Empirical studies of convergence in income, productivity and competitiveness: the experience of Asian economies

Abstract

This thesis comprises three essays on convergence issues in selected Asian countries, focusing on both macro and micro perspectives. These papers encompass issues on convergence testing from time series and panel perspectives, with in-depth coverage on convergence in aggregate and structural ( inter sectors and sub sectors convergence) of the Asian economies. The first essay examines the convergence hypothesis using both time series and panel frameworks, from a macro perspective. Utilizing the concepts of stochastic convergence, the study tests the income convergence hypothesis by using both linear and nonlinear unit root tests. The present study emphasizes the usefulness of the nonlinear unit root tests due to Kapetanios, Shin and Snell (2003) and extended by Chong, Hinich, Liew and Lim (2008) to permit the test of long-run convergence and catching-up hypothesis. Next, the study utilized the recent panel unit root test for the nonlinear heterogeneous panel model proposed by Ucar and Omay (2009). In the present study, despite using the univariate and panel nonlinear tests for convergence, the results suggest strong evidence of divergence among the Asian countries with Japan. Further analysis was undertaken using tests for convergence with two robust tests that do not require stationarity in the data generating process suggested by Nahar and Inder (2002) and recently proposed by Phillips and Sul (2007a). On the one hand, the Nahar-Inder test indicates divergence between the Asian economies with Japan (except for Singapore), however, on the other hand, the Phillips-Sul test suggests that, all the other Asian economies converge towards Japan. The study suggests that since the Asian economies are in iii various stages of development, the Phillips-Sul test for convergence is more appropriate for such transition economies. Thus, testing for convergence using the unit root and cointegration test for transitional dynamics in the data may not be appropriate. Bernard and Durlauf (1996: p.172) have cautioned that “the (time-series) test may therefore be invalid if the data are largely driven by transition dynamics” and in this study, this found support. The second paper tries to bridge the gap between the macroeconomic issue of productivity convergence at the aggregate level and the microeconomic issues of convergence at the industry level. The study investigates structural convergence in selected Asian countries over the period of 1970-2005 using the non-linear time-varying coefficients factor model proposed by Phillips and Sul (2007a). This model has the flexibility to model a large number of transition paths to convergence, and allows for convergence clubs as well. Structural convergence exists if the convergence progress in income is accompanied by convergence at a sectoral or disaggregated level. The study finds strong divergence in income convergence at the aggregate level, and the clustering shows four clubs. To be robust, the study presents three measures of structural convergence, namely productivity, labour shares and value added. Convergence tests on productivity and value added shares indicate divergence in all sectors that leads to possible creation of club convergence. On the other hand, the labour share shows convergence in aggregate in three sectors (manufacturing,mining and construction). Also, the paper reveals that there is strong sectoral club convergence within the manufacturing sector in Asian whereas the evidence of convergence club for services, agriculture, and construction as well as for mining is rather weak. While the integration process is actively geared in Asian, the question of candidates’ suitability for the AEC (Asian Economic Community), as proposed in this study JAKITH (Japan, Korea, India, Taiwan, Hong Kong, and ASEAN) is still a debatable issue. Utilizing the concept of sub-sectoral convergence, the third study tests the convergence in competitiveness for 13 Asian countries for medium/high technology manufacturing sub-industries in three decades. This paper focuses on the issue of competitiveness measured in terms of productivity and labour input efficiency at the industry level within the manufacturing sector of Asian countries. The study applies the non-linear time-varying coefficients factor model that allows for large transition path to converge and also convergence clubs (Phillips and Sul, 2007a). The results indicate that in general Asian countries have a constant increase in its competitiveness yet no aggregate convergence is achieved. Club convergence indicates that labour productivity is likely to iv be driven by high income countries, as opposed to labor input efficiency. However, the low income countries show significant increase of productivity and high labour input efficiency as compared to richer countries. The high labour input efficiency indicates low wages paid to workers, significantly correlated with high density population countries as India and Indonesia. Based on comparative advantage theory, focusing on the industry level, the study also utilized the Krugman specialized index to show the clustering of concentrated industries among countries in Asia. The study found no trace of close to convergence situation, which indicates that the economic activity of Asia is focused or concentrated in specific activities, explaining the divergence in the sub-industries. The growing similarity of Asian economies in terms of overall productivity masks a continued high degree of specialization in particular industries. These findings should help policy makers for both target groups in designing appropriate growth-oriented programme as well as in setting priorities in their implementation

More information

Identifiers

ORCID for Jean-Yves Pitarakis: orcid.org/0000-0002-6305-7421

ORCID for Jian Tong: orcid.org/0000-0001-9367-4853

Catalogue record

Export record