DT-19/22 Estimación del efecto del incumbente en la elección de Intendentes en Uruguay (1971-2020)

The purpose of this paper is to determine the effect of individual and partisan incum-
bency on the election of regional governors in the period 1971-2020 for Uruguay. The
legal framework prohibits the reelection of governors after two terms in office, so there are cases of open elections. By exploiting these situations, which imply the exogenous and unintentional withdrawal of governors, it is possible to estimate separately the effect of the incumbent at the individual level (effect of being governors on the chances of being reelected) and the party level (probability of the governors’s party of winning the next election). The estimates using the diff-in-disc technique yield a positive and significant individual incumbency effect of 73.7%, which is higher than the one reported in the only available precedent. At the same time, I find a negative and statistically non-significant party incumbency effect, in line with the literature. As a way to explain these effects and shed light on the issue, different mechanisms are explored. Additionally, I test the hypothesis of a change in incumbency effects due to the Constitutional Reform of 1997, which entailed the separation of national and departmental elections and the limitation of candidacies. There is a decrease in individual incumbency and an increase in party incum- bency for a period before the implementation of the Reform, a trend that is accentuated after 1997. This allows us to think of the Constitutional Reform as consolidating changes that were already taking place in the Uruguayan political system. The evidence found allows us to affirm that in Uruguay incumbency is a personal, individual matter, and that when open elections take place, the new candidate of the incumbent party is not able to capitalize on his or her party’s performance.
Key words: term limits, incumbency effect, dif-in-disc, Uruguay.
JEL classification: D72, H11, H70, R50.