For modeling the risk of address default over time, two ideal types must be distinguished within the framework of internal rating systems: Point-in-Time and Through-the-Cycle. The risk estimates of Point-in-Time rating systems fluctuate significantly over time depending on economic influences. In contrast, Through-the-Cycle systems aim for a relative stability of classification throughout the entire economic cycle. In many credit risk models, the rating data is further processed. The cyclicality of these models thus also depends on the type of rating input chosen. How can a bank take these cyclical effects into account in its risk management? Starting from the question of how fluctuations over time pose a problem for credit risk measurement, the relationship between credit risk and economic development is examined, and the two rating paradigms are characterized.