I’d Like toMove It! Consumption Rivalry in the EV Public ChargingMarket: Demand Estimation with Deterministic Choice Set Variation

Abstract

Consumption rivalry generates variation in the choice sets decision-makers face. Neglecting such variation may bias demand estimates. To correct this bias, researchers have used aggregated data on stockouts and information from periodic inventory systems to impose probabilistic constraints on the choice set.

I use new data from a real-time inventory system that records all public charging sessions of electric vehicles. Per transaction, this identifies a user's full set of available alternatives. Including this information as deterministic choice set constraints significantly improves demand forecasts for local charging facilities. Estimated differences between observed and latent demand reach up to 40 percentage points.

Keywords: Electric Vehicles, Discrete choice, Preference estimation, Consumption rivalry, Policy Design

JEL Classification: H42, H54, L62, Q41, Q48