Loss Aversion and Time-Differentiated Electricity Pricing

I develop a model of loss aversion over electricity expenditure, from which I derive testable predictions for household electricity consumption while on combination time-of-use (TOU) and critical peak pricing (CPP) plans. Testing these predictions results in evidence consistent with loss aversion: (1) spillover effects — positive expenditure shocks resulted in significantly more peak consumption reduction for several weeks thereafter; and (2) clustering — disproportionate probability of consuming such that expenditure would be equal between the TOU-CPP or standard flat-rate pricing structures. This behavior is inconsistent with a purely neoclassical utility model, and has important implications for application of time-differentiated electricity pricing.