Reopening the Convergence Debate when Sharp Breaks and Smooth Shifts Wed, 1870-2010

Abstract This paper attempts to re-investigate the catching-up (stochastic convergence) hypothesis among the selected 16 OECD countries applying the time series approach of convergence hypothesis with annual data over one century. To reach this aim, we propose a model which specifies a trend function, incorporating both types of structural breaks – that is, sharp breaks and smooth shifts using dummy variables and Fourier function respectively. In order to detect the sharp breaks, we apply the multiple structural break models (Bai & Perron, 1998) and the Fourier function proposed in Becker et al. (2004) to capture the smooth shifts. Our results show that most divergence process occurred over World War I (WWI) and World War II (WWII). Among the 69 estimated break points occurred over the period 1870-2010, 75 % of those break points result in catching-up and the remainder results in divergence.