What would you pay for more reliable electricity?

Getting your Trinity Audio player ready...

Last week there was a power outage in my neighborhood during an intense rainstorm. We were hit with two flavors of outage. Some houses had a brownout (very low power only) from 3:30pm to 8:30pm, while others had a brownout followed by a power outage, from 3:30pm to about 4am. One of my neighbors sent out a note afterwards: “I am wondering if I should invest in a power generator. Last night was miserable. Any advice?” She was willing to spend real money to avoid similar outages. How much money, I wondered? And is she an outlier or the norm? What is more reliable power worth to people?

A Vermont utility, Green Mountain Power, offers its customers two Tesla Powerwalls in exchange for $55/month (for ten years). The utility reserves the right to use the batteries to supply grid power during times of very high demand. With this plan, the customers are able to avoid power outages and the utility can reduce its generation costs. The program has been limited to 500 customers per year until this past summer, when they opened it up after a particularly outage-filled winter. Plenty of Vermonters seem willing to pay $55/month for much better reliability. But what about in other places?

There have been many attempts to estimate peoples’ willingness to pay to avoid power outages. If regulators have a good handle on this number, they can better determine how much to invest in reliability. California’s power grid aims for 1 outage every 10 years, but that applies to supply-side issues (running out of electricity) and not to the problems with local distribution that cause around 94% of our outages (squirrels, mylar balloons, fallen branches). Are we investing enough in hardening the distribution grid? Should we be undergrounding everywhere? Alternatively, should we lighten up on the 1-in-10-year requirement for supply-side shortages because rolling outages aren’t all that bad, especially when compared to these other, much more frequent outages?

As you can see in the chart above, the estimates determined over the years for willingness to pay (aka “value of lost load”) have been all over the map, ranging from $.05/kWh to over $30/kWh for residential use. Such a big spread is almost meaningless. That may be one reason why cost-benefit analyses of building electrification like the one I wrote about in this blog post do not incorporate the value of reliability. Gas is more reliable than electricity, so residents may incur extra outage hassles when they electrify.¹ How significant are they, and to what extent do they change the calculus of electrification?

Researchers use a variety of techniques to evaluate willingness to pay. One approach is to survey customers, asking what they would be willing to pay to avoid different types of outages. (Is the outage in summer or winter? Announced in advance or not? Two hours or eight?) These surveys are relatively easy and inexpensive to do, but answers to hypotheticals are not especially reliable. What users say they will do and what they actually do are often very different. Nevertheless, Texas grid operator ERCOT, which had a horrific power outage in the winter of 2021 that killed over 200 people, will be doing just such a survey in the coming months. They are not sure how much to raise rates to fund improved reliability, or what it might cost them to ask users to reduce power use at peak times. The survey may help.

A different approach is to look at what actions customers actually take in response to outages, which economists refer to as “revealed preferences.” I will briefly cover two such studies, one by Robert Harris on generator sales and one by David Brown and Lucija Muehlenbachs on home battery purchases.

Harris looked at generator sales in the US from June 2012 to December 2020. There were over two million generator sales during this time period.

The sales align pretty well with the risk of high winds, as shown in the Federal Emergency Management Agency (FEMA) map below. 75% of all generator sales occur in the week before and the week after an outage, many of which are due to high winds.

Harris looked at how much power the generators have and how long they are expected to operate, and from that determined what people were willing to pay to avoid outages. He came up with an average figure of $1.57/kWh, or $10.37 per outage. Based on this, he concluded that undergrounding distribution lines rarely makes sense — the costs far outweigh the benefits.

He looked closely at generator purchases in northern California in 2019, when PG&E was proactively shutting off power on transmission lines during high fire danger days. These shutoffs affected millions of people with prolonged outages that could last for days. The average number of outage-hours quadrupled in 2019 to nearly 9 hours for an average California household, and much more for those hit by power shutoffs. Many households were willing to spend over $100 for more reliable power, as shown in the map below. Harris notes that the total willingness to pay to avoid outages in 2019 was over $200M, which easily exceeded the $75 million Governor Newsom allocated to mitigate the damage from the power shutoffs.²

Brown and Muehlenbachs focused exclusively on these power shutoffs, looking at battery storage purchases rather than generators.Their estimate for willingness to pay was much higher than Harris’ nationwide $1.57/kWh, coming in at $4.29/kWh. They surmise this may be due to the fact that the California outages were particularly long and therefore harmful. But it could also be that batteries are expensive.They found that purchases were largely restricted to the top 25% higher-income households. Since their estimates do not include generators or other avoidance or mitigation costs, they expect their willingness to pay metric is still too low for California households. Moreover, neither of these studies ascribes any cost to those households that make no purchases because the options are unaffordable or the residents do not have time to deal with it. Perhaps the survey results could be combined with these “revealed preferences” studies to reflect a more accurate valuation of outage costs.

We don’t need scientists to tell us that outages can be a nuisance. As climate change increases the severity of storms and the risk of wildfire, it is likely that outages will become more common and peoples’ willingness to pay to avoid them will go up. We are seeing outage levels rise across PG&E territory, as well as in hurricane-prone regions.

It will take a while to reinforce the grid, so I expect we should all get better at dealing with outages. What was the experience like for my neighbors during this most recent outage? It proved pretty uncomfortable for my older neighbor who is now considering a generator.

“I went to the kitchen to make some coffee, opened the refrigerator, and found the light dim. I tried to light my gas stove, but the pilot didn’t work. Some lights worked, and some didn’t. Then, the power went out altogether. I made a call to PA Utilities and left my location details. I then checked X (Twitter), where I follow PA Utilities, and found a few postings but nothing specific to my neighborhood. Our heater needs electricity, so our house started getting cold quickly. I don’t tolerate cold temperatures. So I read for a while in my bed, bundled in two sweaters, covered my head with a scarf, and went to sleep.”

Others mentioned a lack of good information about the outage.

“Some lights and other electrical items didn’t work or worked feebly. We figured it out, but the city didn’t catch up. I received text messages several times telling me that the power had been restored when it obviously had not. One text message told me that it would be restored at 8:31. That is the time the power went completely out and didn’t come back up until around 3 am. All of this to say that there should be a better information system with more accurate updates. There was no way to text back anything other than ‘out’.”

and

“We heard two explosions, and our power went out. It came back on very quickly, but many electrical items did not work. Some worked fine, and others not at all. For example, our stove hood LED lights worked, but the fan motor would not turn on. I was very, very concerned that a power surge had gone through the system and destroyed many of our items. Particularly my computer, which would not start, until I replaced the power supply with a new one, and then it worked. I was annoyed and wondering about the consequences.”

For some who lost internet, that  was almost a bigger problem than the power outage.

“The biggest challenge was that the high schooler’s homework was all on the iPad which she could not do without internet.  Oddly a while after we lost power, we got the internet back, so she used our backup battery and patched through to get her homework done.  Homework that should have taken a couple hours ending up spanning about 7 or 8 instead…. When the power came back on shortly after 3 a.m., it awakened us, so a disrupted night’s sleep was an additional bonus.”

Several neighbors used the outage as an opportunity to eat out or order in.

“We heard a loud noise, and our power went out, period, for about  4-1/2 hours. We bundled up, used some flashlights and some candles, and ordered out for dinner.”

and

“Power was out for about 12 hours at our house.  Low voltage for a time and then it went completely dark.  Thankfully, my phone was fully charged. We lit some candles, located a flashlight, and bundled up. An Old World experience. We popped over to California Avenue where they had power for dinner.”

Others opted for something more like camping, enjoying some time “unplugged”.

“For us, the power went out right as I was checking on my roasted chicken in the oven. Ended up finishing it on the grill and carried on with Sunday potluck with friends by candlelight, camping lanterns and low light power too. Came home to half the house lights on at half the power. Settled in and played Trivial Pursuit as a family by half powered lights, candles and flashlights. Honestly, in the grand scheme of what’s going on in our state with flash floods and mud slides and power being out for days for some, let alone destroyed homes and loss of life, I feel fortunate that we were only inconvenienced for a few hours and made the most of it as a family and enjoyed unplugged time together.”

And then there was this approach, which I hope more of us can do in the coming years.

“Last year I bought a used Nissan Leaf to cut down on transportation costs when going about town. I also got a $100 inverter so that I could plug appliances into the car when need be. So everytime the power goes out, I simply run an extension cord from the garage into the house to keep the fridge and router running and we barely even notice the inconvenience (other than having to rely on LED camping lanterns for light after the sun goes down). We even plugged in our TV this last time and streamed a Disney movie for the kids.”

One of my takeaways from reading the studies and hearing from neighbors is that California should worry less about rolling outages, which are planned and last for at most two hours per household, and more about the longer unplanned outages that stem from problems with the distribution grid. Maybe we could save on some of the dirty “peaking” capacity we buy for times of high demand and instead use the money for grid hardening. I was also surprised to read that one in eight households in California has a generator. I suspect many of those households will be hard to fully electrify, depending on how powerful the generator is. So my final thought is that the day cannot come soon enough when it is easy and affordable to use an EV battery to supply power to a house during an outage. I will give an update on that later this year.

How did you deal with power outages during the storm? With climate change intensifying storms and wildfire risk, are you doing anything to make it easier for your household to handle outages?

Current Climate Data

Global impacts (December 2023), US impacts (January 2024), CO2 metric, Climate dashboard

Want to be Notified of New “A New Shade of Green” Blog Posts?

To be notified by email of new blog posts, send an email to notify@newshadeofgreen.com with “Subscribe” in the subject.

Comment Guidelines

I hope that your contributions will be an important part of this blog. To keep the discussion productive, please adhere to these guidelines or your comment may be edited or removed.

• Avoid disrespectful, disparaging, snide, angry, or ad hominem comments.
• Stay fact-based and refer to reputable sources.
• Stay on topic.
• In general, maintain this as a welcoming space for all readers.

Notes and References

1. Many gas appliances require some electricity to operate (e.g., most gas furnaces). But they require much less power than their electric equivalents, so it is easier for a backup power supply like a generator or battery to keep them operating during an outage. ↩︎
2. PG&E has since reduced the scope and duration of these power shutoffs, but it continues to be under pressure in the largely rural communities that are bearing the brunt of these outages. ↩︎

Most Popular

Hyatt House Pleasanton recently sold at auctionJune 17, 2026 1:54 pm

Pleasanton set to be featured on national television5 hours ago

The pathway for horse racing to return to PleasantonJune 16, 2026 12:22 pm