We propose and test an unexplored collateral channel of government investment impacts on the private-sector finance by exploiting the variations created by China’s subway expansions. We purposefully build a geo-financial dataset that links private firms with their nearest new subway stations. Using a firm-level stacked difference-in-difference research design, we find that the introduction of a new subway station is associated with an 12.44 percent increase in the debt/asset ratio of private firms located in the one-kilometer radius of the station. We also find that the values of firms’ productive structures and land increase due to the introduction of the station, suggesting that the financial effect is likely achieved via the collateral channel. This assessment is buttressed by our finding that firms close to a new subway station are more likely to acquire production structures and land than firms far away from any subway station. Overall, our findings suggest that government infrastructural investment has a crowding-in effect on private firms’ finance via the collateral channel.

scoco@ceibs.edu