Dynamic effects on government survival can come in a variety of forms. The form that first drew serious scholarly attention is the incidence of outside “events” – foreign crises, political assassinations, and the like – that occasionally challenge and topple governments. As the analyses reported in Chapter 5 demonstrate, developments of a more systematic sort, such as the changing state of the economy, also have a role to play in government survival. A third type of dynamic effect is autocorrelative: the potential influence of past government durations on present and future ones. Finally, there is the possibility that governments face a changing risk of termination the longer they remain in power, quite apart from any idiosyncratic events or systematic developments occurring in the outside environment. This change could take the form of a reduction in risk over time, indicating a process of consolidation of power; a rising risk over time, suggestive of an unravelling of the reins of power; or even some more complex, nonmonotonic pattern.

This chapter is concerned primarily with the last type of effect, about which almost nothing is known. The “events” assumption that the underlying rate of termination is constant throughout the lifetimes of governments represents the simplest treatment of the issue, but it is also the most optimistic: it assumes that there are not systematic trends apart from those measured by the covariates in the model.