WP-010: Koichiro Ito, "Asymmetric Incentives in Subsidies: Evidence from a Large-Scale Electricity Rebate Program" (December 2014)

To motivate conservation, many policymakers favor implementing subsidy programs to provide a financial incentive for customers to reduce their energy consumption. This paper analyses the impact of California's 2005 residential 20/20 electricity conservation program. The 20/20 program offered a rebate to households that reduced their 2005 summer electricity consumption relative to their summer 2004 consumption. Eligibility for the program was automatic if a household had begun its electricity service by a certain date in 2004. Professor Ito was able to use this strict eligibility cutoff to compare the energy consumption of households who just barely met the eligibility cutoff with those who just barely missed the cutoff. Both sets of households were substantially similar to one another so that the effect he measures can be confidently attributed to the program. The research design allows him to exclude the impact of those who reduced their consumption for reasons not related to the rebate, such as changes in the number of people in a household from year to year. 

Professor Ito finds that the rebate incentive reduced electricity consumption by 4% in the inland areas and caused no reduction in the coastal areas of California. Income, climate and air conditioner saturation significantly drive the difference in response. The inland areas face higher summer temperatures and the income levels are relatively low, both factors that can lead households to be more responsive to the rebate program. In addition, Ito finds that the asymmetric structure of the rebates weakens the incentives for households to respond. The analysis shows that those households that were further from the rebate target showed little response. There was in effect a "giving up" response  - a household was less likely to reduce its consumption when the rebate target looked unattainable. 

In calculating the cost-effectiveness of the program, Ito finds that the overall program cost is 17.5 cents per kWh reduced and $381 per ton of carbon dioxide reduced. However, the cost-effectiveness differs substantially between the coastal and inland areas. Because the program did not induce significant reductions in electricity in the coastal areas, the cost per kWh reduced is quite high - 94.5 cents. Since the inland areas did conserve a lot more electricity, the cost per kWh reduced was 2.5 cents. These results suggest that one way to improve the cost effectiveness of this rebate program is to target lower-income households and households in areas with higher summer temperatures because they are more likely to respond to the rebate.