Good Crises Wasted? The Effects of Crises on Risk Mitigation Through International Organizations

How do global crises affect international cooperation through international organizations (IOs)? While some theories predict that crises generate solidarity and spur institutional strengthening, others expect them to fuel blame, retrenchment, and declining multilateralism. Despite the centrality of crises in contemporary global governance, systematic empirical evidence on how global shocks reshape states’ willingness to resource IOs remains scarce. This paper provides the first empirical analysis of how international crises affect investments in the capabilities of risk-mitigating IOs. We develop an integrated theoretical framework that combines insights from crisis management and international relations. We explain why crises may produce divergent effects across time horizons and levels of aggregation. At the national level, crises increase risk salience and political demand for protection, which can stimulate investments in collective risk management. At the same time, crises also heighten political contestation, blame, and demands for control over international institutions, which can reduce willingness to provide them with flexible resources and autonomy. We therefore distinguish between short- versus medium-term effects and between IO-wide core funding and voluntary, earmarked contributions by affected states, which differ in the degree of discretion and control they grant to member governments. We test our expectations using two original datasets. At the IO level, we track annual budgets and staffing levels of three major global risk-managing organizations from their founding after World War II through 2020. We match these data to domain-specific measures of global crises in public health, finance, and nuclear safety, along with macroeconomic controls and institutional fixed effects. At the country level, we assemble a panel of advanced industrialized states from 1990–2020 that combines detailed data on voluntary and assessed contributions to the WHO with country-specific exposure to severe epidemic mortality, as well as economic, demographic, and political controls. We estimate both panel OLS models and dynamic local projection models to capture the timing and persistence of crisis effects. Across specifications, we find a striking and robust pattern. At the IO level, crises systematically depress the growth of core budgets and staff. This negative effect emerges quickly and intensifies over time, peaking roughly three to five years after a crisis, when IOs experience reductions in annual budget growth of one to two percentage points—large relative to their historical growth rates. These results indicate that crises are followed by sustained constraints on IO-wide, collectively financed capacity. At the same time, affected states respond in the opposite direction in the short run. Countries that experience severe crises increase their voluntary, earmarked contributions to the relevant IO in the immediate aftermath. These increases are large—often amounting to a substantial share of normal annual contributions—but short-lived, fading after roughly two years. Over the medium term, even total contributions (including assessed fees) show signs of decline. Together, these results reveal a systematic divergence: crises generate bursts of targeted, discretionary cooperation by affected states while simultaneously undermining IO-wide core funding. We argue this reflects a post-crisis political economy in which governments seek solutions but resist granting IOs flexible, autonomous resources.