Despite significant spending on ‘resilience’ by international development agencies, no theory-based method for estimating or measuring development resilience has yet been developed. This paper introduces an econometric strategy for estimating individual or household-level development resilience from panel data. Estimation of multiple conditional moments of a welfare function—itself specified to permit potentially nonlinear path dynamics—enables the computation and forecasting of individual-specific conditional probabilities of satisfying a normative minimum standard of living. We then develop a decomposable resilience measure that enables aggregation of the individualspecific estimates to targetable subpopulation- and population-level measures. We illustrate the method empirically using household panel data from pastoralist communities in northern Kenya. The results demonstrate the method and its potential for targeting resilience-building interventions.