Table 4

Difference-in-differences estimation around CEO turnover events

Dependent variableLog(1+patents), t+2Log(1+citation adj.), t+2
(1)(2)(3)(4)(5)(6)
Treated×Post0.228**0.256**0.797**0.267**0.298**1.023**
(2.26)(2.56)(2.33)(2.07)(2.32)(2.36)
Post−0.053−0.204−0.800**−0.016−0.117−1.049**
(−0.30)(−1.35)(−2.37)(−0.08)(−0.69)(−2.47)
Observations9161,1624369161,162436
Adjusted R20.9480.9400.9410.9290.9220.917
Control variablesYesYesYesYesYesYes
Sample typeRawRawMatchedRawRawMatched
Sample window[−3,+3][−4,+4][−3,+3][−3,+3][−4,+4][−3,+3]
Year FEYesYesYesYesYesYes
Event FEYesYesYesYesYesYes

Note(s): The table presents the results of the OLS regressions for the sample of CEO turnovers, where the reason for CEO departure is unrelated to firm performance, such as CEO death, health issues or retirement (Gentry et al., 2021). Columns (1)–(2) and (4)–(5) present the estimation results using the raw sample, whereas Columns (3) and (6) show the estimation results using the matched sample from propensity score matching. Columns (1), (3), (4) and (6) use a 3-year event window around the turnover, whereas Columns (3) and (4) use a 4-year event window. The control variables (not shown) include the logarithm of CEO age, the logarithm of CEO tenure, firm size, the logarithm of firm age, capital expenditure, R&D expenditures and the logarithm of the number of firm employees. Event- and year-fixed effects are included in all columns. Standard errors are clustered at the firm level, with t-statistics presented in parentheses.*, ** and *** denote statistical significance at the 10%, 5% and 1% levels, respectively

or Create an Account

Close Modal
Close Modal