Political Turnover, Dynamic Separation of Power and Efficiency

This paper contributes to a large literature studying the interaction between institution and fiscal outcomes. We provide two general contributions. First, we argue that we should not take it for granted that there exist institutions that work well in all environments: specific institutions may increase spending in certain countries, but not in others. Second, we find that countries characterized by the same parameters and by the same institutions may end up with different spending levels. This suggests that in addition to institutions, multiplicity may help explain the wide variation of spending levels across countries. Model and Results: In this paper, policy makers have a time-inconsistent temptation to increase current spending. We study a legislature where legislators sequentially negotiate over spending and solve for the Markov-perfect equilibria of the dynamic game. Our goal is to investigate which institutions help reduce inefficient spending. The model generates non-trivial strategic interactions between current and future legislators. First, there is strategic complementarity: the perspective that more future policymakers will limit spending provide stronger incentives to be fiscally responsible in the current period. Complementarity leads to multiple equilibria, which are ranked according to total spending. Second, in the model there is also a region of parameters where strategic substitutability is observed: when the vast majority of future policymakers are expected to favor low spending, it is less costly to procrastinate a spending cut. As a result of these strategic interactions, we argue that changes of the political environment may have different effects in countries characterized by different parameters. This is because the type (and strength) of strategic interactions at work in each country may be different. For instance, we obtain that an increase of political turnover reduces spending in countries where the average legislators' spending bias is sufficiently moderate, but increases it in countries where the average bias is severe. Similarly, we find that electoral laws that reduce the number of extreme-party legislators also have opposite consequences on spending levels depending on which country they are applied to. Finally, we investigate the consequences of stronger majority requirements.