|Abstract: ||Recent electoral reforms in New Zealand and two devolved sub‐national parliaments in the U.K.,
coupled with the widespread fragmentation of party systems in Western democracies and shifts toward
greater electoral volatility, give rise to the need to understand the effects of electoral systems on the
incentives leading parties to cooperate (or not) with one another, and the results for government
survival, especially under the circumstance of “hung parliaments,” where no party holds a legislative
majority. This study is especially germane in the Canadian context, where an extended period of
minority governance has renewed the debate about whether to reform the country’s electoral system to
make it more proportional.
The two main approaches in the extant literature on cabinet termination – the empirical “events
history” strand and the rational‐choice theoretical models – remain to be fully unified, hampering the
proper evaluation of hypothesized causal mechanisms linking public opinion shocks – and the systematic effects electoral systems have on them – to government survival outcomes.
This paper looks at the recent evolution of the literature on government duration, focusing
specifically on determining whether electoral systems alter the incentives leading to government termination in hung parliaments sufficiently to reduce the length of government survival. The literature
to date remains insufficiently specified to draw strong conclusions about the effect of electoral systems on the stability of individual minority or coalition governments. Thus, I propose reworking some assumptions in theoretical models of government termination dealing with public opinion shocks. I also recommend that a large‐N statistical analysis of termination hazard rates of the type used by Lanny Martin be carried out using electoral system as the explanatory variable but limiting the sample to minority and coalition governments.|