An experimental investigation of rating-market regulation

We introduce a simple game-theoretical model that captures the main aspects of the repeated interaction between an issuer and a credit rating agency. The scenario is characterized by up-front payments of issuer-fees and regulatory sanctions for false rating. We chose parameters such that in the Bayesian Nash equilibrium the credit rating agency should always provide truthful ratings. Knowing this, the issuer should never request a rating. Conducting laboratory experiments, we find that issuers frequently request ratings, which in turn is reciprocated with a high proportion of untruthful “good” ratings, even though the credit rating agency faces (low or high) financial penalties for being untruthful. Our results are different from the game-theoretical prediction but they are in keeping with a “cooperative solution”, similar to the “deterrence theory” in Reinard Selten’s “chain store paradox” (Selten, 1978).