The existence of geo-economic clubs must meet two conditions. First, per capita output within each club converges to a long-term stationary state. Second, per capita output between clubs diverges so that the rich become richer and the poor poorer. This paper proposes a production model to explain club divergence and uses both the unit root test and the decomposition of an inequality index to prove the existence of geo-economic clubs among the Chinese provinces based on real per capita GDP data from 1952 to 1997. The results reflect a serious problem of regional inequality and growth divergence in contemporary China. J. Comp. Econ., September 2001, 29 (3), pp. 466-484. University of Middlesex, London NW4 4BT; and University of Chongqing, China.

This paper uses a shift-share method to quantify the components of economic growth and structural changes in employment and investments at both the national and regional levels. The results indicate that rapid economic growth has been characterized by significant shifts in employment and investments between industries in China. Such structural changes have a distinct regional pattern which has important policy implications regarding spatial disparities in economic growth and income. The results shed some important light on the understanding of Deng's development thought that is formalized in this paper as a multi-tier and multi-stage development strategy for a large developing economy.

A production model is proposed to explain why economic growth may lead to regional inequality and club divergence. Economic growth starts from a few centres and spills over to other regions. The spillovere ect diminishes as distance rises, causing club formation with a clear geographical pattern.

This paper employs time series techniques, with or without a structural break, to investigate the question of China's regional per capita income convergence. Our results suggest that China's regions, especially the eastern and the western regions, have converged to their own specific steady states over the past 40 years. The Gini coefficient, the ratio of per capita income between regions, and the coefficient of variation confirm our findings. We also identify the big shocks on the relative regional per capita income by allowing a time break. D 2001 Elsevier Science Inc. All rights reserved.