The notion that economic crises induce the adoption of reform ranks among the most widely accepted concepts in the political economics literature. However, the underlying mechanism of the so-called 'crisis hypothesis' has yet to be fully understood. This paper provides a comprehensive survey of the relevant empirical evidence to date, and scrutinizes the operationalization of the hypothesis' key concepts: crisis, reform and the political mediation of reform during crises. We argue that the social perception of both crises and the subsequent cost of reform requires consideration of how these concepts are operationalized. As a product of the broader economic and institutional environment, social perceptions largely determine the manner in which the political mediation of reform during crises works. Present-day methodological approaches fail to adequately reflect social perceptions and consequently compromise the determination of what constitutes both crisis and the cost of reform in the context of the crisis hypothesis. Most notably, the identification of crises by fixed thresholds constructed around macroeconomic variables impedes the interpretation of the hypothesis' underlying mechanism. A fuller treatment of social perception within the operationalization of the hypothesis' key concepts can enhance our understanding of how economic crises influence political dynamics in bringing about reform. time, the impetus for deep, structural reforms of the financial system has faltered ' (Roubini and Mihm, 2011, p. 183).Rather than characterizing the repercussions of the 2008 crisis as not having been 'severe enough' to trigger reform, this statement indicates that the impetus for reform might depend more on a government's (in)ability to manage the immediate short-term consequences of a crisis. Further, it suggests that the determinants which facilitate the initial introduction and the eventual implementation of a reform agenda are not the same. This paper argues that the way the crisis hypothesis commonly operationalizes its key concepts -namely, crisis, reform and the political mediation surrounding the latter -makes it difficult to take such distinctions into account for empirical testing. Empirical modelling then risks conflating the distinct causal connections between these three key concepts, which in turn undermines conclusive interpretation of results.Among prominent theories of the political economy of reform, the correlation between crisis and reform has received great attention but has yet to be fully understood (see Brooks and Kurtz, 2007). The hypothesis has been repeatedly criticized for failing to adequately reflect the complexity of the mechanism that links crisis and reform (for example, Williamson, 1994;Corrales, 1998;Edwards and Steiner, 2000;Campos et al., 2010). In response to such critiques, empirical models testing the crisisreform link have become increasingly sophisticated, leading to the growing recognition that '[i]t is the type of crisis, and not just the existence of one, that is most crucial...