UC Berkeley

In this dissertation I explore two aspects of the economics of energy. The first focuses on consumer behavior, while the second focuses on market structure and firm behavior.

In the first chapter, I demonstrate evidence of loss aversion in the behavior of households on two critical peak pricing experimental tariffs while participating in the California Statewide Pricing Pilot. I develop a model of loss aversion over electricity expenditure from which I derive two sets of testable predictions. First, I show that when there is a higher probability that a household is in the loss domain of their value function for the bill period, the more strongly they cut back peak consumption. Second, when prices are such that households are close to the kink in their value function - and would otherwise have expenditure skewed into the loss domain - I show evidence of disproportionate clustering at the kink. In essence this means that the occurrence of critical peak days did not only result in a reduction of peak consumption on that day, but also spilled over to further reduction of peak consumption on regular peak days for several weeks thereafter. This was similarly true when temperatures were high during high priced periods. This form of demand adjustment resulted in households experiencing bill-period expenditures equal to what they would have paid on the standard non-dynamic pricing tariff at a disproportionate rate. This higher number of bill periods with equal expenditure displaced bill periods in which they otherwise would have paid more than if they were on standard pricing.

In the second chapter, I explore the effects of two simultaneous changes in minimum energy efficiency and Energy Star standards for clothes washers. Adapting the Mussa and Rosen (1978) and Ronnen (1991) second-degree price discrimination model, I demonstrate that clothes washer prices and menus adjusted to the new standards in patterns consistent with a market in which firms had been price discriminating. In particular, I show evidence of discontinuous price drops at the time the standards were imposed, driven largely by mid low efficiency segments of the market. The price discrimination model predicts this result. On the other hand, under perfect competition, prices should increase for these market segments. Additionally, new models proliferated in the highest efficiency market segment following the standard changes. Finally, I show that firms appeared to use different adaptation strategies at the two instances of the standards changing.