Do Buy-Side Analysts Out-Perform the Sell-Side?

Abstract

We examine the performance of buy-side analysts relative to that of the sell-side. Our tests show that buy-side analysts at a large investment firm make less optimistic stock recommendations than sell-side analysts, consistent with their facing fewer conflicts of interest. However, their earnings forecasts are relatively optimistic and inaccurate and returns to their buy recommendations under-perform sell-side recommendations. Large sample tests that compare the performance of sell-side analyst recommendations and portfolio managers who rely exclusively on buy-side research confirm the sell-side's superiority. These performance differences appear to be partially explained by the buy-side's higher retention of poor-performing analysts and by differences in performance benchmarks used to evaluate buy- and sell-side analysts.