Household electricity consumption behavior: a meta-analysis and experimental approaches.

Summary This thesis examines how consumers respond to mechanisms to reduce their energy consumption. This need for reduction stems from the need to meet greenhouse gas emission reduction targets, to increase energy production from renewable energy sources, and to achieve energy savings. These goals require residential demand to be more flexible in the face of changing supply and for energy savings to be achieved by households. The first chapter explores the barriers to acceptance and adoption of smart meters and the incentives they can provide. Significant barriers exist and consumption reductions are far from being realized. Lack of motivation, lack of understanding of consumption information, and inflexibility of daily life are the main barriers that limit household response to the incentives provided by smart meters. The second chapter analyzes the results of field experiments and pilot studies on the impacts of different incentives on residential consumption. The results show that there is wide variation and that, on average, an incentive will result in a 2% reduction in energy consumption. Real-time feedback incentives as well as monetary information have the greatest effect. Finally, the more robust studies report smaller reduction effects. In the third chapter, an experimental set of common resources is used to explore individual responses to price-based incentives and nudges. Individuals are encouraged to reduce their consumption, either by a price increase or by smiley faces evoking their overconsumption. Price is the most effective in encouraging the target level of consumption, but it takes longer to take effect. Nudge is understood quickly but tends to reinforce overconsumption behaviors. The fourth chapter examines the effect of framing on willingness to exert effort. Individuals are asked to perform a simple, repetitive task for which they receive a piece-rate payment in the form of a gain or loss. Framing in the form of gain and loss is combined with three different payment structures: fixed gain, low gain, or high gain with equal probability revealed before or after the effort is completed. The results show that framing has no effect on effort completion, except for a high payoff context revealed before effort completion.