Salomo Hirvonen

Re-election is often assumed as the key objective in our political economy models. One way to empirically study re-election and its determinants is to estimate the incumbency advantage. We discuss how empirical studies of the incumbency advantage estimate different objects under one name. We focus on two. The incumbent-runner-up advantage is the effect of holding office for a term on a politician’s subsequent electoral career; it measures the value of office and speaks to the strength of legislatures. The incumbent-challenger advantage is the advantage incumbency confers over a challenger of equal quality in a single election, the object that theoretical models define. Neither is point identified: for the first, the researcher observes who does not rerun but not why; for the second, one needs the counterfactual electoral outcome of runners-up who did not run again. We show that the upper bounds of the two objects coincide. Thus, the ‘upper-bound incumbency advantage’ can be estimated in any electoral system and is meaningful under both concepts. Compiling it for 37 national legislatures, 33 estimated from candidate-level data, we provide the first cross-country comparison of the incumbency advantage of individuallegislators under a single harmonized estimand with bounds: it rises with economic development and democratic quality.

Regression discontinuity designs (RDD) are widely used in the social sciences to estimate causal effects from observational data. Following recent methodological advances, scholars can choose from various RDD estimators for point estimation and inference. This decision is mainly guided by theoretical results on optimality and Monte Carlo simulations because of a paucity of research on the performance of the different estimators in recovering real-world experimental benchmarks. Leveraging exact ties in personal votes in local elections in Colombia and Finland, which are resolved by a random lottery, we assess the performance of various estimators featuring different polynomial degrees, bias-correction methods, optimal bandwidths, and approaches to statistical inference. Using re-running and re-election as outcomes, we document only minor differences in the performance of the various implementation approaches when the conditional expectation function (CEF) of the outcomes in the vicinity of the discontinuity is close to linear. When approximating the curvature of the CEF is more challenging, bias-corrected and robust inference with coverage-error-rate-optimal bandwidths comes closer to the experimental benchmark than more widely used alternative implementations.