Estimates of the number of people living in extreme poverty, as reported by the World Bank, figure prominently in international development dialogue and policy. An assumption underpinning these poverty counts is that there are no economies of scale in household size - a family of six needs three times as much as a family of two. This paper examines the sensitivity of global estimates of extreme poverty to changing this assumption. The analysis rests on nationally representative household surveys from 162 countries covering 98 percent of the population estimated to be in extreme poverty in 2017. We compare current-method estimates with a constant- elasticity scale adjustment that divides total household consumption or income not by household size but by the square-root of household size. While the regional profile of extreme poverty is robust to this change, the determination of who is poor changes substantially - the poverty status of 270 million people changes. We then show that the measure which accounts for economies of scale is significantly more correlated with a set of presumed poverty covariates (i.e., years of schooling, literacy, asset index, working in agriculture, access to electricity, piped drinking water, improved sanitation).