Skip to content

Do Utilities Like Heat Waves?

As temperatures climb, utilities sell more electricity.

We had a heat wave in Berkeley last month. I know, a high temperature of 88° Fahrenheit doesn’t sound very hot to most people, but we’re not used to it and our house is neither air-conditioned nor super well insulated.

After a couple nights of tossing and turning, my daughter and I headed straight to Target to get fans for our bedrooms. Unfortunately for us, many other shoppers had the same idea, so the shelves were literally bare.

This reminded me of a maxim I learned early in my career. I was a rate analyst at the Department of Public Utilities in Massachusetts and during a Boston heat wave (much less comfortable than the Berkeley version, as both temperature AND humidity exceeded 90), one of my colleagues pointed out that, “utilities love heat waves.” He went on to explain that during heat waves, people ran out to buy fans and window-unit air conditioners, just as we had (Amazon Prime succeeded for us where Target failed). The next time it was hot, they cranked up their new appliances and electricity consumption was even higher.

Lana Del Rey buying a fan


After our Berkeley heat wave, I decided to see if any evidence supported his claim. I found that (a) indeed, temperature drives peak demand, (b) there’s some evidence that adjustments people make to their electricity consumption in response to temporary shocks can have long-run impacts, but (c) the relationship between increased sales and increased profits is complicated for regulated utilities.

Temperature and Peak Demand

A neat paper by co-blogger, Max Auffhammer, with Patrick Baylis and Catherine Hausman (both former Energy Institute grad students) uses data from over 150 areas around the country to estimate the relationship between temperature and both daily average consumption and daily peak consumption. (Max blogged about it earlier here.)

I’ve reproduced one of the figures from their paper below. It shows that in the Texas wholesale market, peak demand (in red) is almost 25,000 MWs higher on days when the average daily temperature is 90° F compared to days when the average daily temperature is more like 60° F.

Source: Auffhammer et al., PNAS 2017

The average daily temperature is the sum of the minimum and the maximum temperatures on a day divided by two. On a 90° F day, for example, the overnight low could have been 80° F and the high 100° F – not good sleeping weather!

My quick analysis of California data supports the temperature-peak demand relationship across years. The figure below plots annual peak demand in the California ISO against the sum of the Sacramento and Los Angeles high temperature on the corresponding day. I’ve normalized each data point to 1 in 1998, so the figure shows temperatures and peak demands relative to what they were in that year. The highest peak demand we’ve seen over the period was in 2006, when temperatures were as hot as they’ve been over the 19-year period. (The California Electricity Commission is looking at peak demands in much more depth. See the last agenda item here.)

Source: Author’s calculations from California ISO data and Weather Underground

Time Dependency

Is there any evidence that a heat wave one year leads to more electricity demand the following year? If you look closely at the California graph above, 2007 has about the same high temperature as 2005, but peak demand is a lot higher in 2007. This could reflect a number of different factors. For example, maybe the temperature on the day preceding the peak was a lot higher in 2007 than in 2005. But, it’s also consistent with the fan-buying phenomenon. July 2006 was really hot, so I’m guessing a lot of Californians bought new fans, which they still owned on the hot day in 2007.

A paper by Francisco Costa and Francois Gerard substantiates this idea, although they come at it by looking at a period when people reduced consumption. They study a 9-month period in 2001 when several regions in Brazil, which depends heavily on hydroelectricity, were in a severe drought. Consumers in the affected regions were assigned consumption quotas and faced fines and possible disconnection if they exceeded their quotas. During the crisis, these measures led consumers to reduce their consumption by nearly 25% on average. More interestingly, customers who faced the quotas still consumed 10% less electricity ten years later.

They go on to show that households reduced their consumption both by getting rid of appliances, including air conditioners, and also by changing habits. For example, customers in the drought-affected areas were less likely to have their electric water heaters set to the highest temperature five years later.

The Relationship Between Sales and Profits is Complicated

The question I’ve posed in the title assigns an emotion to utilities, so let me be more precise. The idea is that utilities “like” to see higher electricity consumption during heat waves since they will earn more money by selling more kWhs.

In fact, for regulated utilities, the relationship between increased sales and increased profits is complicated. Over the past 10-15 years, 19 state public utility commissions (plus DC) have instituted “decoupling” programs for their electric utilities – see the purple and blue states below. The specifics vary, but in principle decoupling guarantees that utilities will earn a specific amount of money (called the “revenue requirement”) to cover their costs. If sales spike during a heat wave, rates will be set slightly lower in the future to offset that increased revenue.

Source: Center for Climate and Energy Solutions

In states that do not have decoupling, utilities will likely be able to keep the extra revenue from increased heat-driven sales. Their costs will also be higher as they run plants with higher fuel costs or purchase expensive power from wholesale markets. But, for most utilities, fuel and purchased power costs are passed directly to ratepayers through fuel adjustment clauses.

Utilities might also experience higher costs operating their transmission and distribution systems as both the heat and increased demand strain the networks. These sorts of costs are harder to pass on to customers. And, to the extent the strain on the system leads to outages (like it did in Los Angeles two weeks ago), utilities will face irate customers.

In the long run, though, higher peaks require more investment in new capacity, which generally increases utility profits. This is true across generation, transmission and distribution, and even true for companies in decoupled states.

So, while the decoupled Massachusetts utilities may now be less effusive about heat waves than my former colleague suggested, I’m guessing that utilities in places like Florida are still cheering – or at least doing silent fist pumps – as temperatures climb. And, as Max and his co-authors have pointed out, we all need to pay attention to these heat-driven peaks since they’re becoming more and more common with climate change.

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.

14 thoughts on “Do Utilities Like Heat Waves? Leave a comment

  1. Catherine, I know in central California you folks are getting FlexAlert warnings just like we are in Burbank. We’re coming up to the season where bright, sunny days, instead of providing a windfall of clean solar energy, threaten to take the grid down as the sun sets and natural gas can’t be piped in fast enough to feed electricity generation plants.

    Now that the California Energy Commission has admitted the closure of San Onofre was responsible for the Aliso Canyon Pipeline Disaster, I’m looking forward to outages later this summer. Having millions in the Bay Area’s tech sector suddenly unable to turn on their computers and charge their smartphones might force CPUC to consider denying PG&E’s application to close Diablo Canyon, the reliable source of 9% of California electricity and 22% of its clean electricity.

    Joni Mitchell is known as an ardent opponent of nuclear power. Wasn’t it she who also penned the lyrics:

    “You don’t know what you got, ’til it’s gone”?

    • Bob Meinetz, there’s been one valid Flex Alert this year early on, and most recently CAISO had overforecasted peak load by a couple thousand MWs, so we coasted through pretty easily. It turns out that the loss of SONGS has had little impact on the system operations and we have plenty of capacity into the next decade. In fact, we’re over resourced with obsolete expensive generation.

      • mcubedecon, the California Public Utility Commission disagrees with you:

        “California’s electric system has 26,000 miles of bulk electric transmission lines ranging from 60 kilovolts (kV) to 500KV and hundreds of electric generation sources that work in concert to continuously maintain system reliability and balance supply and demand. In 2012, the San Onofre Nuclear Generating Station representing 2,246 MW was retired. Solar resources have compensated for much of the energy loss during the daytime hours. However, the use of the gasfired generation has increased during the shoulder hours and to maintain local reliability.”

        “Aliso Canyon Working Gas Inventory, Production Capacity, Injection Capacity, and Well Availability for Summer 2016 Reliability”

        “SoCalGas had been using both the tubing and a much wider steel casing surrounding it to deliver larger volumes of gas. Experts said that was risky because the casing was a safety barrier if the tubing failed. In the case of the blown-out well, the casing is believed to have failed under high pressure, allowing the gas to escape.”

        In fact, 98% of the clean nuclear generation we depended on was swapped out for dirty gas-fired generation, increasing California’s CO2 emissions by 8 million tonnes/year.

        • Bob, SoCalGas’ pipeline system was designed to deliver multiples of current gas consumption. Many gas fired plants have been retired in the LA Basin already, and other replaced with 30% more efficient combined cycle generators. The Aliso Canyon blowout had nothing to do with the recent upswing in generation gas demand (which is only a small portion of total basin gas demand), and everything to do with SoCalGas maintenance practices.

          The issue of reliability and renewables could not be addressed by SONGS because SONGS couldn’t ramp up and down to follow renewable generation patterns (much less provide fast ramp). The need for fast ramp is what is driving the reliability question, not total continuous capacity availability.

          • mcubedecon, Aliso Canyon is not part of SoCalGas’s “pipeline system”. It’s a reservoir, and an injection casing blew out due to the attempt to cram too much gas through it – as I’ve already explained:

            “In October, the time of the blowout, the utility had pumped Aliso Canyon to 93 percent of capacity, the utility said. State records show 2014 was a near-record year for injections, filling 71 billion of Aliso Canyon’s 86 billion cubic feet capacity.
            Former PUC President Loretta Lynch said state regulators should have kept a tighter rein on storage amounts. The utility commission allowed SoCalGas operators “to run the hell out of those pipes, and found in the process it had endangered thousands of people.”


            SONGS was entirely capable of load-following customer demand, so statements like “SONGS couldn’t ramp up and down to follow renewable generation patterns” mistake the problem for the solution. Get rid of nuclear? No, get rid of renewables, and our problem is solved. Renewables cause more problems than they’re worth, and always have.

    • Bob,

      I am not sure who your service provider is, but if it was PG&E they can call for a Flex Alert independent of a CASIO alert. A few years back, 2013/4, I wanted to sign up for PG&E’s demand response Flex Alert program- my wife and I can normally modify our load if needed as we have done it for 11 years as we had an E-7 NEM TOU rate schedule- we were not allowed to sign up for the program.

      I found out that PG&E is allowed to call for a Demand Response (a Flex Alert) up to something like 10 times during the summer to help them manage their costs as well as the supporting grid stability.

      • Mark, for many Californians using electricity when it’s convenient doesn’t impose a huge financial burden.

        For others, like single moms working two jobs, doing laundry or cooking at odd hours to accomodate their utility (aka “demand response”) does impose a significant burden in convenience. Should it be borne by those least able to bear it?

        My service provider is Burbank Water and Power, and supporting grid stability was never an issue before San Onofre closed. Financial records now show 29% of our electricity comes from burning coal (Intermountain in Utah).

        • Bob, the issue of maintain system stability in the LA Basin has been an issue since at least 2009 when the State Water Board ordered modification or closure of the once-through cooling plants (including SONGS), the SCAQMD restricting the use of offsets in criteria pollutants for power plants, and changes in how CAISO looked at local reliability needs. (We conducted a study for the CEC on the issue.) By the way, IPP is scheduled to be closed in 2020 by LADWP, so that coal generation will disappear. Coal plants have been closing across the West will no real consequences on reliability. System-wide outages, or even transmission-level local outages, are very rare events that we focus on too much in electricity planning. Based on analysis we’ve done on both PG&E and SCE, actual distribution level outages are 15 times more likely to occur than hypothetical system outages.

          • mcubedecon, IPP can’t be “closed” by LADWP (the Los Angeles Department of Water and Power doesn’t own it). That coal generation will not disappear until 2025, when it supposedly will be replaced by burning fossil fuel methane – the same fuel which replaced San Onofre.

            Shiny new methane plants will guarantee fossil fuels relevance for at least several decades. That there have been “no real consesquences on reliability” is largely due mild weather, which is rapidly giving way to steadily-increasing average summer temperatures. Binding transmission constraints on both the CAISO grid and Southern California gas backbone, due to the haphazard replacement of San Onofre with intermittent renewable energy, natural gas, and imports, has resulted in the highest frequency of Flex Alerts since the 2000-2001 California Electricity Crisis.

            From Catherine’s colleague at Haas:
            “Market Impacts of Nuclear Power Plant Closure”
            Lucas Davis and Catherine Hausman

            Click to access WP248R.pdf

  2. I’m told that utilities think mostly about the long game, so I suspect that even in decoupled environments, “higher peaks require more investment in new capacity, which generally increases utility profits” will dominate.

    In fact, one can imagine some future in which heat waves are part of the justification of continued utility existence. Pretend there is a lot of customer-site distributed generation and storage. On most days demand on the utility system itself is very much reduced, maybe even zero. Along comes a heat (or cold) wave that requires more energy than the sized-for-average use systems can deliver and suddenly the utility’s idle capacity save’s the day. The utility business becomes a kind of insurance business. Far fetched?

  3. Heat waves often entail power outages – few utilities relish dissatisfied customers. Also, Catherine mentions “decoupling” (which PG&E has had in some form for its electric business since 1982), but another aspect is the repeated nature of general rate cases. Periodically (often on two- or three-year cycles) utilities, their regulators, and usually “the public”, relitigate the “revenue requirement”. The effects of heat waves, or under-normal temperatures, on past revenues are one of the many elements in these regulatory free-for-alls. No utility ignores that it is in a complex, evolving repeated game.

  4. I see evidence of these effects in the water sector, too. However, decoupling has complicated effects and tradeoffcs too, and is only one mechanism among others to manage load under normal and non-normal conditions.

%d bloggers like this: