I investigate causal machine learning (CML) methods to estimate effect heterogeneity by means of conditional average treatment effects (CATEs). In particular, I study whether the estimated effect heterogeneity can provide evidence for the theoretical labour supply predictions of Connecticut's...