Evaluating competing multifactor asset pricing models involves comparing the statistical significance of their mean pricing errors (alphas). Unfortunately, this comparison favors imprecisely estimated models because p-values tend to be higher in more noisy models. To avoid false impressions of relative success at tests for zero mean pricing errors, we develop a notion of comparative p-values and suggest comparing these instead of the raw p-values. This comparison gives more precisely estimated models a fairer chance or, equivalently, quantifies how much easier it is for imprecisely estimated models, by comparison, to pass the test.

A sovereign creditrating is a function of hard and soft information that should reflect the creditworthiness and the probability of default of a country. We propose an alternative characterisation for the subjective component of a sovereign credit rating – the parts related to the ratee’s lobbying effort or its familiarity from a United States point of view – and apply it to S&P, Moody’s and Fitch ratings, using both traditional ordered-logit panel models and machine learning techniques. This subjective component turns out to be large, especially for the low-rated countries. Countries that are rated as investment grade tend to be positively influenced by it, and vice versa. Subjective judgment in credit ratings does have predictive value: it helps in identifying chances of sovereign defaults in the short-term. Still, the impact of subjectivity in sovereign ratings on borrowing costs is very limited on average.