There has been a great deal of discussion of China's four trillion yuan stimulus policy and the side effects it has caused: overcapacity, rising local government debt, and non-performing loans, even air pollution. What is less is clear are the mechanisms through which the stimulus policy has affected different types of firms, state-owned enterprises (SOEs) and non-state-owned enterprises, for example. Liang et al. (2017) find that local government debt in China has increased rapidly after the global financial crisis, and, at the same time, the leverage level of SOEs has climbed, while that of the non-SOEs has declined. They argue that this implies a correlation between local government debt and firm leverage. Because the local governments become major investors and their investments are mainly financed by borrowing in the financial markets, the local government debt has “crowed out” private lending, and, at the same time, “crowded in” SOE lending, thus causing a deterioration in the credit misallocation in China. According to standard macroeconomic theory, when a government increases borrowing to finance public spending, it will drive up interest rates. These higher interest rates are likely to discourage private sector lending and investment. Liang et al. provide a broader view of the “crowding out” effect. They identify two channels through which local government debt may affect firm level borrowing activities: one is the traditional financing cost channel through the credit market; and the other is a reallocation channel through government intervention. Liang et al. study these two channels and find that the second channel is more important. China has not fully liberalized her financial markets and interest rates are still not determined by market clearing. The government has close ties with the financial market and can influence the borrowing decision of market players in a very profound way. Liang et al. have not opened this “black box”, but their study can shed some light on what is happening in China and may stimulate more studies in this area. Liang et al.’s basic conclusions are convincing, but the real story may be more complicated than they suggest. First, the definitions of SOEs and non-SOEs are not always white and black. The Chinese government has a recent policy of encouraging the so-called “mixed economy”. What happened was mainly SOE investment in non-SOEs, and consequently, some of the non-SOEs are no longer classified as non-SOEs. This may affect the growth rate of SOE investment as well as non-SOE investment. Liang et al. also mention that large SOEs and local SOEs are quite different in many ways. Large SOEs can borrow nationally and are not constrained by local credit markets, while local SOEs are more dependent on local banks. But it is also true that at the local level, the behavior of SOEs are often quite similar to non-SOEs, because some of these local SOEs are competing with non-SOEs both in domestic and international markets, unlike the national SOEs which are often in monopolistic industries. Second, Liang et al. use Chengtou bonds as a proxy for local government debt. It is true that recently, the market for Chengtou bonds has developed rapidly, but the bulk of local government debts are in fact borrowings from banks. These local government borrowings are difficult to trace, and more prone to corruption and misallocation. The Chinese government has introduced new policies regulating Chengtou bonds, so the correlation of Chengtou bonds with overall local debt may have weakened. Third, we should be very careful in interpreting the policy implications of Liang et al.’s study. The rising SOEs leverage ratio and declining non-SOEs ratio clearly show that SOEs are less efficient, but there are also good reasons to believe that public investment may not necessarily “crowd out” private investment, especially after a financial tsunami. Maybe what matters is the type of public investment, and the way local governments chooses to finance their public investment. Even if local governments had not increased their debt levels, private investment would probably have remained sluggish, because of pessimistic market sentiment. If we agree that the recent decline of private investment is mainly caused by the recession, then policy makers should continue to use proactive fiscal policy, increase public investment, and, at the same time, find other ways to encourage private investment. For example, the government can liberalize some of the heavily regulated and tightly controlled service sectors.