Smart meter devices and the effect of feedback on residential electricity consumption: Evidence from a natural experiment in Northern Ireland

Abstract

SummaryUsing a unique set of data and exploiting a large-scale natural experiment, we estimate the effect of real-time usage information on residential electricity consumption in Northern Ireland. Starting in April 2002, the utility replaced prepayment meters with "smart" meters that allow the consumer to track usage in real-time. We rely on this event, account for the endogeneity of price and plan with consumption through a plan selection correction term, and find that the provision of information is associated with a decline in electricity consumption of up to 20%. We find that the reduction is robust to different specifications, selection-bias correction methods and subsamples of the original data. At £15-17 per tonne of CO2e (2009£), the smart meter program delivers cost-effective reductions in carbon dioxide emissions. Keywords Abstract:Using a unique set of data and exploiting a large-scale natural experiment, we estimate the effect of real-time usage information on residential electricity consumption in Northern Ireland. Starting in April 2002, the utility replaced prepayment meters with -smart‖ meters that allow the consumer to track usage in real-time. We rely on this event, account for the endogeneity of price and plan with consumption through a plan selection correction term, and find that the provision of information is associated with a decline in electricity consumption of up to 20%. We find that the reduction is robust to different specifications, selection-bias correction methods and subsamples of the original data. At £15-17 per tonne of CO 2 e (2009£), the smart meter program delivers cost-effective reductions in carbon dioxide emissions.JEL Classification: Q40, Q41, D8

Transient and Persistent Energy Efficiency in the US Residential Sector: Evidence from Household-level Data

“…Policy measures based on nudges, social norms (Allcott, 2011;Allcott and Rogers, 2014), new information devices (Gans et al, 2013;Jessoe and Rapson, 2014) or information campaigns may attain improvements in the level of transient efficiency in the use of energy. By contrast, improvements in the levels of persistent efficiency would be typically sought through energy efficiency regulations on homes and equipment, incentives promoting a change of habits in the use of equipment, and/or by offering incentives on the purchase of new, high-efficiency equipment (Alberini et al, 2015;Alberini and Towe, 2015) and structures, and/or by introducing.…”

Sensitivity of price elasticity of demand to aggregation, unobserved heterogeneity, price trends, and price endogeneity: Evidence from U.S. Data

“…review a number of studies, and suggest that differences might be due to the sample period, the nature of the data-such as panels (Maddala et al, 1997;Metcalf and Hasset, 1999;Garcia-Cerrutti, 2000;Bernstein and Griffin, 2005; v. pseudo-panels (Bernard et al, 2011), cross-sections (Nesbakken, 1999;Krishnamurthy and Kriström, 2013;Quigley and Rubinfeld, 1989;Boogen et al, 2014;Reiss and White, 2005;Gans et al, 2013), or time series (Kamerschen andPorter, 2004, or Dergiades andTsoulfidis, 2008)-geography, and level of aggregation of the data. In more recent studies, the price elasticity of electricity consumption ranges from as low as -0.06 (Blazquez et al, 2013) to as high as -1.25 (Krishnamurthy and Kriström, 2013).…”

Estimating the Impact of Time-of-Use Pricing on Irish Electricity Demand

“…For evidence on Denmark see Gleerup et al (2010). 3 While Gans et al (2011) analyse the impact of enhanced feedback on electricity consumption in Northern Ireland, they focus just on a group of households who were already cognisant of their electricity consumption due to their choice of the prepayment option of payment. The extent to which their results are generalisable to other household types is debateable.…”

“…In terms of indirect feedback, historic feedback (comparing with previous recorded periods of consumption) appears to be more effective than comparative or normative information (comparing with other households, or with a target figure). Gans et al (2011) Northern Ireland (from 1990 to 2009), which is merged with price and plan information from the electricity utility, and weather data (the final sample size is over 45,000 observations). They find that households using the keypad use 15-20 per cent less electricity than other households, even controlling for housing type, heating, household characteristics and selection into the plan.…”