Observational equivalence and a stochastic cointegration test of the neoclassical and Romer's increasing returns models

Abstract

In this paper, we suggest a time series test based on the idea of stochastic cointegration to compare endogenous growth models with those based on exogenous technological progress, and apply it to the neoclassical and Romer's models. First, the non-stationarity and cointegration implications under these two models are derived. They provide the basis for an examination of the phenomenon of observational equivalence and for empirical analysis. We then apply the test to the data of Japan, UK and USA. Empirical results suggest that in each country, there is (are) missing unit root process(es) in the production technology and further investigation is needed to understand the mechanics of economic growth.