摘要: Presidential campaign visits for members of the House are important events during campaigns. The role president in elections serves a dual purpose. First, it may increase probability being majority party Congress. Second, ensures that Congress owe some debt to president. Assessing eectiveness such presidential interventions, however, is very dicult. Due powerful selection eects and unobserved confounding, dicult estimate these visits. Both regression matching based estimators ill-suited this context due weak set observables predicting treatment. In analysis follows I use Manski-type bounds on Bayesian treatment estimator assess