Prescription drug use under Medicare Part D: A linear model of nonlinear budget sets

Medicare Part D enrollees face a complicated decision: they dynamically choose prescription drug consumption in each period given difficult-to-find prices and a nonlinear budget set. We use Part D claims data to estimate a flexible model of consumption that accounts for nonlinear prices, dynamic responses, and salience. We use reduced form price responses from a linear regression of consumption on coverage range prices to compare performance under several models of behavior. We find small price elasticities, substantial myopia, and that salient characteristics impact consumption beyond their effect on prices. A hyperbolic discounting model with salience fits the data best.