Understanding the impact of local government debt on economic development has emerged as a focal issue for both academic research and policymakers. This study adopts a financing structure perspective and utilizes panel data from 214 cities and 3,228 A-share listed companies in China (2017–2023) to empirically investigate the impact of local government debt on corporate “shift from virtual to real economy” and its underlying mechanisms. The expansion of local government debt significantly promotes enterprises “shift from virtual to real economy”. The positive impact of local government debt on enterprises” transition from financialization to the real economy is more pronounced among firms in first tier and new first-tier cities, non-state-owned enterprises, and labor-intensive industries. Further analysis indicates that local government debt drives capital reallocation from financial investments to real investments by alleviating corporate financing constraints. This study proposes policy recommendations including optimizing debt fund allocation, further optimizing the financing environment, implementing differentiated regulatory measures. These suggestions provide both a theoretical foundation and practical references for synergistically advancing debt governance and real economy revitalization.

We provide firm-level evidence on the real effects of shadow banking in terms of technological innovation. Firm-to-firm entrusted loans, the largest part of the shadow banking sector in China, enhance the borrowers’ innovation output. The effects are more prominent when the borrowers are subject to severer financial constraints, information asymmetry, and takeover exposures. A plausible underlying channel is capital reallocations from less productive but easy-financed lender firms to more innovative but financially less-privileged borrower firms. Our paper suggests that shadow banking helps correct bank credit misallocations and thus serves as a second-best market design in financing the real economy.

We provide firm-level evidence on the real effects of shadow banking in terms of technological innovation. Firm-to-firm entrusted loans, the largest part of the shadow banking sector in China, enhance the borrowers’ innovation output. The effects are more prominent when the borrowers are subject to severer financial constraints, information asymmetry, and takeover exposures. A plausible underlying channel is capital reallocations from less productive but easy-financed lender firms to more innovative but financially less-privileged borrower firms. Our paper suggests shadow banking helps correct bank credit misallocations and thus serves as a second-best market design in financing the real economy

We study credit allocation across  rms and its real e ects during China's economic stimulus plan of 2009-2010. We match con dential loan-level data from the 19 largest Chinese banks with  rm-level data on manufacturing  rms. We document that the stimulus-driven credit expansion disproportionately favored state-owned rms and  rms with lower average product of capital, reversing the process of capital reallocation towards private  rms that characterized China's high growth before 2008. We argue that implicit government guarantees for state-connected  rms become more prominent during recessions and can explain this reversal.

Based on a DSGE model embedded with a stock market, we inspect interconnection between China's financial markets and macroeconomic cycles. We find consumption, investment and capacity utilization display significant and positive responses to stock market booms triggered by financial and confidence shocks, however, inflation responds insignificantly. We perceive a counteractive and significant reaction of China's monetary policy rule to credit-to-GDP gap at business cycle frequency. We decompose stock price into fundamental value influenced by the financial shock and speculative bubble driven by the confidence shock, and the confidence shock's contribution to stock price fluctuations is estimated to be about 14.8%. Model validation based on the DSGE-VAR framework indicates no serious structural model misspecification.