Number of mergers between institutional blockholders.

This figure plots the number of effective mergers between institutional blockholders in each year during 1995–2012. The merger information and blockholders’ holding information come from SDC database and Thomson Reuters Institutional (13F) Holdings database, respectively. According to [15], mergers should satisfy the following criteria. First, both acquirer and target in the merger are located in the U.S. and should be listed in Thomson Reuters Institutional (13F) Holdings as institutional blockholders that hold more than 5% of shares in at least one firm. Second, Thomson Reuters Institutional (13F) Holdings database stops recording the fillings of the target institutional blockholders in the same year as that of the merger’s announcement date. Third, the completion period of mergers should not exceed one year. Fourth, both the target and acquirer hold firms in the same industry at one quarter before the merger. After that, we link firms affected by these mergers to financial data from COMPUSTAT. In order to make the analysis comparable to related works on CSR, firms in financial (SIC code: 6000–6999) and utility (SIC code: 4900–4999) industries are excluded. Observations with missing data are also excluded. If no firms are from a merger case because they are from financial and utility industries or with missing data, the merger will be dropped out of the merger cases. Finally, 36 effective merger cases are included in this quasi-natural experiment.