Vulnerability is a multidimensional concept incorporating notions of risk and poverty. While it has been established that higher incidence of poverty in developing countries exacerbates vulnerability, the role of risk requires closer inspection. Developmental interventions in these countries target poverty reduction, which in turn, could reduce vulnerability. However, a key question is whether developmental interventions reduce the vulnerability and risks faced by households. To answer this, the present study empirically examines the impact of developmental interventions on the vulnerability of households in a rural Indian setting. The major advantage hence is that it not only looks into the impact on aggregate vulnerability but also its different components such as poverty, covariate, idiosyncratic and unexplained risks. Empirical analysis is based on a survey of 800 households in the drought-prone villages of western Odisha, India, where a key developmental intervention, Western Orissa Rural Livelihood Project was implemented during the last decade. Adopting 'vulnerability as expected utility' approach, this study reveals three major findings. First, both aggregate risk and poverty are the dominant sources of vulnerability, with the former accounting for a sizable share. Second, the households that benefited from livelihood interventions are found less vulnerable. Third, the other major determinants of vulnerability are education, access to social network, family size and crop-diversification. From a policy perspective, results support continuation of these programmes, but realigning these also target risk reduction.