Working Paper: CEPR ID: DP6584
Authors: Sudipto Bhattacharya; Gabriella Chiesa
Abstract: We examine the implications of optimal credit risk transfer (CRT) for bank-loan monitoring. In the model, monitoring improves expected returns on bank loans, but the loan-portfolio return distribution fails to satisfy the Monotone-Likelihood-Ratio Property (MLRP) because monitoring is most valuable in downturns. We find that CRT enhances loan monitoring and expands financial intermediation, in contrast to the findings of the previous literature, and the reference asset for optimal CRT is the loan portfolio, in line with the preponderance of portfolio products. An important implication of optimal CRT is that it allows maximum capital leverage. The intuition is that the lack of MLRP makes debt financing suboptimal, so the bank is rewarded for good luck rather than for monitoring, and it faces a tighter constraint on outside finance: incentive-based lending capacity, given bank capital, is smaller. Optimal CRT exploits the information conveyed by loan portfolio outcomes to shift income from lucky states to those that are more informative about the monitoring effort. Thus, monitoring incentives are optimized and incentive-based lending capacity is maximized. The role for prudential regulation of banks is examined.
Keywords: credit risk transfer; monitoring; incentives; prudential regulation
JEL Codes: D61; D82; G21; G28
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Monitoring (M) (E40) | Expected returns on bank loans (ER) (G21) |
| Optimal credit risk transfer (OCRT) (G33) | Monitoring incentives (MI) (O31) |
| Monitoring incentives (MI) (O31) | Incentive-based lending capacity (ILC) (G21) |
| Debt financing (DF) (G32) | Tighter constraint on outside finance (COF) (G39) |
| Portfolio outcomes (PO) (G11) | Monitoring incentives (MI) (O31) |