This article uses a VAR model to empirically test Chinese coal mine safety regulations based on a theoretical analysis of their outcomes. The results show that the mine safety regulations are effective over the long term, represented by the long-term drop in casualty rates per million tons of coal. This is offset in the short term by the adverse behavior of coal miners, meaning that regulatory institutions must make trade-offs between long-term and short-term effects. The empirical analysis also finds that increases in coal output will lower the casualty rate per million tons of coal in the short term, but increase it in the long term. Based on these empirical results, the paper offers policy recommendations for improving China's coal mine safety regulations.