A Note on the Scale Efficiency Test of Simar and Wilson

Simar and Wilson (2002) procedure to test for returns to scale has a drawback as it may generates inconsistent values. We use a counter example to show that their returns to scale test procedure generates non admissible pseudo-values. Consequently, the test results are not consistent and cannot be used to test for returns to scale. We propose a remedy that consists in a correction of the simulation algorithm to ensure the consistency of the estimated test statistic.