Policy Choices in Assembly versus Representative Democracy : Evidence from Swiss Communes

This paper investigates whether the form of the legislative institution - assembly versus parliament - affects the level and composition of local public expenditure. We collect data at the commune level in Switzerland over the period 1945-2010 and use two research designs: fixed-effects and regression discontinuity (RD) based on local population. Analyzing communes that switched the form of their legislative institution over time, we find that introducing a parliament leads to a 12 percent increase in both general administration and education spending per capita and an increase in total spending and revenue of about 6 percent. In contrast, regression discontinuity estimates cannot be distinguished from zero for any spending category or overall. These contrasting results highlight the local nature of discontinuity estimates since population is an order of magnitude larger in our switcher sample compared to the RD sample. To understand the mechanism at play, we run a survey among assembly participants and document a sizeable under-representation of 20- to 40-year-olds as well as of women in town meetings compared to both the electorate and to voters in elections. Switching from assembly democracy to parliament thus increases the representation of two demographics that are known for their relatively high preference for education spending.