This chapter studies calculative risk-assessment practices in credit rating agencies. It identifies two fundamentally different methodological approaches for producing ratings, which in turn shape the respective conceptions of credit risk. The traditional approach sees ‘risk’ as an only partially calculable and predictable set of hazards that should be avoided or minimized. This approach is particularly evident in the production of country credit ratings and gives rise to ordinal rankings of risk. By contrast, structured finance rating practices conceive of ‘risk’ as both fully calculable and controllable; they construct cardinal measures of risk by assuming that ontological uncertainty does not exist and that models can capture all possible events in a probabilistic manner. This assumption—that uncertainty can be turned into measurable risk—is a necessary precondition for structured finance securities and has become an influential imaginary in financial markets.
