The severe and rapid economic slowdown due to the spread of coronavirus provides an opportunity to test the impact of lifestyle changes on electricity use. There are some insights that help policymakers calibrate the impact of different CO2-reduction policies.

As the quarantine was imposed and many businesses shut down, electricity use in Seattle decreased. Measuring that decline helps determine the size of potential electricity reductions from a significant lifestyle change.

Every day, Seattle City Light projects anticipated electricity demand to ensure it has the capacity to serve its customers. These short-term projections are based on several factors. In an e-mail, Scott Thomsen from City Light told me their estimates are based on “prevailing weather, day-of-week and other seasonal factors that drive electricity consumption.” Those projections don’t include the impact of the quarantine because they are new and difficult to predict. Measuring the gap between projected electricity demand and actual demand helps determine the effect of changing the behavior of most of the city.

The impact of the quarantine on total electricity demand is remarkably small. Thomsen told me, “We estimate that the impacts range from 2 percent to 6 percent reduction in load from what we otherwise would expect.” The data bear that out.

The chart below shows electricity generated, purchased, and used by Seattle City Light for the week of March 21 through March 28. The highlighted portion shows weekdays when we would expect to see the largest change. The solid blue line represents actual demand, and the dashed blue line shows Seattle City Light’s projected demand. The gap between those two lines provides an estimate of the impact of the reduction in demand due to the quarantine.

For example, on the morning of Thursday, March 26, demand was nearly 15 percent below projection. A few hours later, in the middle of the day when demand typically declines, electricity use was back to about 99 percent of projected demand.

The basic implication for policy here is that even a significant change in behavior has a relatively small impact on overall electricity demand. Seattle has witnessed an incredible shock in how people live and work, but the impact on electricity demand has been small. It will take time to determine why this is the case, but it seems like lifestyle changes are like squeezing the toothpaste tube – the demand doesn’t necessarily go away, it just shifts.

A debate central to many of the competing climate policies is the question of whether to emphasize behavioral change or technology innovation. The left generally stresses approaches that require people to change their lifestyle – pushing people out of cars and onto bikes and buses, and into high-density communities rather than spacious suburban homes.

These numbers demonstrate that a dramatic change in behavior can have small impacts on electricity demand. By way of comparison, technology solutions can yield significant reductions without an economic shock.

Working with Google Nest, Portland General Electric (PGE) studied the potential effect of using smart thermostats to reduce demand during peak hours. PGE adjusted the thermostats of homeowners who volunteered by up to two degrees during periods of high demand, rewarding them with a small rebate on their bill. In the winter, PGE found they could reduce demand by 20 to 33 percent in the first hour and 15 to 30 percent in the second hour, and about 15 percent in the final hour. The results were better in the summer, reducing demand during peak hours by up to 40 percent.

There was some rebound in demand after PGE gave control back to users. Thermostats turned on to heat or cool the house when the peak-demand hours had passed. The research, however, demonstrates that overall demand was reduced as well. Other studies find that once provided with short-term financial incentives, people become habituated to reduce electricity consumption, reducing total demand.

The comparison of reduced demand due to the behavior change from the quarantine as opposed to using incentives and technology is, admittedly, rough. This also doesn’t measure the impact of the quarantine on transportation fuels, which is significant. Nor does it measure the cost of the quarantine compared to the reductions in emissions, which is extremely high.

This rough look provides us with a general sense of the magnitude of potential alternatives and demonstrates that poorly calibrated lifestyle changes can have a very high cost with little to show for it. This is particularly true when some are pointing to the quarantine’s impact on air quality as evidence to support tough climate policy.

It also demonstrates how powerful innovation can be at reducing electricity use and environmental harm. After a couple weeks, the reductions in demand caused by the quarantine are smaller than those found when people are given the technology and incentives to conserve.