Cabinet Survival And Competing Risk Essay, Research Paper
Cabinet Survival and Competing Risk
Daniel Diermeir and Randy Stevenson attempt to solve a recent controversy in the study of cabinet termination pertaining to the shape of hazard rates. There were two analysis one by Warwick which provided evidence that cabinets are more likely to terminate the longer they are in office. Alt and King did the other analysis, which suggest that hazard rates are consistent over the lifetime of a cabinet. King’s approach used Brown’s constant hazard rate, but allowed it to depend on a set of time invariant covariates which was suggested by Strom. In other words Strom suggest a bargaining model to explain the cabinet terminations. Strom’s analysis suggests that rather than focusing on cabinet survival, it is important to distinguish between cabinets that end in election and those that are replaced without elections. This issue is of some importance considering a constant hazard rate would support the model of the cabinet termination due to Brown, while an increasing hazard rate would favor Strom’s approach. What is in dispute however, is the substantive significance of the results. By using various equations and other literature they obtained some results and began comparing them. Specifically, King show that, for Warwick’s pooled sample, the slop of the hazard rate changes only by ten- percent mean value over the whole range. This means that while there is some evidence for increasing hazard rates in samples that does not distinguish between types of failure, all the evidence suggest is that these increases are small enough to be considered substantively flat. The only one mode where a cabinet terminated the case was that the cabinet was replaced without a new election. Hazard rates aren’t flat in the case of replacement, but increase over the life of the cabinet. They used analysis, which demonstrate the processes that correspond to each failure type are markedly different. Standard models of cabinet termination are best when replaced without a new election. They conclude that though their analysis merely focused on the replacement hazard where as Strom focused on sizes of dissolution hazards, their findings were however, that their concluded evidence supported Strom’s approach.