Information versus Automation and Implications for Dynamic Pricing

This paper studies whether information, communication or automation technologies can enable the consumer demand response necessary to expand dynamic pricing in resource markets such as electricity. Treatment effects from household random assignment to technologies and different price distributions reveal substantial demand reductions during critical and peak times for all technologies, with automation yielding the greatest reductions. Household random assignment, however, tests for and measures long-run elasticities, as opposed to the short-run price elasticities relevant to dynamic pricing. We illustrate how demand from a randomly assigned control group can be used to form a non-parametric control function estimate of demand that eliminates the bias that otherwise exists when estimating short-run demand curves with household random assignment. We find that information and communication technologies alone yield nearly perfectly inelastic short-run demand curves. Only consumers treated with automation technology exhibit short-run demand elasticities that can justify using short-run price variation to reduce peak or critical demand. Without automation, the longer-run conservation we observe may be more effectively initiated by messaging, rebates on durables, or permanent price changes as opposed to exposing consumers to temporary price shocks to which they are unable or unwilling to respond. Automation Technology, Demand Response; Short-run Elasticities; Utilities; Energy; Electricity; Field experiments.