The issue of zombie firms is one of the key concerns of China’s supply-side structural reform. To clarify the causes of zombies is of great significance in disposing zombie firms. This paper looks into the causes of zombie firms from the perspectives of government-banks relationship in the context of the local government intervention along with Chinese-style fiscal decentralization and the historical role of state-owned banks as the ″second fiscal department″. The formation of zombie firms is more likely to be institutional. The average probability of a state-owned enterprise becoming a zombie is 3.1% higher than that of a private one, which indicates that the status of ″state-owned enterprise″ increases the probability by around 10.9%. In addition, the coefficient of the ratio of regional state-owned-enterprises’ output to actual GDP is significantly positive, indicating that the more the regional economy relies on the state-owned sector, the more likely local enterprises will become zombies. Furthermore, the stronger the intervention motives of local governments are, the higher the probability that local state-owned firms will become zombies. For every one-standard-deviation change of fiscal decentralization, the probability will change by 4.41% for state-owned enterprises, equivalent to 9.8% of the mean, the economic significance of which should not be ignored. On one hand, the increase of the level of competition in the regional financial market significantly reduces the possibility for enterprises under various types of ownership to become zombies. For one-standard-deviation change of the intenseness of competition, the probability changes by 1.46%, which is 5.14% of the mean. From the regression results of the nonlinear model, the less intense the competition, the stronger the improvement effect of raising the level of competition. The difference in the improvement effect on state-owned and private enterprises narrows along with the increasing intenseness of competition. For state-owned enterprises, the mean marginal effect of competition increases with the level of itself, but the rate of increase drops. On the other hand, the local government’s stronger motive of intervention reduces the effect of financial market competition, and indirectly increases the possibility of turning local companies into zombies For state-owned enterprises, the above-mentioned increasing effect is 60 percent higher than that of private enterprises. Companies with lower total factor productivity and those who are not involved in R&D activities are more likely to become zombies, especially for state-owned ones. Lower productivity, as a competitive disadvantage, is one of the mechanisms to explain why local governments keep interfering with financial markets and supporting zombie firms. These findings are meaningful for the comprehensive understanding of the causes of zombie firms, for the reflection on the orientation of regional competition, for the unhooking of government and financial enterprises, and for the governance of state-owned enterprise. We believe that breaking the symbiotic relationship between government, banks and enterprises, promoting openness within and beyond financial industries, financial disintermediation, the deleveraging and classification of zombie firms, are all essential to getting out of the dilemma of zombies in China. Meanwhile, adopting sustainable-growth assessment indicators such as Inclusive Wealth Index and guiding local governments to abandon the practice of distorting resource allocation will be institutional guarantee for removing the soil which nourishes zombie firms. Research up to now has applied empirical models or game theory models to explore the links between government-enterprise, bank-enterprise relationship and zombie firms, but little research has been done on the interaction between the local government and banks, neither has it provided empirical evidence. In this research, we focus on the impact of government-bank relation, rather than the former two kinds of relation. This complements the discussion on the causes of zombie firms. In addition, we compare the changes in the number of zombie firms with the benchmark interest rate during the same period, which further examines the applicability of the CHK method to China’s firm-level data.