Unobservable, but Unimportant? The Influence of Personality Traits (and Other Usually Unobserved Variables) for the Estimation of Treatment Effects

A large and highly used number of treatment effects estimators rely on the unconfoundedness assumption ("selection on observables") which is fundamentally non testable. When evaluating the effects of labor market policies, researchers need to observe both variables that affect treatment participation and labor market outcomes. Even though many countries now offer access to (very) informative administrative data concerns about the validity of the unconfoundedness assumption remain. The main concern is that the observable characteristics of the individuals may not be enough to properly address potential selection bias. This is especially relevant in light of the research about the influence of personality traits on economic outcomes. We exploit a unique dataset that contains a rich set of administrative information of individuals entering unemployment in Germany, as well as several usually not observed characteristics like personality traits, attitudes, expectations, and job search behavior. This allows us to empirically assess how estimators based on the unconfoundedness assumption perform when alternatively including or not these usually not observed variables. Our results suggest that these variables, which appear as relevant for the participation probabilities, do not matter for the estimation of effects of the programs evaluated.