Another Look at the Identification at Infinity of Sample Section Models

It is often believed that without instrument, endogenous sample selection modelsare identified only if a covariate with a large support is available (see Chamberlain,1986, and Lewbel, 2007). We propose a new identification strategy mainly based onthe condition that the selection variable becomes independent of the covariates whenthe outcome, not one of the covariates, tends to infinity. No large support on thecovariates is required. Moreover, we prove that this condition is testable. We finallyshow that our strategy can be applied to the identification of generalized Roy models.