Assessing Local Government Debt Risks in China: A Case Study of Local Government Financial Vehicles

25 Pages Posted: 22 Sep 2015

See all articles by Kunyu Tao

Kunyu Tao

Central University of Finance and Economics

Date Written: September–October 2015


Strong credit expansion in China after the recent global financial crisis has brought local government financial vehicles (LGFV) into the spotlight. Rapid growth of LGFV has triggered concern about local government indebtedness, banks' asset quality and, more broadly, China's medium‐term financial stability and sovereign risk. This paper constructs a unique firm‐level dataset to evaluate the country's local government debt. We find an uneven distribution of LGFV, which are concentrated in the coastal areas, and a deterioration of their debt repaying ability from 2010 to 2012. We use principal component analysis (PCA) along with multivariate discriminate analysis (MDA) to identify the credit risk of LGFV based on conventional financial variables as well as local governments' fiscal status. We also estimate the safe boundaries of debt bearing at the provincial government level. The estimations reveal more severe local government debt risks in the middle‐western provinces and higher risks associated with LGFV at the municipal and county levels. Although it is very unlikely that there will be a national debt crisis in China, the high risk of LGFV should be noted and effectively controlled by improving the fiscal transparency of local governments and reforming the fiscal system.

Keywords: credit risk, fiscal risk of China, local government financial vehicles

JEL Classification: E62, H63, H81

Suggested Citation

Tao, Kunyu, Assessing Local Government Debt Risks in China: A Case Study of Local Government Financial Vehicles (September–October 2015). China & World Economy, Vol. 23, Issue 5, pp. 1-25, 2015, Available at SSRN: or

Kunyu Tao (Contact Author)

Central University of Finance and Economics ( email )

39 South College Road

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics