Seat-vote elasticity and the provincial distribution of government spending in Turkey

Abstract

The main question to be explored in this paper is, whether results of parliamentary elections have an impact on the provincial per-capita government spending. A firmly related question to which the literature has sought an answer and which also assists the effort to answer the question at the center of this study is how a government, interested in maximizing its electoral support, allocates government spending across administrative units (i.e. provinces for Turkey, states for the US). A possible answer to be given to this question relies in the seat-vote elasticities in provinces. The term stands for a coefficient that shows the relation between changes in percentage of seats in response to changes in the percentages of votes. It has been argued that electoral systems, with predominantly majoritarian characteristics, tend to display higher levels of seat-vote elasticity whereas systems mainly revealing proportional traits are characterized by lower levels of seat-vote elasticity. Turkish parliamentary electoral system is proportional election. Yet, levels of seat-vote elasticities may still be a determinant in the allocation of government spending across provinces in Turkey. With regards to the relation between seat-vote elasticity and per capita government spending, it is a widely held belief that the administrative units with a high level of seat-vote elasticity should benefit more from government spending, and those with low level of seat-vote elasticity should benefit less. The paper begins with the calculation of the seat-vote elasticities for 67 provinces for the six general elections held between 1990 and 2013. Then these seat-vote elasticity figures are used as explanatory variables in a fixed effects regression in which the dependent variable is the shares of per capita government spending across provinces. Results will be compared with the findings of the literature that conclude as highly elastic seat-vote relation demands more per capita government spending.