Does financial deepening promote risk sharing in China?

Abstract

This paper addresses the issue of whether financial deepening has promoted interregional risk sharing in China. It first examines the evolution pattern of overall risk sharing across China as a whole and in its coastal, central and western areas in different periods. The level of risk sharing exhibits a declining pattern over time, most of which is attributable to the coastal areas. This is in striking contrast to the rapid development of financial markets and financial intermediaries over the decades. By analyzing whether the development of the credit and capital markets affects consumption risk sharing, we conclude that financial development has not promoted risk sharing for the nation as a whole. While the expansion of the capital market has promoted risk sharing in the coastal areas in spite of the overall declining trend in that region, there is no evidence to show that this is true for the western areas.

Keywords:

• regional risk sharing
• Chinese economy
• capital market
• credit market

JEL classifications:

• O16
• O53
• R11

Acknowledgements

We are indebted to three anonymous referees for their helpful comments that greatly improved this paper. We would also like to thank Zhigang Li, Guang-Zhen Sun, Shang-Jin Wei and Xiaobo Zhang for their helpful discussions and comments. Financial supports from the Asian Development Bank Institute, the Research Grants Council of Hong Kong SAR Government (project no. CUHK442709), the National Social Science Foundation of China (10CGL011) and the Key Projects in Philosophy and Social Sciences Research Program of the Ministry of Education of China (08JZD0011) are also greatly appreciated. The paper was initiated when Qing He visited the Hong Kong Institute of Monetary Research. The views are those of the authors and do not reflect those of the Hong Kong Institute of Monetary Research, its Council of Advisors or the Board of Directors.

Notes

1. During the reform period, per capita output growth was about 9.02%, whereas per capita consumption growth was about 7.86%. The t-statistics for the difference between output growth and consumption growth is 5.707, which is significant at the 1% level.

2. To maintain data consistency, we exclude Tibet and Chongqing from the original list of 31 provinces.

3. Chen (2006) suggests that household deposits are much less influenced by central government policies than are total savings and loans.

4. The coastal areas include Beijing, Fujian, Guangdong, Hainan, Hebei, Jiangsu, Liaoning, Shandong, Shanghai, Tianjin and Zhejiang. The central areas include Anhui, Heilongjiang, Henan, Hubei, Hunan, Jiangxi, Jilin and Shanxi. The western areas include Gansu, Guangxi, Guizhou, Inner Mongolia, Ningxia, Qinghai, Shaanxi, Sichuan, Xinjiang and Yunnan.

5. Crucini and Hess (2000) report consumption volatility for Canadian provinces and Japanese prefectures to be about 2%.

6. Following Athanasoulis and Van Wincoop (2001), we define the national counterpart as for any regional variable x_i ; θ_j is the relative weight of region j’s endowment to the national endowment.

7. See Sørensen et al. (2007).

8. Because the estimates of (1 − β_t ) vary from year to year, we smooth them by estimating the average over a nine-year rolling window. For example, the extent of risk sharing in 1989 is equal to (1 − β_t ), and β_t represents the estimates for the panel regression from 1981 to 1989.

9. This is obvious, as the degree of risk sharing attained by region i is equal to . See Sørensen et al. (2007).

10. As the consumption correlation and volatility measures are derived from a long period of time for each province, conducting panel data analysis similar to that in Section V is not feasible.