Abstract
Abstract
Global estimates indicate that low- and middle-income countries account for nearly 89% of the world's flood-exposed population, yet the relative importance of flood exposure and structural vulnerability in explaining regional welfare differences remains poorly understood. This study examined the associations and relative importance of population flood exposure and multidimensional structural vulnerability across 127 districts of Pakistan. District-level data were analysed using ordinary least squares regression with province fixed effects and Conley spatial heteroskedasticity and autocorrelation consistent (HAC) standard errors. Shapley value decomposition was used to quantify the relative importance of population flood exposure, six structural vulnerability domains, and long-term changes in temperature and rainfall in explaining two welfare outcomes: average district consumption and average consumption among poor households. Structural vulnerability was negatively associated with both welfare measures. For average district consumption, the six structural vulnerability domains collectively accounted for more than 90% of the explained variation, whereas population flood exposure contributed less than 1%. For consumption among poor households, long-term temperature change was the largest individual contributor, although the structural vulnerability domains collectively explained a substantially larger share of the variation. Overall, population flood exposure alone provided limited insight into regional welfare differences, while structural vulnerability emerged as the principal explanation. These findings suggest that integrating structural vulnerability alongside flood exposure may strengthen climate adaptation and disaster risk reduction planning.