The role of credit rating agencies (CRAs) in national and international capital markets has evolved over a period of time from an ‘information provider’ to various financial players to a ‘quasi-regulator’ of the market. This paper argues that the CRAs are ill suited to assume this role in the present form because of the ratings’ lack of accuracy in determining the creditworthiness of investments. This paper develops a game theory model to address the question: would an increase in competition among the CRAs improve the quality of the ratings? From the model, it is concluded that excessive competition in the ratings industry may not necessarily be healthy, as the reputational cost of getting labelled as dishonest decreases with the number of players. On the other hand, with only two or three dominant players in the market, the possibility of collusion among them increases, leading to the likelihood of inflationary ratings by the agencies. The model proposes that honest behaviour among CRAs can be achieved by ensuring an optimum number of players in the market, which prevents collusion, but at the same time, keeps each CRA accountable through a significant reputational cost.