Revealed preference analysis of characteristics models

Characteristics models have been found to be useful in many areas of economics. However, their empirical implementation tends to rely heavily on functional form assumptions. In this paper we develop a revealed preference approach to characteristics models. We derive the necessary and sufficient empirical conditions under which data on the market behaviour of heterogeneous, price-taking consumers are non-parametrically consistent with the consumer characteristics model. Where these conditions hold, we show how information may be recovered on individual consumers' marginal valuations of product attributes. In some cases, marginal valuations are point identified, and in other cases, we can only recover bounds. Where the conditions fail, we highlight the role which the introduction of unobserved product attributes can play in rationalizing the data. We implement these ideas using consumer panel data on the Danish milk market.