A frequently used metric associated with bank vulnerability is the non-performing loans ratio (NPLs), and many studies have analyzed and developed frameworks for forecasting this variable. The primary critique against many of the previous work is the failure to account for the nonlinearities inherent in the relationship between bank NPLs and the economic cycle. As a result the generalized conclusions on the linkage between NPLs and gross domestic product (GDP) may be misguided, especially when prudential rules impose natural asymmetries that may cause the association to change along different phases of the economic cycle. This study therefore, reexamines the relationship between NPL and GDP for Barbados using a regime switching approach. In doing so, the authors analyzed bankimpacts from a mix of macroeconomic and idiosyncratic data under different regimes to determine whether NPL outcomes across regimes and across institutions are homogenous. Further, the authors examined whether changes in systemic vulnerabilities are driven by all banks or only a subset. Answers to these critical problems, therefore provides a valuable framework that guides the authorities in developing appropriate policies to help strengthen the banking sector against systemic failures. The overall findings suggest that a non-linear approach to analyzing the NPL-GDP nexus is valid, but the relationship is not homogenous across institutions in different regimes. In addition, some institutions appear to be of greater systemic importance than others.