Good-Bye I/B/E/S (or Not?)
研究了I/B/E/S数据库在2018年声称对分析师ID匿名化后的实际影响，发现主要影响非美国公司预测，且匿名化早于声称时间，同时提供了逆向还原匿名ID的方法。
ABSTRACT In October 2018, I/B/E/S claimed to have anonymized the identifications (IDs) of 88 brokers and their analysts in its Detail History Files. This paper examines to what extent the change affects the post-2018 Detail History File. The main findings are as follows. First, there is no evidence of a large-scale ID anonymization for the annual and quarterly EPS forecasts for U.S. firms. Second, the anonymization predominantly affects the forecasts of non-U.S. firms. A significant percentage of forecasts for non-U.S. firms have been anonymized since January 2017 (at least one year earlier than what was claimed). Third, there is no evidence of a large-scale ID reshuffling of forecasts between the 2015 and 2021 Detail History Files. Last, because the anonymization has very little effect on the Recommendations Detail File, I am able to offer detailed steps to reverse engineer the anonymized analyst IDs in the Unadjusted Detail History File. JEL Classifications: G00; M41.