279 /342
臺大管理論叢
第
27
卷第
2S
期
279
4.3 Robust Check
For robustness, we use level value with one-year lag instead of changes in all
independent variables and control variables when dependent variable is change in
reinsurance demand. All results for the variables of interest including CEO turnover, routine
CEO turnover, non-routine CEO turnover, forced CEO turnover, and voluntary CEO
turnover are qualitatively similar (the results are not tabulated). We also obtain similar results
on interest variables when using the original value for total reinsurance ratio, reinsurance
ratio from affiliated reinsurers, and reinsurance ratio from non-affiliated reinsurers rather
than change value as dependent variables and original value for all independent variables and
control variables with one-year lag (the results are not tabulated). In addition, we consider
the products portfolios
32
(e.g., percentage of commercial lines) in this paper. We next separate
the percentage of long-tail lines and percentage of short-tail lines (reference variable) into
four categories (e.g., Choi, Park, and Ho, 2013): percentage of commercial long-tail lines,
percentage of commercial short-tail lines, percentage of personal long-tail lines and
percentage of personal short-tail lines (reference variable). The evidence (not tabulated)
shows percentage of commercial long-tail lines is positively and significantly related to
changes in total reinsurance ratio and reinsurance ratio from affiliated reinsurers. More
important, the results of variables of interest controlling products portfolios are similar to
those in Table 3.
5. Conclusion
The paper investigates the impact of CEO turnover on the demand for total reinsurance,
demand for reinsurance from affiliated reinsurers and demand for reinsurance from non-
affiliated reinsurers in the U.S. property casualty insurance industry from 2000 through
2010. Our evidence shows that insurers with CEO turnover are more likely to increase
reinsurance demand than insurers without CEO turnover after CEO turnover. Specifically, an
insurer with new CEO is more likely to have a more conservative strategy and thus increase
demand for reinsurance because new CEO does not have much track records with the board
and trust from the board.
More detailed analyses indicate that insurers with non-routine (forced) CEO turnover
are more likely to increase reinsurance than insurers without CEO turnover, but insurers with
routine (voluntary) CEO turnover are not likely to change reinsurance policies after CEO
32 We thank a reviewer for this valuable comment.