The important factors that determine the living standards of adults are labor force participation, living arrangements, and health status. For the elderly, lifetime asset accumulation is another key determinant of living standards in old age. In a majority of less developed countries (LDCs), especially in the rural areas, asset accumulation levels of the elderly are very low. Employer-based or public pension income is accessible to only a tiny percentage of the elderly. A crucial aspect of empirical and policy interest is how do rural LDC elderly obtain economic welfare in the absence of private asset accumulation or formal public programs. A majority of the elderly continue to work in very old age, supplying labor in the market, working in family-owned enterprises, or participating in domestic duties. A majority of the rural LDC elderly, especially in South Asia, also live in extended households, which typically include their children, and the childrenv¢s families. How this living arrangement impacts the elderly has been examined only sparingly. Indeed, national policy on older persons in India, for instance, has implicitly assumed that the extended household works well in delivering old age support. In light of these facts, the current study presents an empirical economic profile of rural elderly in Northern India by examining their labor supply in detail. The empirical study carries out three tasks. Elderly labor supply is estimated via a structural framework that is robust to the endogeneity of marginal productivity, and includes a wide range of demographic, household-level, and infrastructure covariates. The framework is used to examine the extended household as an old age support mechanism. A key contribution of this study is that rural work behavior is observed separately for wage-based labor, production on household farms, and household-owned business enterprises. Secondly, the study attempts to situate the labor supply analysis in the larger context of intrahousehold economics. A specific question asked is, do rural elderly work out of preference? Or do they engage in labor activities primarily compelled by poverty? This question is interpreted as a ``poverty hypothesisv¢v¢ for elderly labor supply. The hypothesis was originally developed in the economics literature on LDC child labor. Congruent with child labor, a general form of the hypothesis states that the householdv¢s poverty compels the elderly to work. The corresponding empirical implication is that the own-wage elasticity of the elderly will be negative. Finally, labor supply in domestic work, and other non-market work activities, is included in the analysis. Reduced-form estimates are derived for the labor participation decision in any type of work. The empirical findings suggest that living with adult children does not have statistically significant influences on the hours supplied by the elderly in wage-based labor. The non-elderly working coresiding members are, on the other hand, predicted to work more wage hours if elderly are present in the household. The own-wage elasticity of annual hours of elderly labor supply is estimated to be negative for wage work. Since the estimates are structurally robust, this can be suggestive of a poverty hypothesis for elderly in rural North India. The reduced-form model for labor participation predicts that elderly men coresiding with at least one adult son and daughter-in-law are less likely to choose to work. Elderly women coresiding with adult daughters are less likely to be working. The elderly not coresiding with adult children are less likely to be working when small children and young teenagers are present in the household. The empirical results provide strong support for formal social security programs, and for elderly-targeted policies to be combined with rural poverty alleviation and employment generation.