research is an ongoing debate. Numerous studies suggest that sell-side analysts issue overly optimistic recommendations to curry favor with covered firms and attempt to win investment banking business for their employers. Such behavior, if it is pervasive, may significantly reduce the investment value of sell-side research. However, several recent studies show that recommendations provided by analysts with potential conflicts of interest do not underperform recommendations provided by other analysts, casting doubt on the importance of such conflicts.1 We approach this issue from a new direction. We examine whether institutional investors affiliated with investment banks use stock recommendations issued by their own sell-side analysts. Suppose we discover that an investment bank’s money managers disregard (or worse, trade against) the recommendations of the bank’s sell-side analysts. Such behavior would be clear evidence that the bank’s sell-side recommendations are viewed as not useful (at best) by institutional investors who are particularly well-positioned to judge. Our primary findings are that money managers affiliated with investment banks use sell-side research in general, and they particularly listen to their own bank’s analysts. Specifically, we study 1.1 million holdings changes for institutional investors affiliated with 58 major US investment banks around 70,000 stock recommendations from 1993 to 2003. We find that these investors increase their holdings after positive initiations and upgrades, and they decrease their holdings following downgrades. More importantly, the changes in holdings are economically and statistically larger when the recommendation is from an affiliated analyst.2 Further tests show that our results are more pronounced for recommendations on small and low-analyst-coverage stocks. To our knowledge, we are the first to explore how institutional investors respond to recommendations issued by affiliated and unaffiliated analysts.3 Our results show that sell-side research is used by institutional investors. This fact suggests that relatively knowledgeable market participants view sell-side research as valuable, despite the existence of potential conflicts of interest. Our results are robust to alternative explanations. For example, we consider the possibility that analyst recommendations and institutional investor holdings changes are joint responses to public events, but we find no evidence to support this. We also explore the hypothesis that affiliated money managers “cherry pick” recommendations based on private information. We find no support for this story either; there is no differential performance between recommendations that are followed and those that are not followed. The remainder of the paper is organized as follows. We discuss the related literature in Section 2 and sample selection and data in Section 3. Section 4 provides our results. We conclude the paper in Section 5.

We investigate how institutional investors affiliated with investment banks use stock recommendations from sell-side ana- lysts that are affiliated with their own banks. We find that sell-side recommendations appear to be used by institutional investors in general, though we cannot completely rule out the possibility that analysts and investors are reacting to the same information. How- ever, and more importantly, we find that institutional investors re- spond more strongly to recommendations issued by analysts affiliated with their own investment banks, particularly for initia- tions and upgrades. Thus, we find that investment banks do take their own advice. Our results are more pronounced among small and low-analyst-coverage firms. We examine whether our results are driven by affiliated institu- tional investors’ superior knowledge about the bias and invest- ment value of recommendations issued by analysts affiliated with their own banks. We find that, on average, analyst stock rec- ommendations have investment value for upgrades, but not for ini- tiations or downgrades in the long run. However, regardless of recommendation type, we find no significant difference in returns between recommended stocks followed and not followed by affil- iated institutional investors, so we see no evidence of ‘‘cherry pick-ing’’ by affiliated investors. We test whether analyst recommendations and institutional holdings changes are driven by publicly-available information events by restricting our sample to quarters in which only one ana- lyst issues a new recommendation for a stock. For this sample, it is less likely that analyst recommendations are a response to a spe- cific event, and we find that our results hold in this subsample, meaning that institutional investors appear to listen to sell-side analysts, and they particularly listen to the analysts affiliated with their banks. Our results contribute to the debate on the extent to which con- flicts of interest affect the investment value of sell-side analyst re- search. Our findings indicate that there is no systematic evidence that analysts knowingly issue overly optimistic recommendations despite their own negative opinions on these stocks. Our results also show that the potential conflicts of interest do not invalidate the investment value of sell-side analyst stock recommendations in the eyes of institutional investors.