Does Saving Energy Lead to More Energy Consumption?

"Rebound" is an argument occasionally used by critics of energy efficiency policies. They assert that the more efficiently you use energy, the more energy you will use. Since efficiency drives down the costs of energy, the argument goes, people demand more energy, which negates some or all of the efficiency gains.

The most extreme variation of the rebound argument is "backfire:" efficiency causes energy consumption to rise above the pre-efficiency level. Accordingly, if backfire is true, inefficiency results in less energy consumption. Or, to paraphrase an Amory Lovins joke about backfire, consumers should be required to waste energy in order to save energy.

Let's get down to brass tacks: If I buy a hybrid-electric car with better mileage than my current wheels, rebound theorists say that I will drive more as a result, offsetting efficiency gains through increased gasoline consumption.

Or, if I add insulation to my house in order to stop Btu's from flying through the walls, rebounders say that I will turn up the thermostat in winter, taking back those efficiency gains by burning more natural gas.

But how much unmet demand for energy consumption does efficiency actually free from cost shackles? If I buy a more efficient washing machine, would I wash my clothes more often? If I replace incandescent bulbs with compact fluorescents or LEDs, would I leave lights on longer? If I switch from a desktop computer to a more efficient laptop, would I spend more time reading postings from self-important bloggers like myself?

Entering the world of "rebound" arguments leads one into a tangled world of questions, assertions, conjectures, and ideology.

In a paper published on ElectricityPolicy.com earlier this month, a Natural Resources Defense Council team tried to sort it all out by putting rebound hypotheses to an empirical test.

The team's conclusions:

• Many rebound hypotheses are stated in such sweeping terms that they are not testable. They are more akin to opinions littered with ambiguities, not rigorously expressed, falsifiable hypotheses that can be tested by evidence.
• Other hypotheses are framed with a bias assuming that only negative rebound effects would occur. Take the example of the hybrid-electric car purchase. What if, instead of driving more, the purchaser uses money from fuel savings to pay down a credit card balance?
• Rebound effects, to the extent they have been measured by data, use up less than half the efficiency gains and don't persist through time.

The NRDC study quotes analysis from the International Energy Agency and the U.S. Department of Energy concluding that rebound ranges from 0 for residential appliances to 10 to 40 percent for residential water heating and 0 to 50 percent for home cooling. For cars, an EPA-Department of Transportation analysis estimated a 10 percent rebound from fuel efficiency gains.

At a macro level, the study relied on California Energy Commission analysis to show that the Golden State's broad efficiency policies have resulted in per-capita energy consumption levels substantially below those of the other 49 states.

Energy intensity - amount of energy used per unit of gross domestic product - has dropped in the U.S. and other industrialized countries even as the economy has grown. In the U.S., primary energy intensity fell, in constant 2005 dollars, from 15,390 Btu per dollar of GDP in 1973 to 7,400 Btu per dollar in 2010, according to Energy Information Administration data.

Rebounders argue that higher energy prices would be more effective than efficiency policies for cutting energy consumption and the unwanted emissions it causes. A price on carbon is certainly needed, but persistent market barriers to efficiency show that pricing by itself would not be sufficient for acquiring all the cost-effective efficiency savings that remain unexploited.

NRDC's study suggests that if rebounders want to press their arguments more effectively, they'll have to offer more empirical support for them. As Amory Lovins likes to joke: "In God we trust. All others bring data."