Residential access to modern energy and lighting is considered to be important for development as it improves living standards and productivity. However, developing countries can face severe constraints on available electricity. As a consequence, residential electricity use is limited and households pay prices that can be as high as those observed in wealthier nations. In the developing world, where lighting is a major component of residential electricity consumption, energy efficient technologies such as compact fluorescent lamps (CFLs) can offer electricity savings and allow more energy services. Still, where CFLs have been introduced in developing countries through mass replacement programs at a zero or subsidized price, their use is often discontinued once the consumer is required to pay the full price. This represents a major constraint to realizing the potential impacts of energy efficient lighting.
To better understand the adoption, continued use, and diffusion process of the CFL technology, this study uses randomized experiments to inspect two major mechanisms: (1) the ?rebound? effect, which is the result of behavioral responses that boost consumption and offset CFLs? technologically feasible electricity savings, and (2) the role of ?peer? networks in spreading information on a new technology. Although the initial adoption of CFLs may be limited because the willingness to pay (WTP) is lower than the market price, their diffusion may be restricted if information is too noisy or if a greater importance is put on negative rather than positive information.
The study furthers knowledge on the behavioral barriers that arise from end-user interactions with energy-efficient technologies. The findings are relevant to several aspects of energy conservation policy. Examination of the rebound will inform the optimal combination of innovation and price policies and the trade-off between welfare and electricity conservation. Examination of peer- networks will discern between different learning mechanisms and tell how best to advance CFLs? diffusion. Lastly, WTP studies assist in developing strategies for optimal pricing and subsidies. Beyond the policy relevance for CFL programs deployed recently in developing countries, lessons learned regarding responses to energy efficient technologies have behavioral implications that are also relevant to developed countries.
Introduction Residential access to modern energy and lighting is important for development in that it improves living standards and productivity (World Bank, 2006). However, developing countries often face severe constraints on available electricity, which result in blackouts. In the developing world, where lighting is a major component of residential electricity consumption, energy efficient technologies, such as compact fluorescent lamps (CFLs), can reduce overall household electricity demand while permitting utilities to simultaneously reach a greater number of customers with existing supplies and meeting conservation goals. Still, when CFLs have been introduced through mass replacement programs at a zero or subsidized price, their use is often discontinued once the consumer is required to pay the full price. This represents a major constraint to realizing the potential impacts of energy efficient lighting. In order to better understand the adoption, continued use and diffusion process of the CFL technology, the study uses randomized experiments to inspect two major mechanisms: (1) the "rebound" or behavioral responses that boost consumption and offset CFLs’ technologically feasible electricity savings and (2) the role of "peer" networks in spreading information, including whether the information source is too noisy or if negative information on outcomes, quality, and maintenance requirements weighs more heavily or spreads faster than positive information. As initial take up of CFL lamps may be limited because the willingness to pay (WTP) for them is lower than the market price, WTP must be elicited to study these two mechanisms. Randomized Design Three randomized experiments were conducted to assess, respectively, the savings and rebound effects in energy consumption brought by CFL use, the peer effects on the dissemination of efficient CFL lighting technologies, and the demand and willingness to pay (WTP) for CFLs. The study was implemented near Bishkek, the capital of Kyrgyzstan. Using the electricity utility’s dataset of approximately 40,000 households, 110 neighborhoods of households were randomly selected to receive differing intensities of treated households. Within these neighborhoods, 1,000 households were randomly selected to participate in the study. Of the entire selected sample, households were randomly assigned to one of three groups: the control group (457 households), the group receiving the light bulbs for free (280 households), and the group receiving the opportunity to purchase the light bulbs at a randomly drawn price (263 households). Baseline data were collected for all 1,000 households through an extensive household survey. Light bulbs were distributed to those households eligible to receive the energy efficient technology either for free or by purchase (543 households in total). Based on the data collected in the baseline household survey, household characteristics are compared across all three groups. Preliminary analyses show there are very few statistically significant differences between the groups, indicating that the randomization process worked and the sample is balanced. Moreover, no pre-trends and no pre-intervention differences in electricity consumption between treatment and control groups were identified. Demand for Light Bulbs To measure household willingness-to-pay (WTP), randomized experiments provide a rigorous alternative to contingent valuation and other non-experimental approaches. Experiments use actual decisions in real market situations where individuals face real benefits and budget constraints. We conduct a randomized experiment to measure the household demand for one, two, three and four CFLs. Preliminary analysis of the correlations between the per unit prices that households bid for light bulbs and baseline household characteristics indicates that certain household characteristics are positively correlated with the per unit bids for light bulbs. Years of education of adults and households head, total income, household expenditures, and number of rooms and light bulbs in the household are all positively correlated with the household per light bulb bid. The results can inform government programs promoting energy efficiency in developing strategies for optimal pricing and subsidies as well as characteristics of households to target. Behavioral Response to CFLs According to preliminary analysis, households that received free-bulbs have significantly lower average consumption post-intervention than control groups. No significant impact on electricity consumption is found among households that purchased the CFLs. However, when looking at average quarterly consumption post-intervention a more complex pattern arises. The results are unlikely to be explained by the existence of pre-intervention trends. The impact of treatment assignment on pre-intervention electricity consumption is not significant. The estimated coefficients become significant only in the post-intervention period as the variance in electricity consumption declines for both treatment groups. Peer effects To study the role of peers in household decision-making, the project exploits two types of variation in density of informational networks. These results will be complete after the final round of household surveys is completed in early 2014. This analysis provides a better understanding of how best to advance their diffusion and make such efforts more cost-effective.
