Investigating the Convergence Hypothesis in the Eu: More Evidence Accounting for Structural Breaks

Abstract

The aim of this paper is to test the convergence hypothesis for the EU countries with Netherlands which is considered as the benchmark. The empirical analysis uses Gross Domestic Product (GDP) per capita in Purchasing Power Parity (PPP) in constant prices of 2005 and covers the period 1950-2010. The empirical approach complementarily employs unit root tests for stochastic convergence and a test proposed by Tomljanovich and Vogelsang (Empirical Econ 27: 49-62, 2002) and Nieswiadomy and Strazicich (Econ Inq 42: 323-340, 2004) that is based on Carlino and Mills' (J Monet Econ 32: 335-346, 1993) methodology. Applying the unit root tests for the relative per capita real GDP series for each country, we are able to reject the unit root hypothesis for 6 out of 14 countries. However, our results suggest there is stronger evidence in favor of convergence when we account for one or two endogenous structural breaks in the intercept and slope of the trend function. Based on Carlino and Mills methodology, we found strong evidence of a catch-up process until the mid-1980 towards the Netherlands for all countries except for the UK, although, after 1985, it is clear that the economies show different behavior which is probably associated with differences in the growth process.

Convergence of Output per Worker in G7 Countries: Evidence from the Club Convergence Test

“…Applied studies on developed countries in the literature testing the Convergence Hypothesis are generally based on OECD countries (Liu and Ruiz, 2006;Caggiano and Leonida, 2009;Marattin and Salotti, 2011;Ceylan et al, 2013;Bahmani-Oskooee et al, 2017), The European Union (EU) and the subgroups of the countries that make up the EU (Monfort, et al, 2013;Borsi and Metiu, 2015;Ceylan and Abiyev, 2016;Chapsa et al, 2015, Cabral, et al, 2019, Bolea, et al , 2018Cavallaro and Villani, 2021), countries classified as high-income by the World Bank (Desli and Gkoulgkoutsika, 2021) and homogeneous country groups such as G7 countries (Cellini & Scorcu, 2000;Ceylan, 2010a). In studies employing methodological approaches based on stochastic approach, deterministic approach or the combination of the two, it is seen that the findings on the power of convergence vary depending on the methodology applied, countries selected, and the time period examined, but generally are in favor of the Convergence Hypothesis.…”

Economic growth

“…King and Ramlogan-Dobson (2015) used Fourier-type tests, to examine stochastic convergence in eighteen Latin American countries, finding that in almost all, growth is consistently related to the U.S. growth, but they had converged to a quite low income level. Chapsa et al (2015) tested for stochastic convergence in the EU countries, using various unit root tests, with the Netherlands as the benchmark country. The unit root, and the convergence hypothesis were rejected in 6 out of 14 countries.…”

“…Nonetheless, the first part of the time samples contains more instances of convergence than the second one. Chapsa, Katrakilidis and Tabakis (2015) test for stochastic convergence, using unit root tests and a test proposed by Tomljanovich and Vogelsang, ( 2002) and Nieswiadomy and Strazicich, ( 2004) that is based on Carlino and Mills' (1993) methodology, in the E.U. countries with the Netherlands as the benchmark.…”