The widely held view that larger families tend to be poorer in developing countries has influenced research and policies. But the basis for this"stylized fact"is questionable, the authors argue. Widely cited evidence of a strong negative correlation between size and consumption per person is unconvincing, given that even poor households face economies of size in consumption. The authors find that the correlation between poverty and household size vanishes in Pakistan when the size elasticity of the cost of living is about 0.6. This turns out to be the elasticity implied by a modified version of the food-share method of setting scales. By contrast, some measures of child nutritional status indicate an elasticity closer to unity. Consideration of the weight attached to child versus adult welfare may help resolve the nonrobustness of demographic profiles of poverty. The authors show that the incidence of severe child stunting is more elastic to household size than their Engel curve estimate suggests, although the latter is still a fair predictor of child wasting. A consideration of the purpose of measuring poverty - notably the extent to which it is used to inform policies aimed at promoting child welfare - may go some way toward resolving the issues.