A 2005 subsidy to encourage more efficient energy use in sunny California worked remarkably well to reduce energy usage in lower-income communities, a new study determined. But the program had little effect on people’s behavior in cooler and wealthier coastal residences, so the study raises questions about the program’s overall cost-effectiveness.
Because substantial rebates were paid to the richer, coast-dwelling participants in the program, which is no longer in effect, the study’s author concludes that it was not as cost-efficient at reducing carbon emissions as it may have been.
The paper was produced by the Energy Policy Institute at Chicago, the research group that was behind another recent, controversial study questioning the cost-effectiveness of energy efficiency to cut carbon emissions (ClimateWire, July 8). Released at a time when the Obama administration is placing big bets on energy efficiency to help meet its climate goals, the new study also drew criticism from energy efficiency advocates, who said the program it analyzed was not designed to provide climate benefits.
Study author Koichiro Ito, an assistant professor at the University of Chicago’s Harris School of Public Policy, said his findings show that even though energy efficiency subsidies may be more politically popular than a carbon tax, they aren’t necessarily cheaper.
"If you change the price of electricity or price of energy, that’s going to give a right incentive for everyone to conserve electricity, so that’s very efficient policy," Ito said.
Although efficiency programs may be more politically favored than a carbon tax, for example, Ito said, "whenever you do the subsidy program, you have to remember the money is going to be paid by someone, and typically that someone is going to be a customer or an individual household."
$9M for vacation time?
The study was published this month in American Economic Journal: Economic Policy.
Ito’s analysis looked specifically at California’s 20/20 program, an electricity rebate program most recently put in place in 2005. It gave residents a 20 percent discount on their monthly electricity bill if they cut usage by 20 percent compared to the previous summer.
The advantage of using data from this program, the study explained, was that all ratepayers with electricity accounts for a year or more were automatically enrolled in the program. That way, there was no self-selection bias, and this also provided Ito with a good control group — that is, the ratepayers who barely missed the deadline to participate in the program were otherwise easily comparable to the participating ratepayers.
Ito learned that both temperature and income had a big impact on whether or not the program changed people’s willingness to ease up on their air conditioners.
"I find that the rebate incentive reduced electricity consumption by 4 percent in inland areas in California, where the summer temperatures are persistently high and the income levels are relatively low," Ito wrote in the study, adding that people in these areas continued to deliberately save energy during the three summers that followed the program.
But on California’s cooler, richer coast, it was a different story. In these areas, there was no measurable difference in behavior between people participating in the 20/20 program and those who did not.
This doesn’t mean people in these communities didn’t receive any rebates — the study notes that coastal customers received a total of over $9 million through the 2005 20/20 program.
Why might that be the case? Ito said it could have been because some coastal households were trying to save energy, but it’s just as likely that they reaped rewards accidentally, without changing their long-term energy habits.
For example, "let’s say you happened to travel this year but you didn’t travel last year," he said. "You can get the rebate for a reason other than your conservation."
Ultimately, the program’s cost was dramatically different depending on climate and income level, at 94.5 cents per kilowatt-hour reduction on the coast compared to 2.5 cents per kWh reduction for inland homes.
Energy efficiency advocates say study compares apples and ‘hamburgers’
Ito concluded that targeting lower-income electricity users could make future, similar programs more cost-effective. He also challenged the idea of choosing energy efficiency programs over more direct economic incentives to reduce energy use.
He said, "subsidies programs are very politically accessible, but we have to keep in mind that those policies tend to be very inefficient in the sense that we have to pay a lot of money to someone who did not conserve anything."
Steven Nadel, the American Council for an Energy-Efficient Economy’s executive director, said that while the analysis on how low-income, inland households were more likely to participate in the program was "useful," he believed Ito went too far with the study’s implications.
Nadel said the program was designed to address California’s electricity crisis in 2005, not climate change.
"The purpose of the program was to reduce peak demand, so where does carbon come into that?" Nadel asked. "The program had no impact on the price of hamburgers — but who cares?"
Nadel also downplayed the program’s cost, noting that it accounted for a relatively small fraction of the state’s spending in 2005 to tamp down on overall energy use.
But states and federal agencies will almost certainly be spending more money on energy efficiency to address climate change in the near future under U.S. EPA’s Clean Power Plan.
Michael Greenstone, director of the Energy Policy Institute at Chicago, said he was heartened by EPA’s approach to energy efficiency under the final Clean Power Plan, which he said included safeguards to ensure programs will actually result in emissions reductions.
In accompanying documents, Greenstone said, EPA "emphasize[s] the use of randomized control trials … to measure the true energy savings from energy efficiency investments."
"That’s an enormous step forward," Greenstone added. "The way energy efficiency investments are usually credited is based on engineering models of what people predict will happen; here they’re saying, ‘Actually, we’re going to look back and find out what happened.’"
