Working Paper: NBER ID: w14511
Authors: Ernesto Dal B; Martn Rossi
Abstract: We evaluate the effects of the duration of legislative terms on the performance of legislators. We exploit a natural experiment in the Argentine House of Representatives where term lengths were assigned randomly. Results for various objective measures of legislative output show that longer terms enhance legislative performance. We use a second experiment in the Argentine Senate to determine whether our results are specific to a particular chamber and a particular time. The results from the Senate reinforce the idea that longer terms enhance legislative productivity. Our results highlight limits to classic theories of electoral discipline (Barro 1973, Ferejohn 1986) predicting that shorter terms, by tightening accountability, will incentivize hard work by politicians. We discuss and test possible explanations. Our results suggest that the "accountability logic" is overcome by an "investment logic."
Keywords: legislative performance; term length; natural experiment; Argentina
JEL Codes: H1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| longer legislative terms (D72) | enhance performance (D29) |
| four-year terms (K16) | performance advantage over two-year counterparts (D29) |
| term length (C41) | legislative productivity (D72) |