Estimating Candidate Valence

We estimate valence measures for candidates running in U.S. House elections from data on vote shares. Our estimates control for endogeneity of campaign spending and sample selection of candidates due to endogenous entry. Our identification and estimation strategy builds on ideas developed for estimating production functions. We find that incumbents have substantially higher valence measures than challengers running against them, resulting in about 9.2 percentage-point differences in the vote share, on average. We find that open-seat challengers have higher valence measures than those running against incumbents by about 5.3 percentage points. Our measure of candidate valence can be used to study various substantive questions of political economy. We illustrate its usefulness by studying the source of incumbency advantage in U.S. House elections.