We assess the extent to which discretion, unexplained variations in the terms of a loan contract, has varied across time and lending institutions and show that part of this discretion is due to private information that lenders have on their borrowers. We find that discretion is lower for secured loans and loans granted by a larger group of lenders, and is larger when the lenders are larger and more profitable. Over time, discretion is also lower around recessions although the private information content is higher. The results suggest that bank discretionary and private information acquisition behavior may be important features of the credit cycle.

This paper investigates how political influence affects firms’ financial flexibility and speed of adjustment toward target leverage ratios. We find that at the macro level, firms in environments with high political advantages, proxied by provincial affiliations with heads of state as well as political status and party rank of provincial leaders, adjust faster. At the micro level, firms that are state-owned, have CPC members as executives, or bear low exposure to changes in political uncertainty adjust faster. When interacted, the micro-level political factors have more significant impact.

We empirically investigate whether taking senior bank loans would enhance market discipline and control risk-taking among borrowing banks. Controlling for endogeneity concern arising from borrowing bank self-select into taking senior bank debt, we document that both the spreads and covenants in loan contracts are sensitive to bank risk variables. Our analysis also reveals that borrowing banks reduce their risk exposure after their first issuance of senior bank debt. We also find that lending banks significantly increase their collaboration with borrowing banks and increase their presence in the home markets of borrowing banks.

We show that borrowing firms benefit substantially from important enforcement actions issued on U.S. banks for safety and soundness reasons. Using hand-collected data on such actions from the main three U.S. regulators and syndicated loan deals over the years 1997–2014, we find that enforcement actions decrease the total cost of borrowing by approximately 22 basis points (or $4.6 million interest for the average loan). We attribute our finding to a competition-reputation effect that works over and above the lower risk of punished banks post-enforcement and survives in a number of sensitivity tests. We also find that this effect persists for approximately four years post-enforcement.