The nonlinear ARDL approach and productivity bias hypothesis: Evidence from 68 countries

In this paper, we test the Productivity Bias Hypothesis for each of the 68 countries for which we were able to collect data. When we used the linear ARDL approach, we found support for the hypothesis in 16 countries in the long run. However, when we used the nonlinear ARDL approach, we added 11 more countries to the list. Furthermore, the nonlinear approach supported the short-run asymmetric effects of productivity ratio in 47 countries that lasted into long run asymmetric effects in 29 countries in the sample. Seventeen additional countries were added to the list when both models were augmented with a few additional determinants of the real exchange rate. (c) 2021 Board of Trustees of the University of Illinois. Published by Elsevier Inc. All rights reserved.