Abstract:
Energy efficient technologies are a major tool for reducing electricity usage as well as
greenhouse gas emissions. Nevertheless, serious market barriers impede the diffusion of these
solutions. Many utilities offer programs and incentives to surmount these barriers. However,
few empirical studies exist that quantitatively document the impacts of such initiatives. This
project investigates a utility rebate program aimed at encouraging the adoption of domestic solar
hot water (DSHW) systems in northeast Florida residences. Four years of household
consumption, temperature, and one year of block group level demographic data is analyzed using
multiple and multilevel, or hierarchical, regression analysis techniques. Four estimators are
developed to estimate the causal impact of the program. The first estimator utilizes a random
selection of non-participating households within the utility’s service territory and a subset of
participating households for that year to establish control and treatment groups. The remaining
estimators represent a random effects model, a fixed effects model, and a mediating multilevel
model of the causal effect of the program for the years 2003 through 2006, respectively. These
latter models derive control and treatment groups from pre- and post-treatment cohorts of
program participants. Each estimator exhibits particular strengths and weaknesses. In aggregate,
they provide a bound for the average impact of DSHW technology on monthly household
electricity consumption. The results of this analysis indicate a direct causal relationship between
the adoption of DSHW and energy savings. In particular, it estimates a statistically significant
reduction in monthly electricity consumption ranging from 71 to 118 kWh for the average
household, corresponding to 3-8% of total usage. Finally, economic and environmental policy
implications of these results are considered and indicate that financial incentives as well as other
strategies are necessary to aid the diffusion of DSHW technologies.
