Policy experiments using large microeconomic datasets have recently gained ground in macro- economics. Imposing rational expectations, we examine robustness of evidence derived from ideal natural experiments applied to atomistic agents in dynamic settings. Paradoxically, once experi-mental evidence is viewed as su¢ ciently clean to use, it then becomes contaminated byex post endo- geneity: Measured responses depend upon priors and the objective function into which evidence is fed. Moreover, agentsípolicy beliefs become endogenously correlated with their causal parameters, severely clouding inference, e.g. sign reversals and non-invertibility may obtain. Treatment-control di§erences are contaminated for non-quadratic adjustment costs. Constructively, we illustrate how inference can be corrected accounting for feedback and highlight factors mitigating contamination.