Which nesting structure for the CES? A new selection approach based on input separability

Abstract

Nested CES production functions are very popular in economic applications, but their empirical validity has been questioned. In this paper, we outline a simple method that can be used to test whether a given dataset is consistent with a CES production technology and that informs on which nesting structure describes it more closely. The method is based on empirical testing input separability conditions, derived for all the feasible 3-input and 4-input nested CES. A Monte Carlo simulation environment is exploited to assess the accuracy of the test.