Canadian household prescription drug expenditures are studied using different years of the Statistics Canada Family Expenditure Survey. Master files are used, expanding the number of available years and permitting provincial rather than regional identifiers. Nonparametric Engel curves are estimated. Difference-in-difference mean and 80th percentile regressions examine budget shares by low-income and high-income households before and after the introduction of provincial prescription drug programs. The evidence is consistent with the view that unlike senior prescription drug subsidies, nonsenior prescription drug subsidies are probably more redistributive than an equal-cost proportional income transfer.