The author examines the potential trade-offs that may arise between poverty alleviation and unemployment reduction. He discusses various analytical arguments that may provide a rationale for their existence, and uses three alternative methodologies to assess their relevance: a vector autoregression framework (which is applied to Brazil and Chile), cross-country regressions, and simulations with a structural macro model linked to a household survey. Impulse response functions to output and wage shocks indicate no short-run tradeoff. between unemployment and poverty. By contrast, regression results, which control for a variety of determinants of poverty rates across countries, suggest that such a trade-off may indeed exist. Simulations with the structural model show that labor market reforms may induce both short- and long-run trade-offs between the composition of unemployment and the incidence of poverty among household groups.