The Non-Linear Effect of Chinese Financial Developments on Energy Supply Structures

Currently, oversupply coal and coal-based power in China poses a great challenge to energy structure optimization and emissions reduction. The energy industry, however, is closely linked to the financial sector. In view of this, using a non-linear Panel Smooth Transition Regression (PSTR) model, this paper examines the threshold effects of financial developments on energy supply structures for 17 energy supply provinces in China observed over 2000–2014. The main results are: (1) The ratio of coal supply (LCSR) specification is seen to be a four-regime PSTR model with added value in the financial industry/GDP (LFIR) as the threshold variable. The LFIR and LCSR show a positive correlation, and the elastic coefficients change between 0.02 and ~0.085; the impact of financial institutions’ loan balance/GDP (LLAN) on LCSR takes on an inverse U-shaped curve: first positive, then negative, and again positive with the financial crisis in 2008 as the turning point; (2) The ratio of thermal power generation (LTPG) specification is seen to be a two-regime PSTR model with investment in the coal industry/GDP (LCIR) as the threshold variable. Results show that LFIR has a negative effect on LTPG, and the coefficients in the low regime tend to be 0.344%, then gradually decrease to 0.051% in the high regime. The influence of LLAN on the LTPG is positive before and negative after the financial crisis. The influence of the foreign direct investment GDP proportion (LFDI, the degree of financial openness) on the LCSR and LTPG both remain negative. Therefore, in the process of formulating energy conservation policies and adjusting energy-intensive industrial structures, the government should fully consider the effect of financial developments.