We construct a growth model consistent with China's economic transition: high output growth, sustained returns on capital, reallocation within the manufacturing sector, and a large trade surplus. Entrepreneurial firms use more productive technologies, but due to financial imperfections they must finance investments through internal savings. State-owned firms have low productivity but survive because of better access to credit markets. High-productivity firms outgrow low-productivity firms if entrepreneurs have sufficiently high savings. The downsizing of financially integrated firms forces domestic savings to be invested abroad, generating a foreign surplus. A calibrated version of the theory accounts quantitatively for China's economic transition.

The extant corporate investment literature has documented that information asymmetry and agency conflicts between managers and outside investors prevent firms from making optimal investment decisions. In this study, we investigate whether government intervention, as another form of friction, distorts firms' investment behavior and leads to investment inefficiency. Using Chinese data, we test this by measuring government intervention at two different levels. First, we compare investment efficiency between SOEs and non-SOEs. We find that the sensitivity of investment expenditure to investment opportunities is significantly weaker for SOEs. Second, we measure government intervention by whether a firm is politically connected through the employment of top executives with a government background. We find that political connections significantly reduce investment efficiency in SOEs. However, we do not find such evidence in non-SOEs. Taken together, our findings suggest that government intervention in SOEs through majority state ownership or the appointment of connected managers distorts investment behavior and harms investment efficiency.