Lottery‐demand models using aggregate data are often used to make inferences regarding individual behavior, the most important being the distributional burden of lottery‐ticket expenditures. It is shown here that estimates for the income elasticity and the cross‐price elasticity will only be representative of individual behavior under extremely restrictive assumptions. In fact, estimation of aggregate‐demand models presupposes that the income elasticity is equal to one. Cross‐sectional analyses using microlevel data face similar restrictions on consumer behavior. Remedies are discussed, but more conclusive evidence on the distributional burden of lotteries will remain elusive until better individual‐level data become available. (JEL D11, H71, H22)