251 /342
臺大管理論叢
第
27
卷第
2S
期
251
incentives to reduce risk resulting in purchasing more reinsurance based on managerial
professional quality than routine new CEOs.
On the one hand, insurers with non-routine CEO turnover tend to purchase more
reinsurance than insurers with routine CEO turnover because new CEOs resulting from
routine turnover are from the board and they are familiar with the direction of the board. In
addition, if the new CEO is promoted through routine turnover, she/he should be better
trusted than non-routine CEO. Routine-turnover CEOs have worked with other directors in
the past. If routine-turnover CEOs maintain the current reinsurance policy and unexpected
huge losses occur, the board members are more likely to attribute the losses as unexpected
losses rather than poor reinsurance decisions. New non-routine CEOs may want to purchase
more reinsurance to transfer their firms’ risk. When huge losses occur, non-routine turnover
CEOs are more likely to be blamed for insufficient reinsurance because they have yet to
establish their credibility with the board.
On the other hand, insurers with non-routine CEO turnover are more likely to take more
risk and thus purchase less reinsurance. The reason is that non-routine CEO turnover are
generally involved with certain negative events of insurers (e.g., termination of old CEOs
because of poor performance). Goel and Thakor (2008) suggest that the board’s decision to
retain or fire a CEO is based on his/her observed performance. The board is more likely to
tolerate new CEOs appointed as a result of non-routine CEO turnover that failed a high risk
project than CEOs who have been with the insurers for a while. The two conflicting
arguments above lead to the following null hypothesis:
Hypothesis 2: The reinsurance decision of insurers with routine or non-routine CEO
turnover is not different from that of insurers without CEO turnover
after CEO turnover.
2.3 Forced CEO Turnover, Voluntary CEO Turnover and Reinsurance Demand
Voluntary CEO turnover can emerge from normal CEO turnover (i.e., health,
retirement, or death). However, no evidence of a significant relation between voluntary CEO
turnover and prior corporate activities or performance has been obtained (e.g., Huson et al.,
2001; Huson, Malatesta, and Parrino, 2004). In contrast, forced CEO turnovers may be
associated with poor prior performance (e.g., DeFond and Park, 1999). Čihák et al. (2009)
find that a significantly positive relation between a forced executive (i.e., president,
chairperson, CEO and COO) turnover and higher default risk, because the executive’s
exposure to be forced out job risk. Huson et al. (2004) find that top management turnover