When Does Static Willingness to Pay Mislead? A Framework for Dynamic Hedonic Valuation

Many policy counterfactuals depend on how consumers value product attributes such as sugar, caffeine, alcohol, or emissions. Standard hedonic and differentiated-products models typically impose time-separable preferences. But when attributes are habit forming, current consumption can shift future marginal valuations, so static willingness-to-pay may be insufficient for policy counterfactuals. I develop a nonparametric revealed-preference framework for dynamic hedonic valuation, deriving necessary and sufficient conditions for rationalising observed prices and choices. Using cereal scanner data, I show that the hedonic representation restricts prices, while habits improve behavioural coherence conditional on that representation. The framework diagnoses when static willingness-to-pay is defensible for policy.