Provincial Conditional Income Convergence in China, 1953–1997: A Panel Data Approach

Why is Chinese provincial output diverging?

“…The fact that these trends seem to be related to patterns in progress with market reforms in these provinces is consistent with the recent results of Weeks and Yao (2008), who find that the rate of technological progress differs significantly between coastal and interior provinces.…”

Regional Growth and Regional Decline

“…The literature on China, in particular, is extensive. It typically finds divergence for the 1970s, followed by a period of convergence in the wake of agricultural reforms, and then further divergence during the rapid industrialization of the 1990s (for example, Weeks and Yao 2003). For the 1990s onwards, the fast growth of the coastal provinces is often emphasized, consistent with a story in which market access has promoted industrial development.…”

Income inequality between Chinese regions: newfound harmony or continued discord?

“…Many of the regions are the same size as entire countries, but share institutions, language, and cultural traits which could cloud convergence tests in a cross-country setting. Previous empirical analyses of the convergence question also mirror the methodological divide mentioned above: cross-sectional studies have provided evidence in favor of economic convergence (Jian, Sachs, and Warner, 1996;Chen and Fleisher, 1996;Raiser, 1998;Weeks and Yao, 2003). In contrast, time series studies have documented economic divergence (Pedroni and Yao, 2006;Westerlund, Edgerton, and Opper, 2010).…”

“…Chen and Fleisher (1996), and Li, Zinan, and Rebelo (1998), found evidence to support both absolute and conditional convergence in the period 1978-1993. Cai, Wang, and Du (2002 and Raiser (1998) added support for conditional convergence as a consequence of market reform, and more recently, Zou and Zhou (2007) documented convergence within one group "developing" and one group of "developed" provinces, with faster convergence noted in the latter Weeks and Yao (2003) note that the cross-sectional methodology to measure convergence can su¤er from bias if the variables included fail to account for: i) unobserved province-speci…c heterogeneity in initial technology levels and ii) endogenous explanatory variables. Using Generalized Method of Moments estimation with province-and time-speci…c e¤ects, they found that provinces were diverging in the pre-reform period, but that the interior and coastal regions converged to parallel underlying steady-states after 1979.…”