In recent years, the discussion about welfare state reform has often focused on the effectiveness of social security schemes. This debate is torn between calls for more effective poverty alleviation on the one hand and concerns about welfare dependency and the need for stronger targeting of social security benefits on the other. Means-tested benefits have been a major subject in this debate, since they are explicitly directed toward low-income groups. Basically, means-tested benefits are characterized by income and/or asset tests (examples are social assistance or housing benefits in most countries). In spite of the growing concern regarding the effectiveness of means-tested benefits in alleviating poverty, our knowledge on the impact of means-tested benefits remains inadequate. Although a number of studies have assessed the effects of tax and transfer systems as a whole and have provided accounts of the effectiveness of specific social security schemes or the impact of redistribution on specific groups of the population, evidence on the effectiveness of means-tested benefits is still limited. We know astonishingly little about the effects of means-tested transfers on poverty. How effective are they in reducing poverty rates? To what degree can they fill the gap that income redistribution through taxes and non-means-tested social security benefits leaves? How large is the impact of means-tested transfers in private household budgets, and how does the role of means-tested benefits vary across countries? The Luxembourg Income Study (LIS) offers detailed micro-data of private household incomes and facilitates the analysis of the impact of means-tested transfers in a comparative perspective. This paper will explore the potential of these data for assessing the effectiveness of means-tested benefits. Three countries have been chosen for this purpose: Germany, Sweden and the United Kingdom. The evaluation of the effectiveness of means-tested benefits in these countries addresses variations in the effectiveness of means-tested benefits across countries. Can we identify specific patterns in the impact of this kind of benefits? How can we relate these patterns to the institutional settings found in these countries?