Skip to content

Electricity Outages Lead to Substantial Backup Generator Purchases

Preliminary survey results from California suggest that long outages are leading people to buy backup generators, possibly instead of rooftop solar.

The COVID-19 pandemic has shone a very bright light on the importance of making well-informed tradeoffs. Imagine how comforting it would be if every mayor, governor and national leader had a calculator that told them exactly how many days of lockdown they should enforce to balance the public health risks against the economic losses from shutdowns?  

Californians faced another set of tradeoffs last fall with the wildfire-related electricity outages. The state’s utilities shutoff power to millions of people for multiple days during wildfire season, arguing that de-energizing the electricity grid, while inconvenient, would prevent larger losses from fires caused by the electricity system. Just as we’re seeing with COVID-19 shutdowns, a number of commenters wondered whether these “Public Safety Power Shutoffs” (PSPS) were really worth it.

Map showing projected outage areas (in turquoise).

Revealing the Cost of Blackouts

Unfortunately, the people making the decisions about the blackouts were pretty much – shall we say – in the dark. We know very little about the costs of power outages to the economy. Recognizing this, colleagues Fiona Burlig, Duncan Callaway and Will Gorman and I have set out to try to fill some of the knowledge gaps. 

One way to measure the costs of outages is to try to identify the major cost categories, like wasted food, health impacts, business’ lost sales, etc., then measure those costs and add them all up. Think of this as a bottom-up approach. 

Another way, which I will talk about today, is to try to understand how much customers are willing to pay to avoid all of the costs associated with future outages. This is what economists call a “revealed preference” approach. Basically, if a number of customers race out to buy backup generators after the outages, this is a summary measure of how costly the outages were to them: so costly that they’re willing to spend thousands of dollars to avoid the effects of the next one.

My colleagues and I, supported by a fantastic team of undergraduates, have been surveying residential customers in California to understand their experiences with the outages. In order to construct a revealed preference measure of economic losses, we’re asking people a series of questions about their investments in non-grid power sources.

UC Berkeley undergraduates Amy Li, James Wang and Elise Angelich conducting surveys. Photo by Kathy Nagel.

Which Actions Were Due to the Outages?

To be able to say that the outages are what led people to buy generators, we have to try to understand what they would have done absent the outages (i.e., establish the counterfactual). One idea is to compare them to people who didn’t experience outages, but we have to tread carefully. For instance, if we just compared people who live in the Sierra foothills — many of whom lost power for days — to those who live in downtown LA — many of whom didn’t have any outages — the resulting difference in generator ownership would reflect both differences in outage experiences as well as underlying differences between foothillers and Angelinos. Maybe there are a lot more stores that sell generators near the foothillers, for instance. This wouldn’t necessarily say anything about the impacts of the outages.

With the PSPS, though, neighbors – likely similar to one another in many ways – had very different outage experiences. For example, my neighbors, Fred and Evelyn, who live a block and a half away, experienced more than 50 hours of outage while we didn’t experience any. This dramatic difference had to do with the configuration of the distribution grid and which circuits were turned off. But, along a lot of other dimensions, we’re pretty similar to Fred and Evelyn – we both have two kids (who went to the same pre-school), shop at the same stores, etc. So, to guess what Fred and Evelyn would have done if they hadn’t experienced outages, we can look at what their neighbors, like my family, did. This is the approach we’re taking: comparing households who experienced PSPS-related outages to their nearby neighbors who didn’t.

Early Results

We’re still completing the survey, but here are a couple preliminary results from the first 320 people we’ve talked to (they’re spread around the state, but all live close to the borders of the outages):

  1. Lots of backup generators. Among the people who experience outages, we’re seeing almost 15% of them buying backup generators in the 6 months since the outages. On average, they’re paying over $1800 for those generators. This is a lot more generator purchases than I would have expected from residential customers. We need to dig into this number further and understand how representative the people who are answering our survey are, but it suggests the costs of outages are pretty high.
  2. Backup generators instead of solar. If you believe that our comparison group (e.g., my family) describes what people (e.g., Fred and Evelyn) would have done had they not faced the outages, then the figure below suggests that people who experience outages are buying generators instead of home solar.7 percent of the people in the no outage group report installing rooftop solar since the outage while only 2 percent in the outage group have done so. It’s not immediately clear why we’re seeing this pattern. Maybe the costs of installing a solar system that will also provide backup power (with battery backup and a smart inverter) are prohibitive for people who really value having a backup electricity source?
  3. Outages are painful. One of the valuable things about doing surveys instead of using data that others have collected is that you get to hear the stories behind the numbers. We’ve heard from an elderly woman who was hospitalized after she fell grasping for her flashlight in the dark, a couple who couldn’t access their cell phone to report a medical emergency during the outage. These are heartbreaking and help put a human face on the economic figures. We’re also planning to complement our surveys with archival data – on outcomes like hospitalizations, business closures, and foot traffic from cell phones – but the human stories emphasize just how much some people’s lives were impacted.

I started this blog post talking about making informed tradeoffs. Armed with accurate measures of the value that customers place on reliable electricity, we will be able to help policy makers and utilities decide how much investment to make in outage-prevention technologies – things like remote sensing, undergrounding lines, strategically placed micro-grids on the utility system, more vegetation management, more line inspections, etc.

It’s likely that there are cheaper and more environmentally benign ways to keep the electricity on and prevent fires than having customers install their own backup generators. When the utility makes an investment, it preserves power for all its customers – rich, poor, residential, commercial. These utility-side investments are kind of like vaccines for the electricity system, but some are a lot more expensive than others, so we need to figure out how many of them are worth it. Based on our results so far, though, I suspect that there’s a lot more investment that does make sense.

Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.

Suggested citation: Wolfram, Catherine. “After a Long Electricity Outage, Substantial Investment in Backup Generators” Energy Institute Blog, UC Berkeley, May 26, 2020,


Catherine Wolfram View All

Catherine Wolfram is Associate Dean for Academic Affairs and the Cora Jane Flood Professor of Business Administration at the Haas School of Business, University of California, Berkeley. ​She is the Program Director of the National Bureau of Economic Research's Environment and Energy Economics Program, Faculty Director of The E2e Project, a research organization focused on energy efficiency and a research affiliate at the Energy Institute at Haas. She is also an affiliated faculty member of in the Agriculture and Resource Economics department and the Energy and Resources Group at Berkeley.

Wolfram has published extensively on the economics of energy markets. Her work has analyzed rural electrification programs in the developing world, energy efficiency programs in the US, the effects of environmental regulation on energy markets and the impact of privatization and restructuring in the US and UK. She is currently implementing several randomized controlled trials to evaluate energy programs in the U.S., Ghana, and Kenya.

She received a PhD in Economics from MIT in 1996 and an AB from Harvard in 1989. Before joining the faculty at UC Berkeley, she was an Assistant Professor of Economics at Harvard.

25 thoughts on “Electricity Outages Lead to Substantial Backup Generator Purchases Leave a comment

  1. How does the solar adoption in the untreated group compare to general solar adoption trends in CA? Are these people buying battery storage also, or just rooftop solar?

  2. “If the cost of electricity outages is valued as anything other than, length of outage*applicable price to the customer class, it can only be because electricity is not an orthodox market good. Yet, many influential economists still pretend that there are electricity ‘markets’.”

    Agree, Russ – when electricity’s end-consumers have no choice, the phrases “market price” and “competitive” have no meaning. Free-market economics has never been applicable to electricity (or any other commodity provided by a natural monopoly).

  3. Another thought on the private and social economics of this decision — backup generators (BUGs) can cause a number of significant risks:
    1) Substantial emissions, especially particulates and NOx that directly cause and exacerbate respiratory problems. The state Air Resources Board has estimated that 125,000 BUGs were operated during
    the October 2019 PSPS events. From only 50 hours of operation, the BUGs emitted diesel particulates equivalent to driving 29,000 heavy duty diesel trucks on California roadways for a month, or 87 million total miles of travel.
    2) Risk of death from carbon monoxide, when generators are operated in enclosed areas. BUGs kill an average of 80 Americans each year from carbon monoxide poisoning
    3) Fire risk: According to state Fire Marshall data, BUGs start an average of 40 fires per year in California. During PSPS, they would be used in exactly at the time and in the locations where power is shut off TO PREVENT FIRE RISK!

    It would be interesting for an economist to put a value to these risks and include them in the total calculation, not just of energy-not-served but of responses to power outages.

    Again, all this is in my Vote Solar report at

  4. Solar offers limited support as blackouts can and do happen when the sun is not available.

    Batteries require up-to-date electrical systems in homes and many homes don’t have that. So one needs to add in the cost of that.

    With quite a bit of lag… Northern CA – one of the wealthiest regions in the world… with an enlightened population – has decided to follow in the footsteps of India (poor)! Apologies for the snark.

    Solar+batteries have a long long way to go before it becomes a true distributed resiliency alternative.

  5. It never ceases to amaze me that top-class economists bend themselves into pretzels when it comes to electricity. No-one asks what’s the cost of toilet paper or lysol wipes “outages”; it’s the price of these items multiplied by the relevant quantity (e.g. estimates of shortfalls in some period). I’m also surprised after the work of Kahneman and Tversky has become widely recognized, plus Thaler’s Nobel, that anyone gives any credence to Contingent Valuation surveys. If the cost of electricity outages is valued as anything other than, length of outage*applicable price to the customer class, it can only be because electricity is not an orthodox market good. Yet, many influential economists still pretend that there are electricity “markets”.

    • Russ
      You make two contradictory statements. The first being we should try to measure the value of an outage by any means other than the price times the lost quantity (which, BTW, is true because it ignores the lost consumer surplus) so that should be true for electricity, and the second being that electricity doesn’t have a “market” price. These can’t both be true statements. Your second statement is what is true, which is why your first statement is false and why we have to use other methods to derive the value of outages.

%d bloggers like this: