The aim of this work is to empirically test the validity of the median voter model for the Brazilian case. The test was made feasible by the possibility of reelection established in 1997 by the Reelection Amendment. With municipal information from the Demographic Census and from TSE for the year of 2000, we identified the bundle offered by candidates running for reelection as the expenditures realized during their first terms. Next, we estimated the demand for local public services. Relaxing the assumption of perfect information of candidates in relation to the median demand, and imposing a weaker hypothesis, of rational expectations, we justified the estimation of a measure of misperception of candidates for re-election in relation to median demand. Next, we estimated the impact of the module of this measure on the electoral performance of mayors incumbents using a selection model. The result suggests that the median voter model is valid for the case of Brazilian municipalities.
demand for public services; median voter; reelection; selection models
