Revisiting the Unit Root Hypothesis
A Historical and Empirical Study
Keywords:
Unit Root Hypothesis, Post-KeynesianAbstract
This work presents empirical evidence supporting the Unit Root Hypothesis for Argentina real gross domestic product (GDP) between 1810 and 2004 (which is the longest series available). Particularly, six tests will be
conducted, Augmented Dickey Fuller (ADF), Dickey Fuller Generalised Least Square (DFGLS), Phillip-Perron (PP), Kwiatkowski-Phillips-Schmidt-Shin (KPSS), Elliott-Rothenberg-Stock Point-Optimal (ERS) and Ng-Perron
Test. The relevance of that hypothesis is linked with the long-run consequences of policy shocks, fiscal and monetary, as suggested by some Post-Keynesian models of growth where the principle of effective demand is incorporated in
the long run and money is not neutral1. The main finding is that all tests applied favor the existence of a unit root in Argentine GDP when including the whole time series between 1810 and 2004.