I propose two new tests of Falk and Knell's (2004) prediction that individuals' reference income increases with ability. To overcome the difficulty that the reference incomeis not observed in existing large data sets, I extend Falk and Knell's model to establish a link between immigrants' reference income and their return visits to their countries of origin. I derive the (arguably counter-intuitive) predictions that more educated immigrants are less likely to have returned to their country of origin for a visit, and that they are more likely to have difficulty feeling at home when they do return for a visit. Both predictions are confirmed on four waves of data from the German Socio-Economic Panel. The estimated marginal effects are statistically significant, but rather small in size.