(Mis)Judging Energy Hogs
Household electricity consumption is a poor indicator of responsible usage.
California’s path to lower volumetric (per kilowatt-hour) electricity prices and significant monthly fixed charges (that increase with income) has triggered some important policy discussions: What is the true marginal cost of supplying electricity? What are the alternative mechanisms for funding climate mitigation and adaptation? What’s the appropriate role of utility rates in addressing income inequality? What impact do electricity price structures have on vehicle and home electrification? But it has also revived a problematic narrative about residential electricity usage: that households consuming large quantities of electricity are less prudent or ethical in their energy use, typically characterized as “energy hogs”.
This argument has been around for decades. Twenty years ago, it was made in support of increasing-block pricing by the then-president of the CPUC. But there seems to have been little or no analysis to actually back up this viewpoint, and compelling grounds for skepticism. There are many obvious reasons one household would consume more than another that few people would suggest are wasteful or imprudent – such as more people living in the house.
Should we really expect the same electricity usage? (Source, Source)
Thus, when the current debate about fixed and volumetric charges started being presented as an issue of moral character – with one leading opponent of the proposed reform saying it would “reward the energy hogs and penalize the energy misers” and an NGO filing testimony claiming “high fixed charges [result] in the ratepayers with the least efficient consumption patterns realizing the greatest amount of savings” – I dug into some data to figure out how fair those characterizations actually are. In a new Energy Institute working paper, I present analysis that sheds some light on how much we can infer from household electricity consumption.
The first part of the paper, which I’ll discuss today, analyzes the extent to which “benign” factors drive that consumption. I’ll be back next week to talk about the second part of the paper, which looks at the role of household electricity consumption in economy-wide energy use and climate impact.
(Source)
What makes a hog a hog?
In a way, it’s understandable that those concerned with energy use would zero in on household-level electricity (and, to a lesser extent, natural gas). It’s the one component of consumption that a single entity – the distribution utility – knows (almost) completely, unlike gasoline, air travel, or all the energy embodied in the goods and services we buy. But that precision is misleading when it is used to judge whether one customer is more or less prudent or responsible in their energy use.
Piggybacking on earlier work with Meredith Fowlie and Jim Sallee – which we’ve discussed in this blog a few times – I use data from California’s Residential Appliance Saturation Survey (RASS) to examine the electricity consumption of about 32,000 households of the three large California investor-owned utilities, Pacific Gas & Electric, Southern California Edison and San Diego Gas & Electric. I start by looking at each IOU’s “energy hogs”, defined as the 20% of households with the largest net consumption (that is, after netting out rooftop solar generation if the household has solar), because that is what virtually all households are actually billed for. Each one percentile of those households is represented by a dot in the percentile line below. For brevity, I just present the results for PG&E here, because they are the middle case in most of the analysis, but the results are similar for SCE (slightly smaller) and SDG&E (slightly larger). By this definition, the “hog” households are all bunched at the top 20 percentiles of household net consumption.
—
—
Who’s a good hog?
But households have different numbers of occupants, so let’s convert the usage of each household to per capita consumption using the information in the RASS and then ask where those households originally characterized as hogs now fall among the distribution of all households. The lowest one percentile of the “hogs” on average is now at the 32nd percentile of all households, the second lowest one percentile is on average at the 41st percentile, and so forth. The “hogs” start to look a lot more like other households.
—
—
Prosumption isn’t Consumption
Now if we are talking about prudent usage, whether or not a household is also a producer of electricity because they have rooftop solar, isn’t really relevant. Electricity from a customer’s rooftop solar system is part of the grid generation mix, just like electricity another customer buys from a solar farm. So I next look at where those same households fall when the metric is gross per capita consumption, that is, adding back in an estimate of the rooftop solar production that their bill netted out. This effectively separates out the production side of what some call “prosumer” households in order to just focus on consumption for this discussion of hoggy-ness. With these two adjustments, more than half of the original “hogs” are now outside of the top 20% of consumers.
—
—
Hot hogs
Of course, with air conditioning being a major contributor to electricity consumption, we’d expect that living in a hotter climate would drive up usage even if those households are just as prudent as their halo-wearing brethren on the coast. So, I then examine where these “hogs” lie in the distribution when we also control for the average differential in usage across climate zones.
The distribution of “hogs” moves even further to the left and about one-tenth of them are now all the way down in the bottom 20%. In fact, the distribution of households originally characterized as “hogs” doesn’t look that different from the distribution of consumption among all households.
—
—
In some further analysis, I also control for demographics — children and seniors in the house, whether the home is only used seasonally, and whether an occupant works from home — and appliance fuel choices — electric heat, hot water, stove and vehicle. These all drive consumption, but have a smaller impact on where the so-called hogs lie in the distribution.
Hogs (and Angels) are a lot like the rest of us
The takeaway from this analysis is that the usage of the top 20% isn’t actually that different from other households after adjusting for three factors that few people would argue constitute hoggyness. The parallel analysis of “energy angels”, those in the bottom 20% of net household consumption, shows they also look much more like the rest of the population after these adjustments.
How much like? For a simple numerical measure, I look at the how the ratio of consumption by higher versus lower users changes with these adjustments. The average net household consumption of PG&E households in the top half of users is 212% higher than the average of those in the bottom half. But after these three corrections, it is only 54% higher. In other words, about three-quarters of the difference between the heavier-use half of households and the lighter-use half is due to the higher users having more household occupants, fewer solar panels, and living in a hotter climate. An even larger share of the difference between energy “hogs” (top 20%) and “angels” (bottom 20%) disappears when one adjusts for these three factors.
The changing face of hoggyness
Finally, I explore how these adjustments affect the demographics of households labeled energy hogs and angels. Perhaps not surprisingly, I find that the common narrative of imprudent energy hogs and socially-responsible energy angels over-represents wealthy and white households among the “angels” and underrepresents them among the “hogs”, because it fails to adjust for differences in number of occupants, rooftop solar and climate. At the same time, it greatly under-represents Latinx families among the energy “angels”.
When it comes to electricity rate design, regulators need to confront thorny factual and policy questions. It would be great if we could stop clouding those important issues with facile judgments of responsible “energy angels” and wasteful “energy hogs”.
Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.
Suggested citation: Borenstein, Severin. “(Mis)Judging Energy Hogs” Energy Institute Blog, UC Berkeley, August 21, 2023, https://energyathaas.wordpress.com/2023/08/21/misjudging-energy-hogs/
Categories
Severin Borenstein View All
Severin Borenstein is Professor of the Graduate School in the Economic Analysis and Policy Group at the Haas School of Business and Faculty Director of the Energy Institute at Haas. He received his A.B. from U.C. Berkeley and Ph.D. in Economics from M.I.T. His research focuses on the economics of renewable energy, economic policies for reducing greenhouse gases, and alternative models of retail electricity pricing. Borenstein is also a research associate of the National Bureau of Economic Research in Cambridge, MA. He served on the Board of Governors of the California Power Exchange from 1997 to 2003. During 1999-2000, he was a member of the California Attorney General's Gasoline Price Task Force. In 2012-13, he served on the Emissions Market Assessment Committee, which advised the California Air Resources Board on the operation of California’s Cap and Trade market for greenhouse gases. In 2014, he was appointed to the California Energy Commission’s Petroleum Market Advisory Committee, which he chaired from 2015 until the Committee was dissolved in 2017. From 2015-2020, he served on the Advisory Council of the Bay Area Air Quality Management District. Since 2019, he has been a member of the Governing Board of the California Independent System Operator.
Energy Institute Blog 
Severin:
Great job on showing that the “energy hogs” issue is largely a faux concern espoused by those who are trying to perpetuate a system with low fixed charges and high volumetric charges that favors the residential solar contracting industry and the owners of residential solar who skew towards the high end of the income and wealth scales. Those interests are apparently willing to raise this bogus concern in order to maintain the benefit of a low-cost connection to the grid while avoiding their share of actual grid costs, which are overwhelmingly fixed rather than volumetric.
Howard Gruenspecht
Curious: What support is there for such customers to leave the grid all together meaning having adequate battery and such an unplugging their generation from the grid? Is there a penalty of some kind? Seems to me that is the option that someone might pursue if they really want to go on their own.
I have 20 kWh battery and 5kW panel field. I buy power from the grid a few days a year. I am actively exploring ways to improve the off-grid experience.
I am actively opposing the IGFC proposal because it will delay electrification and increase utility bills
The ONLY energy HOGS are the California privately owned utilities themselves. For years they told us “Light the Night” and put up more outdoor lighting, “Ready Kilowatt” makes your life easier with more electrical appliances. Pushing Californians into getting larger and larger electrical services on their homes. The more energy they sold you, the bigger the bottom line for the utilities. Now they want to use the homeowner’s income to determine the monthly service charge even if the homeowner conserves energy or generates their own through solar power from their roofs. The CEOs of the three California utilities get 100 million dollars or more in salaries and compensation. Californians have one of the highest commodity charges in the country and the utility HOGS want more money. Soon, everybody will just stop paying them and go off grid with their own micro grid systems then bye, bye PG&E, SDGE and SCE.
“Ethical energy use. ” That’s a new term fraught with peril, reeking of presumptive judgements and pregnant with possibilities. While the point that blithely applying a stereotypical judgement laden term like “energy hog” is unquestionably lazy, it is also difficult to see how a non-peer reviewed “working paper” counts as “Research that Informs Business and Public Policy.” Perhaps this blog post will later be cited as research driven evidence justifying a per capita rate structure that requires disclosing the number of people in a home in the name of equity.
I don’t think “ethical energy use” is a new term, and it certainly is not a new concept. “Ethical” and/or “moral” dynamics — and I would add “virtue signaling” — have permeated discussions of energy use and sources since the California energy crisis and the concomitant and growing concern about climate change. Severin’s paper actually seeks to move beyond that by focusing on data, though no such study of this sort can entirely escape value judgements. And yes, his “working paper” is “circulated for discussion and comment purposes,” and is not peer reviewed, as noted on its cover page. That said, every EI@H working paper I’ve read is transparent and laden with data and methods for readers to take issue or agree with. They certainly aren’t “lazy.”
Hear! Hear!
Good comment. Thanks.
What is “unethical”? Leaving the TV on when you fall asleep or are not in the room? Running the air conditioner with a window open? How on earth can PG&E (or Mr. Borenstein) have enough data to determine who is or isn’t an “Energy Hog”?
In fact, my electric bill is determined mostly by my use of large load appliances like air conditioning/heat pump, electric oven, or charging an EV. Since none of these is unethical per se, no one is unethical. Which is more or less the conclusion of the piece.
Heating a hot tub or a swimming pool might be. Or perhaps pouring an electric vehicle which is more than average and capability such as an oversized Jeep
This is a useful discussion. There are many factors that affect electricity consumption. The household type and climate zone is already addressed by the California Baseline formula. Single family homes in hot climates get a larger baseline allowance than apartments in coastal zones.
The discussion however evades the foundation of most baseline rate designs in other jurisdictions: apportionment of the limited supply of low-cost power.
In Washington State, the Commission adopted baseline rate principles in 1980, specifically to apportion low-cost hydropower. They remain in effect, giving each household a share of the low-cost hydro at a cost-based price, and the ability to buy more expensive new resources at higher rates as needed.
The current Avista rate in Washington is an example of this:
Customer Charge: $9.00
First 800 kWh: $.082/kWh
Next 700 kWh: $.097/kWh
Over 1,500 kWh: $.115/kWh
The end-block rate is generally associated with space conditioning usage (heating and cooling), which has a very poor load factor and high peak coincidence. It should be more expensive.
Taken together, we have a simple cost-based rate design. A customer cost to recover customer-related costs of metering and billing. An initial block to recover low-cost hydropower plus delivery. A middle block to recover newer power resources plus delivery. An end-block rate to reflect the cost of serving peak-oriented space conditioning loads.
Yes, a time-varying rate can also capture this weather-sensitive load in the third block, putting the capital and operating costs of resources built to serve peak demands into the on-peak rate. With modern metering, that may be a better way to price peak-oriented usage. But the inclining block rate produces a similar result, and made a lot of sense in 1980 when advanced metering was unavailable at reasonable cost.
But really — does it matter to most households if their space conditioning load is priced in a third block at a higher price, or in a time-of-use rate at a higher price? The end result is pretty much the same.
Well, yes it does today. Households that charge their electric car will have higher usage, pushing them into the third block of the Avista rate design. BUT, if instead we offer them a time-varying rate, they can charge the car off-peak, when the system is not stressed, and enjoy economic savings while the grid operator avoids peaking capacity needs.
I know that Dr. Borenstein prefers rates that reflect marginal costs, but this Washington approach is closely connected to actual costs reflected in the utility revenue requirement for actual resources used to serve actual loads. Unlike the California income-graduated fixed charge proposal, it prices electricity based on the cost of providing electricity service.
The question that Dr. Borenstein’s post raises is whether the allocation of hydro (or other low-cost resources) should be done on a per-customer basis, or a per-capita basis. Should larger families get a larger baseline.
Which raises a question that goes far beyond the issue of electricity. Some large families live in small dwelling units, and some small families live in large dwelling units. Should the “energy budget” for a baseline be computed per capita, per square foot, or per electric meter? Should people be rewarded or penalized (or neither) for having large families (or for compressing multiple families into a single dwelling because housing is expensive)? Those are generally questions for a different forum; California addresses this in electricity through the CARE and FERA discount programs for lower-income households.
Is there some way to convince heavy users or electricity that they should generate their own via solar as opposed to taxing the grid?
I like the paper–well-written and persuasive.
Wandering outside the scope of Dr. Borenstein’s post, I wonder how the electricity hogs would fare if the metric was not solely “electricity consumption”, but also “energy consumption”, which would include both electricity and natural gas. Each kWh or therm could be converted into Btus (i.e., kWh * 3412 Btu/kWh, and therms * 10^5 Btu/therm).
Residential electric energy when used indoors is converted to heat inside the home, e.g., when powering lights or appliances or computer devices. That heat reduces usage of gas furnaces in winter, so a hog on electricity consumption might be less hoggy on a energy (Btu) basis. It would be interesting to know if switching from a electricity basis to an energy basis actually made a significant difference in identifying hogs. In the case of an electricity hog who uses a heat pump, it might make a big difference.
A metric that reflected the size of the house, Btus per year per square-foot, might also be useful to identify hogs, though, as done in the paper, it would be a good idea to correct for climate zone.
Just an idle thought–thank you.
While I agree hogs and angels are more diverse for all the reasons indicated, there is a the root of this discussion a moral impudence that implies a specific meaning to “responsible” or “prudent” use. Any metric like use per sq. ft. or per person or pre degree day, makes a moral judgement for which there is no consensus. Is it somehow more moral to live in a $5million property on the beach and consume less energy than a home in the San Joaquin valley that costs a tenth of that? Somehow it is suddenly immoral to invest in PV because you have to be affluent to do it?
The pulbic policies here can have two layers. Surely the most important one has to the look at the public costs and benefits of the usage. Per kWh that may vary with time of day (e.g. duck cruve) , but it surely does not vary with the imputed morality of the consumer. We should fundamentally let the price do its job and not moral persuasion.
The second layer can take account of fairness issues of providing help to low income families, but it should be targetted and not available to more affluent ratepayers. This gets much harder to do, of course, when other “worthy” programs are required to be funded by ratepayers.
A large and under-reported aspect of residential electricity consumption is the Home Idle Load.
It typically accounts for one third of a home’s electric use and often much more (over 50%).
It is easy to calculate using smart meter data, varies tremendously from home to home, and is climate-independent.
Years ago we worked with the NRDC to produce this related report: https://www.nrdc.org/sites/default/files/home-idle-load-IP.pdf
We also did an analysis comparing Home Idle Loads to home size:
https://1drv.ms/b/s!Ag7eOV5ifY5CgkddpxHuz9xLBCEH
Prof. Severin,
Obviously, if you control for all the factors that lead to more electricity usage (# of people in house, AC usage), you’ll find that people use similar amounts electricity. That doesn’t mean that fixed charges aren’t impacting some people more than others, proportionally.
In the discussion around AB 205, I have yet to hear a good argument for why someone like me — who lives in an apartment on the coast and uses <150 kWh / mo on average– should see my electricity bill jump by a factor of 3x to 4x simply because my household earns the "wealthy" amount of $200k/year, an income on which I challenge you to buy a home with in the Bay Area today. I've heard you cite the various incentives (NEM) for adoption of rooftop solar by wealthy homeowners as partial justification for these charges… well, I can't add solar to my roof, as I don't own my living space, and my roof is a ceiling, so I will have seen none of the benefit and will bear of the cost. I've calculated this policy is essentially adding 0.5-1.0% to my all-in effective CA tax rate.
Why should I subsidize via a fixed charge someone living in the Central Valley (with huge electricity usage), or someone charging an EV (whose usage is high enough that the new fixed charges will likely net out with reduction in rates)?
As I'm sure you've already noticed, this policy is going to be extremely unpopular across a wide range of consumers.
Brilliant.
And WHO has solar? The rich people with large homes, a swimming pool, and a Tesla who benefitted the most from the previous 4 level tiered system and so bought solar. If it is a social aim to subsidize the poor’s energy consumption, at least don’t mess up the individual behavior by messing with the rates. Why don’t we have different rates and a requirement to file your 1040 for gasoline? For groceries? So much complexity creating so many unintended consequences.
The previous 4- and 5-tiered tariff rating system forced high users to get rooftop solar to get out of the 48 cent per kilo watt hour 24/7 tariffs utilities charged and paid a premium price to buy their rooftop solar system. Utility, State and Federal incentives also helped get 1.2 million California homes to support the grid with their excess power. CARE and FERA supported the Poor from “other” rate payers already rather than from a government program made to help the poor. I am not rich and have a 1,235 sq. ft. home but had 16,000 watts of solar, 8,000 of rooftop on grid and 8,000 off grid with batteries, until they all burned up in a fire caused by illegal fireworks on July 4th, 2023, and since State Farm says the off-grid system is considered associated buildings and fences in nature, the limit how much they will pay toward replacing the system. PG&E disconnected the Natural gas and cut off the pipes at the concrete so I will no longer get Natual gas service, and they may not even give me back electrical power without increasing my electrical service to 200 Amps. The utilities are the HOGS. Onk, Onk.
Maybe it’s a case of “half full vs half empty” but I see a very different relationship in your graphic (which is a nice innovation in presentation). I see that 50% of the “hogs” are in the top 23% of energy users. While maybe not as extreme as the original look, it’s still highly skewed.
As for the adjustments, a linear per capita usage adjustment probably isn’t correct. There are strong economies of scale when living in a house. It takes the same energy to cool a house regardless of the number of people in the house. (And even if it scales with household/house size, it’s not linear.) Lighting has a similar relationship as people share living space simultaneously. I remember seeing a study that showed that wash loads were not highly correlated with household size. And with Prop 13 property incentivizing older residents to remain in larger houses, household size may not be strongly correlated with house size.
An important point about prosumption–residents are paying directly for a large share of their generation, and much of their generation (if not all) is not leaving their neighborhood where its picked up by their neighbors instead of using grid power. Given the very high marginal costs of transmission and distribution (those average costs are rising so marginal costs must be at least equal to or higher than average costs), that leads to additional grid savings.
Is there a rationale for bundling households together in clusters who share consumption and generation making them nearly independent of the utility? Yes, there’s a need for using local wires to shunt power back and forth and reimbursing utility for associated maintenance and repair of said wires, but other than that what’s needed is a cost sharing arrangement among members of the cluster. I would think it would be too hard to measure and bill members of cluster for all the costs turning it into a small scale competitor of utility, but maybe not?