Panel A: Performance measures
|
Return on assets (ROA)
|
Quarterly ROA (net income divided by book value of assets) over the four post-turnover quarters (Q1, Q2, Q3, and Q4) beginning on the first full quarter following the turnover announcement
|
Compustat
|
Change in ROA
|
The change in ROA from the pre-succession quarter over the period one, two, three, and four quarters following the CEO turnover
|
Compustat
|
Tobin’s Q
|
Quarterly market value of assets divided by book value of assets over the four post-turnover quarters (Q1, Q2, Q3, and Q4) beginning on the first full quarter following the turnover announcement
|
Compustat
|
Return on equity (ROE)
|
Quarterly ROE (net income divided by common equity) over the four post-turnover quarters (Q1, Q2, Q3, and Q4) beginning on the first full quarter following the turnover announcement
|
Compustat
|
Stock returns
|
Buy-and-hold returns for each quarter (Q1, Q2, Q3, and Q4) following the turnover announcement, beginning on the first month following the turnover announcement date
|
CRSP
|
Panel B: Control variables
|
Pre-succession ROA
|
ROA in the quarter immediately prior to the turnover announcement. Note that pre-succession Tobin’s Q and pre-succession ROE are calculated similarly using Tobin’s Q and ROE, respectively
|
Compustat
|
Pre-succession ROA change
|
The change in quarterly ROA during the 4 quarters prior to the turnover announcement. Note that pre-succession Tobin’s Q and pre-succession ROE are calculated similarly using Tobin’s Q and ROE, respectively
|
Compustat
|
Industry pre-succession ROA
|
The average quarterly industry ROA, as measured by 2-digit SIC code, for the four quarters prior to the turnover announcement
|
Compustat
|
Note that industry pre-succession Tobin’s Q and industry pre-succession ROE are calculated similarly using Tobin’s Q and ROE, respectively
|
Industry post-succession ROA
|
Quarterly industry ROA, as measured by 2-digit SIC code, in the post-succession period. Note that industry post-succession
|
Compustat
|
Tobin’s Q and industry post-succession ROE are calculated similarly using Tobin’s Q and ROE, respectively
|
Firm size
|
The log of book assets in the quarter prior to the turnover announcement
|
Compustat
|
CEO age
|
The log of predecessor age the year of the turnover announcement
|
Execucomp or hand collected
|
CEO tenure
|
The log time (measured in years and months) that the predecessor held the CEO post
|
Execucomp or hand collected
|
CEO ownership
|
The percentage ownership of the predecessor, as measured at the period prior to the turnover
|
Execucomp or hand collected
|
Inside CEO Succession
|
A dummy set equal to one when the appointed successor comes from within the firm
|
Execucomp or hand collected
|
Diversification
|
Jacquemin and Berry (1979) measure calculated for the year of the turnover. The ratio of sales in each industry-segment (measured by 4-digit SIC code) is calculated relative to total sales. This measure is then multiplied by the minus log of the same measure
|
Compustat
|
Leverage
|
Average quarterly leverage ratio (book debt scaled by the sum of book debt and market equity) over the four quarters prior to the turnover announcement
|
Compustat
|
Industry instability
|
We first calculate total industry sales (as measured by 4-digit SIC code) over the 3 year period prior to the turnover announcement. For each industry, we then regress the 3 year industry sales on time to obtain the standard error of slope. The industry instability is the standard error divided by average sales over this period
|
Compustat
|
CEO duality
|
A dummy variable set equal to one when the CEO was also the chairman of the board
|
Hand collected
|
Board ownership
|
The percentage ownership of all board members (excluding the CEO), as measured at the period prior to the turnover
|
Hand collected
|
Percentage of outside directors
|
The percentage of outsiders on the board of directors, as measured at the period prior to the turnover
|
Hand collected
|
Board size
|
The number of board members, as measured at the period prior to the turnover
|
Hand collected
|
No succession plan
|
A dummy variable set equal to one if the firm lacks a president/COO separate from the CEO
|
Hand collected
|
Interim succession
|
A dummy variable set equal to one if the firm appoints a temporary interim succession at the time of the turnover announcement
|
Hand collected
|
Died
|
A dummy variable set equal to one if the predecessor died while holding the CEO designation
|
Hand collected
|
Illegal
|
A dummy variable set equal to one if the predecessor departs due to illegal activities
|
Hand collected
|