A Model of Endogenous Loan Quality and the Collapse of the Shadow Banking System

I develop a macroeconomic model with a financial sector, in which banks can finance risky projects (loans) and can affect their quality by exerting a costly screening effort. Informational frictions regarding the observability of loan characteristics limit the amount of external funds that banks can raise. In this framework I consider two possible types of financial intermediation, traditional banking (TB) and shadow banking (SB), differing in the level of diversification across projects. In particular, shadow banks, by pooling different loans, improve on the diversification of their idiosyncratic risk and increase the marketability of their assets. Due to their ability to pledge a larger share of the return on their projects, shadow banks will have a higher endogenous leverage compared to traditional banks, despite choosing a lower screening level. As a result, on the one hand, the introduction of SB will imply a higher amount of capital intermediated. On the other han d it will make the economy more fragile via three channels. First, by being highly leveraged and more exposed to risky projects, shadow banks will amplify exogenous negative shocks. Second, during a recession, the quality of projects intermediated by shadow banks will endogenously deteriorate even further, causing a slower recovery of the financial sector. A final source of instability is that the SB-system will be vulnerable to runs. When a run occurs, shadow banks will have to sell their assets to traditional banks, and this fire sale, because of the limited leverage capacity of the TB-system, will depress asset prices, making the run self-fulfilling and negatively affecting investment. In this framework I study how central bank credit intermediation helps reduce the impact of a crisis and the likelihood of a run.