Essays on credit ratings
Through this dissertation I aim to extend and, at the same time, challenge, the existing research on credit ratings, which has been increasing exponentially since the Credit Rating Agencies (CRAs) have been considered as one of the main culprits in the Financial Crisis of 2009. In the first paper, I investigate whether credit ratings can measure financial constraints better than the existing (indirect) proxies, and I find that this is the case. The second essay suggests that credit ratings can trigger significant market reactions even if are unrelated to an actual change in issuer credit risk, meaning that there is a "labeling" effect. Third, I show that sovereign ratings have become more conservative over time, i.e. a country rated AA in 1990 would only receive BBB+ (2 notches less) in 2018, even if it has exactly the same risk as in 1990. In the final paper, I question the interpretation of year trend as "rating conservatism", and, by using a 2-step approach, i show that the original method overestimates the actual magnitude of "conservatism", since most variables exhibit a common secular trend.