In this paper, we examine how various financial vulnerabilities evolve in the lead-up to and in the aftermath of banking crises in various advanced and emerging economies. We develop a holistic framework to track financial imbalances that may render the financial system highly vulnerable to shocks to the economy. Our paper belongs to the strand of the academic literature on financial imbalances, financial crises, and systemic risk that has been brought to the forefront by the Global Financial Crisis (GFC). The GFC, which began as banking crises in the United States and the United Kingdom in 2007, ended up quickly spreading to other financial systems around the world. This experience has profoundly changed the global financial regulatory landscape. Central banks and other official institutions, in turn, have established various tools and early warning indicators to monitor financial stability risks.1 Our paper draws from these advancements to put together a comprehensive early warning indicator that covers multiple areas where vulnerabilities can build up, that captures potential spillover and amplification channels of vulnerabilities, and that predicts the onset and severity of banking crises. We posit a view that the advent of a financial crisis can be decomposed into a financial vulnerability or imbalances component and a shock component (as in Gorton and Ordonez (2014)). Understanding how financial vulnerabilities and imbalances evolve in the run-up to a banking crisis provides a better framework to understand the role that the first component plays in the realization of banking crises. Building upon research on how different types of vulnerabilities in the financial system set the stage for an unwinding or dramatic unraveling of financial imbalances (Ferguson, Hartmann, Panetta, and Portes (2007), Reinhart and Rogoff (2009), and Aikman, Kiley, Lee, Palumbo, and Warusawitharana (2017)), our aim is to shed light on whether both the occurrence and severity of banking crises are correlated with the level of vulnerabilities present in the financial system prior to banking crises. We extend the framework of in Aikman, Kiley, Lee, Palumbo, and Warusawitharana (۲۰۱۷) that maps vulnerabilities in the U.S. financial system to a broader set of 27 advanced and emerging economies. The key contribution of Aikman, Kiley, Lee, Palumbo, and Warusawitharana (2017) was to develop an algorithmic approach which uses a large set of indicators to monitor vulnerabilities that can identify imbalances in the U.S. financial system. Because of banking crises in the United States have been infrequent, Aikman, Kiley, Lee, Palumbo, and Warusawitharana (2017) could not formally test the predictive power of their index with respect to banking crises. They provided only a narrative. In contrast, since we look at a broader set of vulnerabilities for a panel of nearly 30 countries, some of which have experienced multiple banking crises, we can determine the predictive power of our vulnerability index that is derived from a bottom-up holistic framework. That is, we can establish the power of such an indicator to predict the timing of a banking crisis and the severity and duration of a recession that follows. In addition, we can compare our findings with the performance of the credit-to-GDP gap (CGG), which has been touted as one of the best predictors of systemic banking crises at longer horizons and, hence, is argued to be the benchmark in setting counter-cyclical capital buffers (see Drehmann and Juselius (2014)). We look at how the different vulnerability measures compare when predicting banking crises as defined in Laeven and Valencia (2013), in addition to systemic crises as defined in Drehmann and Juselius (2014).