New evidence on the convergence and regional clusters in China: a weighted continuous distribution dynamics approach

This paper examines income convergence by using a continuous distribution dynamics approach which takes both population and economic size into consideration. The study is based on a panel dataset of 31 Chinese provinces from 1952 to 2013. The impacts of geographical location, capital accumulation, trade openness and industrial structure are also evaluated by conditional analysis. This paper provides a new angle for examining the spatial distribution dynamics of China’s provincial income. The results show that the ergodic distribution of the pre-reform period is multimodal, while the ergodic distribution of the post-reform period is bimodal. However, the ergodic distribution of the more recent period 2000–2013 is found to be unimodal. All three ergodic distributions are right-skewed. The weighted analysis shows that by neglecting population and economic size, the economic performance in the pre-reform period may be overestimated, while the poverty reduction effect in the post-reform period may be underestimated. The results suggest that relocating people and economic activities to high-income provinces is an important approach to reduce relative poverty. Conditional analysis results indicate that geographical location, capital accumulation, trade openness and industrial structure have a significant influence on the distribution dynamics of per capita GDP. However, the observed bimodality cannot be explained by one of the four factors.