Conference Agenda

The integration of renewable energy sources into the grid is increasing the value of domestic demand flexibility. This study evaluates the world’s first and largest domestic and commercial nationwide demand flexibility program. We analyzed this policy by randomizing 2.7 million residential and business customers to receive demand flexibility payments. We also randomized the incentive levels and staggered invitations. We found that these events led to a 25\% reduction in overall peak grid demand, with an average cost of £3,000 per MWh. We also found that reduction per opt-in was fairly consistent; variations in aggregate energy response mostly came through changes in customers' opt-in rate. Opt-in was relatively price-elastic, with a 140\% increase in opt-in for a doubling of payment per demand reduction. Notably, customers did not exhibit fatigue between and within event ``seasons''. The most significant reductions in peak demand were observed among households utilizing low-carbon technologies, such as solar panels, heat pumps, and batteries.

Energy-efficient cookstoves lower the marginal cost of food preparation, potentially leading to an increase in food consumption that increases the demand for cooking fuel. This paper assesses the magnitude of the resulting offset in energy savings, or rebound effect, by examining how exogenous shocks in stove technology affect energy and food consumption in a developing-country context. We analyse data from two randomized control trials that distributed energy-efficient biomass cookstoves to households in rural Rwanda and Senegal. In both samples, we find a statistically significant and substantial reduction in energy use among treated households. In the Senegalese sample, this reduction is offset by a moderate rebound effect, with point estimates ranging between 16 and 25%. In the Rwandan sample, conversely, we find a negative rebound that is of roughly the same magnitude but is statistically insignificant. We discuss mechanisms that potentially drive these differences and the implications for rebound research and energy access policy.

Analysis of many policy interventions depends on the availability of accurate estimates of price, income and other elasticities. Economists typically use revealed preference data to estimate these elasticities ex-post. We propose a simple survey-based contingent behavior approach for estimating ex-ante (expected) elasticities as well as their distributions in the population. Ex-ante and ex-post estimates of eleasticities may differ for legitimate reasons; both are relevant to policymakers. We illustrate this CB approach using data from seven Global South countries. Results show that it is possible to develop contemporaneous country-specific estimates of ex-ante elasticities quickly and inexpensively in data-poor environments. We also discuss how these ex-ante elasticities may inform policy acceptance.

JEL codes: C24, O13, Q41, Q5

Purchase decisions for energy-using durables are complex optimization problems. Low-income households face distinct financial and behavioral barriers that complicate the correct valuation of lifetime energy cost. Analyzing 4,600 purchase decisions from Germany’s largest Energy Efficiency Assistance Program, I show that households from the lowest income decile strongly undervalue lifetime energy cost when purchasing refrigerators. After the 2021 reform of the EU Energy Label, increased salience in differences in energy cost between models significantly increases valuation of energy cost. The results underline the relevance of behavioral barriers for energy-efficient investments in low-income consumers and the importance of a careful design of information instruments.