In 2008, the Chinese government has promulgated a package of ‘reverse economic cycles’ based on the ‘4 trillion yuan stimulus package’, in order to resist the strong impact of international financial crisis on the Chinese economy. From the perspective of the implementation of this plan, it has two distinct characteristics: on the one hand, it focuses on major infrastructure, affordable housing, disaster recovery & reconstruction, independent innovation and structural adjustment, livelihood projects, energy conservation and emission reduction and ecological engineering as well as social cultural industries and other related industries and enterprises; on the other hand, they are inclined to industries and enterprises associated with the disadvantaged groups in the central and western regions of China. After the stimulus package was implemented, it not only quickly got rid of the influence of the global financial crisis and succeeded in achieving rapid growth, but also became the mainstay of promoting the global economy from the doldrums. However, an interesting phenomenon is that while China’s economy is ‘rapidly recovering from low levels’, Chinese firms’ leverage is not only rising but also showing structural feature at the overall level, becoming one of the main structural factors promoting the rise in the overall social leverage rate. Therefore, the ‘4 trillion yuan stimulus package’ injects vitality into the restoration of Chinese economy. However, does it promote the rapid accumulation of the debt level of the whole society, especially the level of indebtedness of enterprises and local governments? The nearly simultaneous historical events of ‘4 trillion yuan stimulus package’ and rising Chinese firms’ leverage during the financial crisis provide a quasi-natural experiment for the study on the factors influencing the Chinese firms’ leverage. At the same time, there are key industries and regions of ‘4 trillion yuan stimulus package’, but China is in a special phase of economic restructuring and covers two heterogeneous micro-entities, namely state-owned and non-state-owned enterprises, also providing a good sample for further tracing the source of ‘high leverage’ in Chinese enterprises and the structural changes in leverage ratio among heterogeneous firms in different industries. Through this study, we not only try to enrich the study of the effect of ‘4 trillion yuan stimulus package’ at the micro-enterprise level and provide experience for the macro-policy practice of future economic development, but also enhance the exploration of origin and antecedents of ‘high leverage’ at the macro-policy level, which provides empirical evidence for the direction and focus of ‘deleveraging’. For this reason, this paper regards the ‘4 trillion yuan stimulus package’ as a quasi-natural experiment, select the sample data of Chinese listed companies from 2005 to 2012 by using the method of double difference estimation to explore the overall impact of the ‘4 trillion yuan stimulus package’ on the leverage ratio of listed companies in key industries and regions, and examines whether there is a significant difference in the leverage ratio between listed enterprises in key industries and those in non-key industries. Then, after considering that the implementation of this plan is more likely to be in the central and western regions, it is probable that whether the ‘4 trillion yuan stimulus package’ produces a relatively significant difference between the leverage ratios of listed enterprises in key industries and one in non-key industries in different regions. Furthermore, since China has a large number of non-listed enterprises, the data limited to the level of non-listed enterprises is not complete enough within the sample year, so that it exists some evaluation deviation on the impact of the ‘4 trillion yuan stimulus package’ on the firms’ leverage. Therefore, this paper continues to extend the research ideas of listed enterprises sample, and further uses the sample of state-owned enterprises to examine the impact of ‘4 trillion yuan stimulus package’ on the leverage, so the results of two sets of samples of listed enterprises and state-owned enterprises become a mutual evidence. From the microcosmic consequences of the implementation of ‘4 trillion yuan stimulus package’, this paper explores the reason for ‘high leverage’ and the structural changes in leverage of heterogeneous firms from different industries. The results of this paper show that, on the one hand, using two sets of sample data from the Chinese listed enterprises and state-owned enterprises as proof of each other, the ‘4 trillion yuan stimulus package’ significantly increases the firms’ leverage of the industry-focused program in these key industries; on the other hand, the empirical results of the two sets of companies complement each other, indicating that, in the long run, although the plan has significantly increased the leverage ratio of listed enterprises in the key industries of eastern region, it also raises the leverage of state-owned enterprises in key industries in central and western regions, especially the non-listed state-owned enterprises in central and western regions. Therefore, this paper argues that, in the direction of ‘deleveraging’, the decision-making bodies should give priority to lowering the firms’ leverage of the unlisted state-owned enterprises which focus on the key industry in implementing the ‘4 trillion yuan stimulus package’, especially those in the central and western regions by the end of 2008. On the one hand, the governments can both broaden the financing channels by speeding up the examination and approval of state-owned enterprises, improve the proportion of direct financing by enterprises in the capital market using stocks and other means, speed up the marketization reform in the central and western regions, perfect the multi-level capital market, and establish a diversified financing channels for non-listed state-owned enterprises in the central and western regions, in order to gradually absorb part of the debt pressure of these enterprises. On the other hand, the governments should continue to reduce the burden on state-owned enterprises, especially those non-listed state-owned enterprises in the central and western regions, and encourage these enterprises to improve their profitability and operating performance so as to make the state-owned capital more optimized, and resolve the risk of ‘high leverage’ by preserving and increasing the value of state-owned assets.