A new study finds a large household response to electricity prices in the long run.
The law of demand is as close as we economists get to an immutable law in our universe. When a product’s price increases, consumer demand for that product should decrease. To my economist’s eye, this law has all the elegance of Newton’s law of gravity or Ohm’s law of electricity. But its real-world application is messier because economic laws govern the behavior of humans instead of particles or electrons.Source: here and here.
Energy economists spend a lot of time thinking about how the law of demand applies to energy markets. If you think this sounds snoozy, think again. Energy demand elasticities, which summarize how consumers respond to energy price changes, are key inputs in the models and analysis that guide big-ticket energy infrastructure investments, electricity market regulations, and climate policy interventions. If these elasticity estimates are wrong, our models will mislead.
Economists have had some success at estimating electricity demand elasticities over the short run (see here and here). These short-run elasticities are the numbers that are typically baked into energy policy modeling (see here, for example). But many energy investments and policy decisions have long-run energy price ramifications. In the long run, I can do more than turn off my AC when electricity prices go up. I can invest in a more efficient one! So we should expect larger demand responses over longer periods of time as people find opportunities to take more significant actions. How much larger has been hard to pin down.
To estimate this long-run response, you might be tempted to compare the electricity consumption across consumers who have historically faced different electricity prices (think California versus Nevada). But that strategy is fraught because there are lots of confounding factors that differ across Californians and Nevadans that could generate differences in electricity consumption, like weather, housing stock, or lifestyles…
A new Energy Institute Working paper by EI alum Jesse Buchsbaum takes a new approach to this challenge. Jesse leverages a quirky feature of California’s retail electricity rate design that generates sustained differences in retail electricity prices across otherwise nearly identical households. Making these apples-to-apples comparisons, he finds that households that have historically faced higher electricity prices consume significantly less electricity.
A Clever Use for Quirky Electricity Rates
Our California readers may not even realize that the electricity prices they pay depend in part on what electricity “baseline territory” they happen to live in. Climate-differentiated electricity pricing is designed to mitigate some of the electricity bill impacts of living in a place that requires lots of cooling or heating to stay comfortable. Jesse uses data from PG&E’s service territory which spans the scorching summers of Fresno and the snowy winters of the Sierras.
The graph below provides an example of how the electricity price structure in my pricing zone (Zone T) differs from my friend Max’s pricing zone (Zone X). You can see that baseline territory pricing means Max can consume more electricity at the lower rate of 20 cents/kWh than I can because his micro-climate is hotter than mine.
Jesse’s paper leverages the fact that, whereas electricity rates change discontinuously when you cross a territory border, other factors (e.g. household income, climate) do not. So any differences in electricity consumption patterns can reasonably be attributed to the electricity price differences. He compares electricity demand across neighbors living very close to these borders to see how demand changes as you move from the low-price zone to the high-price zone.
Do consumers respond to electricity prices?
Jesse starts by looking at how households respond to month-to-month variation in price schedules. Periodic adjustments can move prices differently on either side of a territory border. In the short run, he finds that a 1% increase in retail electricity prices induces a demand reduction of 0.3% (or around 1.5 kWh per month on average for the households in the study). These estimates are in line with prior estimates of short-run elasticity estimates (see, for example, here).
Using more than 10 years of billing data, Jesse can also assess the longer-run impacts of sustained price differences generated by climate zones. He finds some remarkably large effects. Over multiple years, a 1% electricity price difference is associated with more than a 2% difference in household electricity demand (on average).
This is a surprisingly large elasticity estimate. But it is possible if people are considering their electricity bills when it’s time to remodel the house, purchase an air conditioner, replace the water heater, etc. So Jesse’s been doing some econometric detective work to try and figure out what could be driving such large effects.
One part of the explanation appears to be…space cooling. The picture below summarizes how electricity consumption increases with higher temperatures (or cooling-degree days). As the temperature rises, more people turn on their air conditioning which drives consumption up.
You can clearly see that this electricity demand response to warmer temperatures (i.e. more cooling degree days) is significantly higher among households on the lower-price side of zone borders (blue line). This is consistent with people using AC less – or investing more in energy-efficient cooling – when electricity costs more.
Jesse finds a similar -albeit harder to interpret- pattern when he estimates the relationship between average electricity consumption and heating degree days:
In areas that don’t get very cold (at least by my Canadian standards), this graph suggests that households facing higher electricity prices (red line) are investing in more efficient heating appliances and/or have developed the habit of putting on an extra sweater. Beyond 20 HDDs, the patterns are more confusing. The drop in electricity consumption as climates get colder could be partly explained by heating fuel choices. Lucas has a cool paper on home heating fuel choices which finds that households facing higher electricity prices and colder climates are less likely to heat with electricity.
California’s high prices will slow progress on electrification
Throughout this blog, I’ve been referring to the “low” and “high” price sides of pricing territory borders. But it would be more accurate to refer to these prices as “too high” and “even higher”. Retail electricity prices in California– across all climate zones – are much higher than the social marginal cost of electricity consumption because California uses retail electricity prices to raise needed revenues for power system infrastructure, wildfire adaptation, public purpose programs, etc.
Jesse’s paper shows us how high retail electricity prices have impacted household electricity consumption. Impacts over the long run are surprisingly large. Going forward, these too-high prices will discourage investments in electric cars, electric space heating, and electric water heaters. This is bad news in a state where residential sector electrification is at the core of our decarbonization strategy.
If you’re still reading this blog, you must be interested in/have questions about this new research. You are officially invited to Jesse’s Energy Institute webinar this Wednesday where he will be presenting his research and taking your questions. I’ll see you there!
Keep up with Energy Institute blog posts, research, and events on Twitter @energyathaas.
Suggested citation: Meredith Fowlie, “Who (Besides the EI Blog) Pays Attention to Electricity Prices?”, Energy Institute Blog, UC Berkeley, October 17, 2022, https://energyathaas.wordpress.com/2022/10/17/who-besides-the-ei-blog-pays-attention-to-electricity-prices/