Conference Agenda

I examine rebound effects in electricity consumption induced by solar photovoltaic (PV) adop- tion using detailed panel data of 60,000 Swiss single family home residents (2008-2019). I find that solar PV adoption increases a household’s electricity consumption by an average of around 8% post-adoption. The decomposition of the rebound effect, using machine learning based counterfactual prediction, illustrates that household fuel switching accounts for part of this effect. This manifests through household electrification, such as electric vehicle adoption. Further results suggest that rebound effects are mainly driven by a sub-sample of solar PV households that adopt relatively large installations and adjust their consumption profile drastically.

Are budget constraints and inattention relevant in explaining the energy efficiency gap in low-income households? The lifetime energy cost of energy-using durables accounts for a significant share in total cost but is not salient to consumers at the time of purchase. Moreover, more energy-efficient durables with lower lifetime energy cost may not be affordable for budget-constraint customers. Using 20,601 investment decisions in household appliances by low-income households, I examine how these households trade off the lifetime energy cost against the purchase price and investigate the role of budget constraints and inattention in the trade-off. My analysis exploits the effect of exogenous changes in the budget constraint and salience of appliance energy efficiency. I find that households are far from fully considering lifetime energy cost in their investment decision, and both budget constraints and inattention are relevant in explaining the result. The findings are informative for distributional consequences of different policy instruments.

This paper estimates the effect of distributed solar generation on distribution feeders, utilizing unique, proprietary data from individual solar installations and feeders in Connecticut. We find that each additional kilowatt (kW) of distributed solar reduces the annual maximum feeder load and the top one percentile of the load by 0.11 kW. In addition, we find evidence of a 3.6% solar rebound effect, primarily occurring in spring and fall, such that it does not affect the impact of distributed solar on peak feeder loading. The economic value of avoided distribution capacity ranges from $1.6 to $3.1 per MWh, significantly below the cost premium of distributed solar compared to utility-scale solar. Our estimate is an important input to policies for optimal solar investment and to the major debate on replacing solar net-metering policies.

This paper examines household intentions and beliefs in response to higher expected energy costs in Germany. We conduct a tailor-made household survey during the energy crisis of 2022 that led to significantly higher prices for residential heating in Germany. On average, households anticipated a 70 percent increase in energy costs on their next heating bill. In response to higher cost shocks, households planned to reduce their heat demand but were less inclined to view long-term investments as the most effective or practical way to lower energy costs. This indicates that investing in energy efficiency or switching heating fuel types did not become more appealing as coping mechanisms in the aftermath of the crisis.