A study in the latest issue of the journal Energy Policy casts doubt on some of the claims that utilities and smart grid advocates have made for smart electricity meters and dynamic pricing for electricity. Researchers at the University of Waterloo in Ontario examined 20,000 households in southwestern Ontario before and after time-of-use rates, supported by the deployment of smart meters, were introduced in November 2011.
The program, launched by an unidentified distribution company, resulted in "only modestly reduced residential energy demand during the most expensive peak periods," the study concluded. The city of Waterloo, Ontario, and two townships own an electric distribution company, Waterloo North Hydro Holding Corp., that serves 55,700 customers.
Comparing data for nine months before and nine months after time-of-use rates were introduced, the study found that residential demand for electricity fell 2.6% during peak demand periods — less than forecast in utility projections used to justify the expense of deploying smart meters and offering dynamic pricing programs.
"There is a gain, but the gain is very small," said Lukasz Golab, a management sciences professor at Waterloo and one of the study authors, in a May 31 news release accompanying the study.
The findings come as utilities across Europe and North America continue widespread deployments of smart meters. China is also in the midst of a massive rollout of smart meters, expected to grow steadily from the approximately 160 million units installed in 2016. According to a report from Navigant Research, the global penetration of smart meters is expected to reach 53% by 2025, growing from about 30% at the end of 2016. In the U.S. alone, more than 60 million smart meters are expected to be in use by 2020, the Edison Foundation forecasts.
Smart meters "provide customers with greater control over their electricity use when coupled with time-based rates," said the California Public Utilities Commission, while "reducing the need to build power plants, or avoiding the use of older, less efficient power plants as customers lower their electric demand."
Those benefits are likely overstated, the Waterloo researchers found. Among other conclusions, they found that "confounding factors," specifically the impact of abnormal weather, make isolating the true effects of smart meter and time-of-use pricing plans difficult to quantify.
"Properly evaluating the impact of such policies is critical for policy makers trying to reduce demand, reduce emissions and defer new generating capacity," the study concludes. But mild or extreme weather, particularly in summer, can "skew the estimated impact of TOU pricing if the effects of weather are not adequately modelled."
Somewhat surprisingly, the researchers found, "electricity demand is not being shifted to off-peak periods, but is only being conserved." To the extent that it reduces overall energy consumption, that's probably a good thing, but the effects may be too small to confirm that the benefits of smart meter deployments outweigh the costs. The Waterloo researchers were careful not to make that claim, though. "We don't have the data to decide if these kinds of savings warrant the use of smart meters," said co-author Catherine Rosenberg, a professor of electrical and computer engineering, in the release.
One problem is what David Rapson, an economist at the University of California at Davis, calls "rational inattention": even consumers with the latest smart meters installed may find that adjusting their behavior is too laborious and too time-consuming to warrant the investment of time and mental energy.
In any case, the study will cast doubt on the more extravagant claims by utilities and suppliers that smart meters and dynamic pricing can dramatically lower energy demand.
"ToU [rates] achieve a small savings in residential electricity during on-peak periods," Rosenberg said in an email. "Policymakers will want to determine if the savings that were achieved are in line with the goals they set out to achieve."
Previous studies of Ontario's adoption of smart meters and time-of-use rates found a reduction in electricity use during high-cost peak demand periods, but not an overall reduction in energy consumption, just a shift to lower-cost periods. Ontario's auditor general in 2014 said the province's rollout of the smart meter plan was not well-planned and that differences in peak and off-peak prices weren't enough to encourage actual reductions in energy use.