Résumé : Do individuals divorce for economic reasons? Can we measure theattractiveness of new matches in the marriage market? We answer thesequestions using a structural model of the household and a rich panel datasetfrom Malawi. We propose a model of the household with consumption,production and revealed preference conditions for stability on the marriagemarket. We define marital instability in terms of the consumption gains toremarrying another individual in the same marriage market, and to beingsingle. We find that a 1 percentage point increase in the wife's estimatedconsumption gains from remarriage is significantly associated with a 0.6percentage point increase in divorce probability in the next three years. Ina multinomial model, higher values of consumption gains from remarriageraise the odds of divorce and remarriage but not of divorce and singleness.These findings provide out-of-sample validation of the structural model andshed new light on the economic determinants of divorce.