Credit risk models are shown to play a key part in the global credit crisis. We discuss how the credit market has exposed the shortcomings of the credit risk models and we identify their main shortcomings. To overcome the shortcomings, a new causal framework is proposed to build deductive credit default models which have predictive capabilities. The global credit crisis is the failure of major parts of the credit market to function normally, leading to failures of financial institutions because they were unable to obtain funding from the market. The first part of the market to fail was the US sub-prime mortgage market. Subsequent failures of other parts of the market suggest that the problems are more fundamental than simply the mispricing of some mortgage-backed securities.

As there has been no identifiable external shock to the financial system, the crisis appears to have an endogenous origin which could have multiple systemic causes associated with the various steps of the credit process developing over time. Undoubtedly many of the “usual suspects” such as excessive credit growth, lax lending standards, conflicts of interest, complex financial instruments etc. all play some part. In this paper, we consider the key part played by the pricing3 of credit risk. We discuss how the credit market has exposed the shortcomings of the credit risk models. We discuss the main shortcomings and how to overcome them.