This paper examines total factor productivity differences between exporting and non-exporting firms. These differences are documented on the basis of a sample of Spanish manufacturing firms over the period 1991-1996 drawn from the ESEE. The paper also examines two complementary explanations for the superior productivity of exporting firms: 1) the market selection hypothesis, and 2) the learning hypothesis. Nonparametric tests are implemented for testing these hypotheses. Results indicate clearly higher levels of productivity for exporting firms than for non-exporting firms. Evidence favours the hypothesis of self-selection of more efficient firms into the export market. However there is little evidence that these efficiency gains are supportive of the learning-by-exporting hypothesis.