Multidimensional Non-Linear Pricing: Evidence from Energy Consumption with a Mixture Pricing Mechanism

In current business practice, sellers often combine a variety of pricing strategies to segment the market. These practices often leave consumers with pricing schedules that are highly non-linear and multidimensional. This paper develops a novel structural approach for estimating multidimensional non-linear pricing schemes with more than one decision variable. Using a rich and unique panel data of residents’ electricity consumption, we structurally estimate consumer preferences for electricity consumption in a setting with electricity tariffs, using both an Increasing Block Price (IBP) structure as well as a Time-of-Use (ToU) structure. Our model estimation allows us to identify and evaluate household-level elasticities in price elasticity of demand. Our counterfactual analysis shows that a seemingly welfare-improving policy may trigger the strategic behaviour of consumers and, ultimately, be contrary to the original purpose.