Three essays on banking and asset pricing

Chapter 1 of my Thesis studies in a cross-sectional analysis lenders' beliefs. The coauthor and I use a novel loan-level dataset containing borrower-specific probability of default that allows to measure accurately lenders' expectations. We found our empirical analysis on a learning model where bankers endowed with diagnostic expectations observe noisy fundamentals from firms and estimate their probability of default. We provide empirical evidence that financial institutions are subject to expectational distortions: banks tend to overreact to both micro and macro news, overestimating (underestimating) borrowers' defaults after negative (positive) signals. We also document that the degree of overreaction is quite heterogenous among banks. In addition, overreacting bankers decrease (increase) interest rates more than rational ones and the probability of issuing a new loan rises (fall) in light of positive (negative) news. We confirm these results with a structural estimation exercise departing from a model of banking competition where banks' profit function depends on borrowers' creditworthiness, driven by the level of banks' expectation distortion and firm-specific economic news. In Chapter 2 I develop a structural model of loan demand and lender competition to study how transition risk may affect the Italian credit market. First, I show that transition risk is not currently priced by banks, nor that firms likely more exposed to this risk tend to default more frequently. Then, I use the estimated model to study the effect of policies aimed at more tightly integrating climate-related and environmental risks into banks' business planning. Modeling any such policy as in increase in the cost of lending to ``brown'' firms, counterfactual analyses show that if these marginal costs were to increase by one standard deviation interest rates would on average increase by 130 basis points, while quantities would decrease by about 20k EUR. In Chapter 3 together with coauthors I quantify the exposure of major financial markets to news shocks about global contagion risk accounting for local epidemic conditions. For a wide cross section of countries, we construct a novel dataset comprising (i) announcements related to COVID19 and (ii) high-frequency data on epidemic news diffused through Twitter. Across several financial assets, we provide novel empirical evidence about financial dynamics both around epidemic announcements and at daily/intra-daily frequency. Contagion data and social media activity about COVID19 suggest that the market price of contagion risk is significant. Hence policies that mitigate global contagion or local diffusion may be extremely valuable.