Detecting Information Pooling: Evidence from Earnings Forecasts after Brokerage Mergers

Abstract Forecast improvements can be expected if the two partners involved in a brokerage merger pool information and expertise. We examine four large mergers of brokerage firms in the last decade to study the incidence of and explanations for forecast improvements after the mergers. At the brokerage-level, we find that for two of the four mergers, forecast improvements appear more pronounced in subsamples of stocks for which both of the pre-merger analysts were retained in the merged brokerage. At the analyst-level, we find only weak evidence of forecast improvements after the merger. However, we find evidence that after a merger, a stock is more likely to be assigned to an analyst with overall better forecasting performance before the merger. This suggests that analyst selection can be a mechanism generating the post-merger forecasting improvements.