Limitations of Standards
U.S. energy-efficiency requirements for air conditioners illustrate the inherent limitations of standards.
This was a hot summer. July 2019 was the hottest month ever recorded globally. Even in temperate Oakland, California, where I live, there were a lot of people wishing they had air conditioning.
In Oakland it only gets hot a few weeks a year. We don’t need the latest most energy-efficient air conditioner. So why don’t you see my neighbors going out to buy cheap air conditioners?
Because they can’t. In California, we have the same energy-efficiency requirement for new air conditioners as Nevada, Arizona, New Mexico, Texas, Oklahoma, Arkansas, Louisiana, Mississippi, Alabama, Georgia, Kentucky, Tennessee, Virginia, North Carolina, South Carolina, and, Florida.
Weather varies enormously across these states, but air conditioner standards do not. A one-size-fits-all standard requires too much energy-efficiency in states like California and not enough in states like Florida. This coarseness is one of the major limitations of standards, and one of the key disadvantages of standards relative to price-based policies.
Lots of Variation
The map below plots annual cooling hours by state. This is the number of hours per year for which a household should expect to use an air conditioner.
Americans experience an average of 1265 “full-load” cooling hours per year, but there is large geographic variation. Within the continental United States average annual cooling hours range from 310 in Maine to 2,771 in Florida, almost a 9∶1 ratio.
Differences across individual cities are even larger, from below 250 in cities like San Francisco, to 3,900 in Miami and even a whopping 4,500 in Key West, FL (yikes!). Oakland is a bit warmer than San Francisco, but still has only a tiny fraction of the cooling hours experienced elsewhere.
Coarse Standards
If you were expecting standards to reflect this underlying variation, you are going to be deeply disappointed. Up until recently, there was only a single national standard for central air conditioners. It didn’t matter where you lived or what your weather was like, there was a single one-size-fits-all standard.
Since 2015, we’ve had separate standards for “North” and “South”. Many people worked hard to get this regional-differentiation in place. This is a step in the right direction. It just doesn’t go nearly far enough. Two regions can’t capture the underlying richness in demand for cooling.
I don’t think more regions is the answer. Even just within the Bay Area, cooling hours vary significantly depending on exactly where you live, and I can’t imagine the logistical, administrative, and legal challenges of moving to a system with hundreds of different stringencies.
Too Much of a Good Thing?
So what? Can you really have “too much” energy-efficiency? Yes. We’d all prefer an air conditioner that costs less to operate, just like we’d all prefer a car with higher MPG, but it costs more.
For air conditioners reviewed in a 2019 Consumer Reports article, there is a strong positive correlation between price and energy-efficiency. How much does it matter? The EIA central air conditioner calculator assumes than an 11% increase in energy-efficiency from SEER 13 to SEER 14.5 costs $556.
I have no idea if $556 is the right number. The academic research on this is quite mixed (see, e.g., here, here, and here). But if it were free to increase SEER then all the manufacturers would be doing it. In fact, the very existence of the “North”/”South” duel standard is an acknowledgment of a cost-efficiency tradeoff.
The broader point is that, regardless of the exact magnitude of the tradeoff, energy-efficient air conditioners matter more for Key West than they do for Oakland. Standards are not an effective tool for getting this allocation right.
This Isn’t Just About Air Conditioners
Air conditioners are a particularly lucid example. But the exact same issue arises with natural gas furnaces. In the United States, heating hours vary even more than cooling hours. The DOE’s two (or in some cases three) regions are too inflexible to capture this richness.
A similar issue arises also with cars and trucks. Some people have long commutes and drive 20,000+ miles per year. Others bike to work and leave their cars mostly in the garage. With passenger vehicles, we have a “fleet average” standard, which builds in some inherent flexibility, but there is still too little attention paid to this underlying variation in usage.
Yes, even light bulbs. My colleague Catherine Wolfram has written about the incandescent “relic” in her basement that gets only 20 minutes of usage a year (10 minutes at Halloween and 10 minutes at Christmas). LEDs have come down so much in price that perhaps it doesn’t make much difference, but current lightbulb standards lump Catherine’s basement bulb right along with bulbs used 1000s of hours a year.
Better Prices and Better Information
It’s no mystery that economists prefer price-based policies over standards. The ideal would be to have all prices set equal to social marginal cost. Where operating costs are high from a societal perspective, consumers would invest a lot in energy-efficiency, and where operating costs are low from a societal perspective, consumers would invest less.
Better information would help too. All major appliances in the United States are required to be sold with yellow EnergyGuide labels. However, the current labels provide only very coarse information based on national average usage. In a recent paper, economist Gib Metcalf and I use an online experiment to show that better labels would lead to significantly better decisions. The average energy-efficiency of selected air conditioners stays about the same, but the allocation is much better. That is, consumers facing high operating costs invest more in energy-efficiency, while consumers facing low operating costs invest less.
U.S. energy-efficiency requirements for air conditioners are simply too coarse an instrument to reflect the underlying variation in usage. This coarseness misses the variation in greenhouse gas emission reductions, too. This is not the only limitation with standards, however. Economists have also pointed out that standards can discourage consumers from buying a product at all. Come to think of it, maybe I’ll just keep using a fan.
Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.
Suggested citation: Davis, Lucas. “Limitations of Standards”, Energy Institute Blog, UC Berkeley, September 16, 2019, https://energyathaas.wordpress.com/2019/09/16/limitations-of-standards/
Categories
Lucas Davis View All
Lucas Davis is the Jeffrey A. Jacobs Distinguished Professor in Business and Technology at the Haas School of Business at the University of California, Berkeley. He is a Faculty Affiliate at the Energy Institute at Haas, a coeditor at the American Economic Journal: Economic Policy, and a Research Associate at the National Bureau of Economic Research. He received a BA from Amherst College and a PhD in Economics from the University of Wisconsin. His research focuses on energy and environmental markets, and in particular, on electricity and natural gas regulation, pricing in competitive and non-competitive markets, and the economic and business impacts of environmental policy.
Energy Institute Blog 
Lucas, we are tackling some of the issues you raise by complementing minimum energy performance standards with two other strategies that make markets work better for consumers: 1) eliminating the market intransparency barrier with a zero to 100 relative energy efficiency index on every product model, so consumers can choose the most efficient models to meet their needs (including price point) and 2) eliminating behavioral barriers, some of which were mentioned by other commenters with a combination of the Enervee Score, personalized bill and cost of ownership and aggregated reviews (e.g., rational inattention, lay theory that efficiency necessarily costs more, misinterprtation of the ENERGY STAR label as a benchmark for super-efficient products). These strategies are embedded in a choice engine platform that empowers consumers to choose the efficient products and clean vehicles they aspire to, without mass market incentives. As an example, see cars.sce.com and marketplace.sce.com. Our daily-updated market intelligence actually show that when considering retail price points, there are a number of product categories where efficiency is not correlated with price, and in other categories, it’s often possible to find deals on super-efficient products that make them cheaper than less efficient alternatives.
A third strategy is to augment minimum standards with programs that leverage a dynamic reach performance benchmark to stimulate innovation at the cutting edge of technology. We have been discussing this approach with the California Energy Commission and have the platform to roll out super-efficient performance specs across 20 product categories (see enervee.com, blog.enervee.com). Would be happy to discuss with you.
Anne Arquit Niederberger, Enervee
anne@enervee.com | blog.enervee.com
This discussion has gotten hung up on the “efficiency” of an air conditioner as though the “box” is the most important component to costs of occupant comfort and its influence on consumer decision-making. The thermal properties of the building are at least as important. These are far more difficult to tackle especially for low income portions of the population where large upfront investments are needed to produce a stream of future savings. Adding further complexity is that changes in the thermal properties of the building also affect the space heating service. In much of the country natural gas, propane or fuel oil is the dominant space heating fuel, and these direct combustion emissions (and a variety of secondary ones) are a major source of GHG emissions that will not be changing just because the electricity generating fleet is shifting to less carbon-emitting technologies.
Given the current enthusiasm for building decarbonization, what is the proper mix of energy efficiency programs, equipment efficiency standards, and price signals that can move consumers toward socially efficient building decarbonization?
You make an interesting point, but, the way I see it, national efficiency standards are meant to set a floor—possibly the lowest common denominator for most states—while states where tougher standards are necessary and cost-effective can set their own on top of this floor. This kind of a system makes sense to me. Barring the ability to differentiate standards by geography even semi-well (as you say, this can become too administratively burdensome quickly) or price appliances by SMC (also tough, and I presume SMC varies), what do you propose as a better pathway? I definitely think better labeling has promise, but this should be in addition to, not instead of, some baseline set of national standards.
Perhaps U.S. energy-efficiency requirements for air conditioners illustrate the inherent limitations of standards. However, seems to me they also illustrate the limitations of some (most perhaps, but surely not all) economists’ preferred price-based policies over standards. When you can get all prices set equal to social marginal cost please let us know. And even then you’ll still need standards, given the limitations of price based incentives to drive efficient energy related decision making given that energy is rarely the main game in consumer thinking about their choices. This is well covered territory in the energy policy space – coherent and comprehensive policy frameworks are needed rather than one ‘carbon’ price policy to rule over all. And the test of a policy instrument is not that it works in all cases, such as Catherine’s ‘relic’ in the basement, but provides overall social benefit compared to no policy, or the other policy options that are actually ‘actionable’.
Good points all, Iain. It seems discussion here is primarily driven by two social considerations: 1) policy which allows access to air conditioning for as many people as possible during hot weather, and 2) policy which limits the carbon impact of that access.
Note that neither involve efficiency, nor should they. In developed countries, it’s currently popular to consider energy efficiency analogous to energy: using less energy is the same as having more of it. One is the inverse of the other. By extending the logic, its failings rapidly become evident: using one-half as much energy to meet a specific need is patently less useful than access to twice as much energy, to meet any need. Efficiency improvements approach a hard limit, energy generation is theoretically unlimited. Moreover, by establishing efficiency standards for air conditioners, we (ironically) are using horribly-inefficient policy to accomplish that goal: now, we must establish efficiency standards for washing machines, for dryers, for ovens, vacuum cleaners, etc., taking all considerations of location and economic status into account.
Efficient policy prioritizes clean generation. As carbon emissions from generation approach zero, so does the social relevance of energy efficiency standards for all appliances. With manufacturers freed from obligatory standards, free market competition will make efficient, affordable appliances available to everyone.
Increasing access to air conditioning and lowering carbon emissions both begin with efficient policy, not efficient air conditioners.
I agree that calculating social marginal costs can be a fool’s errand.
However, I take issue with the author’s statement “if it were free to increase SEER then all the manufacturers would be doing it”. That works in the minds of economists but not in the real world. As mentioned by Scott Sklar below,
“We should not overlay “rational” behavior with “actual behavior.”
This applies equally to manufacturers as is does to consumers. Manufacturers price their products in the market just as irrationally as consumers make choices about what products to purchase.
For example, appliance manufacturers could have installed standard remote controls on water heaters at very low incremental cost many years ago. They were encouraged to do so many years ago by conservation groups in the Northwest US, where I live and work. This is a pretty close approximation to the “if it were free” statement by the author. This simple additional remote control functionality would allow utilities to implement demand response programs across the continent. This direct utility control of demand response during high demand times of day saves energy on the distribution and transmission lines (by avoided the highest line losses when lines are fully loaded) as well as avoids running higher marginal cost resources during those periods. Manufacturers chose not to do so for many years because of market risk, lack of national standards, lack of customer demand (for something that customers have no knowledge of), primarily reasons having more to do with marketing and marketing risk that production costs.
Nonetheless, if there were a national standard requiring all water heater manufacturers to install these demand response enabled units, with a reasonable lead time, I would find it hard to argue that this would provide too much conservation or that this standard was inappropriate for any place in the continental US. In some cases one-size might actually fit all.
On the other hand the NW Power and Conservation Council leads the way in the NW for measures that don’t seem to be in the one-size fits all rubric. The NW Power and Conservation Council uses three heating zones combined with three other cooling zones as it evaluates the efficacy of measures base on different climatological differences in the region (WA, OR, ID, MT). If other regions would follow that model, some of the Bay Area issues mentioned by the author might be resolved. Marin and SF Counties might fall into zones separate from Contra Costa and Alameda Counites.
It takes time and effort to do the extra work to make the analysis more granular. But this is being done routinely in the Pacific NW. Bonneville Power Administration and energy utilities voluntarily pay that bill in the NW because it has been demonstrated to be an excellent investment over time for the region and its ratepayers. This has been an approach that has worked well in my region for a few decades.
California has something on the order of 16 climate zones, and the building standards and other EE programs are evaluated on the basis of these: https://ww2.energy.ca.gov/maps/renewable/building_climate_zones.html
I think the premise that “its is only hot a few weeks a year, so we don’t need one type of standard” is a faulty premise.
That is why another level of research is needed. We are finding out that people who have air-conditioning use the
AC for extended periods, even when not too hot, and that increases with age of the resident. So what should be a few
weeks turns into four plus months. We should not overlay “rational” behavior with “actual behavior.
Adj Prof Scott Sklar
Energy Director
Environment & Energy Management Institute (EEMI)
and Acting Director, GWU Solar Institute
The George Washington University (GWU)
https://eemi.seas.gwu.edu/ sklar@gwu.edu
Personal email: solarsklar@aol.com Ph 703-522-3049
Two issues to consider:
1) The problem of “agency”–Air conditioners generally installed in 3 situations: when the house is built, when a resident desires more cooling or is remodeling, and when the old one fails. The dominant purchases are during the first and third situations. In the first, the home builder does not fully consider the future operating costs for the home owner (and this is even worse for rental properties.) The recent push for all-electric ordinances reflects this situation. Developers also build for a “lowest common denominator” consumer that desire a certain set of traits that disregard costs. In the third, the resident is looking for a quick replacement and rarely weighs the relative features of different ACs. They go with the “box” that can be installed the most quickly at the cheapest possible cIost. (Both of these observations come from discussions with developers and installers.)
2) The transaction costs of searching for the right price/efficiency combination can be high. Individuals situations also vary greatly and the current labels do little to solve that problem. It’s not always possible for consumers to accurately weigh their relative benefits.
“In California, we have the same energy-efficiency requirement for new air conditioners as Nevada, Arizona, New Mexico, Texas, Oklahoma, Arkansas, Louisiana, Mississippi, Alabama, Georgia, Kentucky, Tennessee, Virginia, North Carolina, South Carolina, and, Florida.”
Only true for minimum SEER, which isn’t a great efficiency metric in the dry southwest US. AZ, CA, NV, NM have an additional requirement for minimum EER, which does better reflect AC performance in hot, dry climates. HVAC industry tried hard to get rid of the EER requirements during last updates of federal standards.
“A one-size-fits-all standard requires too much energy-efficiency in states like California and not enough in states like Florida.”
That depends on the purpose being served by appliance efficiency standards, which came about in the 1970s in response to the OPEC oil embargo. Originally they had nothing to do with emissions – their purpose was conserve energy in the name of national security, when actually running out of energy was perceived as a credible threat. Since, the purpose of efficiency standards has morphed into preventing carbon emissions, and on a one-atmosphere-fits-all planet, a one-efficiency-fits-all standard is appropriate.
Whether the cost of air conditioning should be “fair” is a social argument more than an economic one. Of course, we’d all prefer to have an air conditioner that costs less to operate. We’d all prefer to have free air conditioning, but we can’t. We are, however, reasonably free to move about wherever they like in the U.S; if we move to the South, we can expect to pay more for air conditioning. It’s a choice. In an idealized situation where electricity is generated with no emissions at all – effectively, a socialized marginal cost of zero – standards are unnecessary, and it doesn’t matter where we live. The market will price air conditioners based on the value of having one that costs less to operate, with no adverse environmental effects either way.
Attempting to game demand vs. supply to protect the environment will always require making a free, competitive market for appliances less free and competitive – another argument for the benefits of providing zero-carbon energy from well-regulated, monopoly sources.
First, the social costs of AC use varies by region because GHG emissions per kWh vary. Second, the point of this post is that uniform standards can present economic barriers to some consumers replacing their inefficient units with more efficient ones. The question to ask is whether it is better to have a mix of old inefficient and new single efficiency units, or to have a smaller amount of old units and mix of efficiencies for new units. I’m agnostic on the question that needs to be answered either empirically or with modeling/simulations. It cannot be answered with simple assertions.
And no, none of this justifies continued reliance on poorly regulated central utilities, particularly when they have a continuing history of four decades of poor decisions.
“First, the social costs of AC use varies by region because GHG emissions per kWh vary.”
Far easier and more cost-effective to reduce carbon emissions en masse at centralized sources of generation, than by subsidizing an unlimited number of distributed sources. Germany will miss its 2020 emissions target by at least 20% taking that ill-fated route.
“Second, the point of this post is that uniform standards can present economic barriers to some consumers replacing their inefficient units with more efficient ones.”
Standards, modeling, social costs, etc are irrelevant, for all their appliances, when customers are provided with zero-carbon electricity. Then, the value of efficiency is completely at the discretion of the consumer – as it should be. Mark Jacobson still insists a 100%-renewable grid is possible, but apparently you are now resigned to the fact it’s not. Good start.
“And no, none of this justifies continued reliance on poorly regulated central utilities, particularly when they have a continuing history of four decades of poor decisions.”
I don’t know anyone who doesn’t rely on a central utility, and its self-financed distribution network, to provide a reliable supply of electricity – but of course you’re free to part ways with PG&E. After you’ve spent not four decades, but four days trying to get by without that poorly-regulated utility, you’ll be entitled to extol the virtues of “distributed energy resources.” I eagerly await your report.
The reason why policymakers moved to subsidizing distributed resources was precisely because the process of replacing large infrastructure pieces takes too long. In the U.S., the pressure of distributed resources is now making coal generation uneconomic. See https://www.utilitydive.com/news/majority-of-coal-plants-are-uneconomic-to-nearby-wind-solar-report-finds/551187/ as one example analysis.
Yes, in the far future when we have zero emission energy, all emissions will be equally zero, but of course, then the relative efficiency won’t matter from an emission standpoint! You’re arguing a tautology. We are discussing how we move forward TODAY in TODAY’s conditions, where emissions per kWh vary across local regions. In TODAY’s case, the social costs vary by region.
I would assert that the idea that customers with economic barriers are not replacing their older inefficient air conditioners is not that they would if they had the money or if the air conditioner cost less. It is that they will not replace their AC until it breaks.
I agree (except when they are remodeling). This raises the “agency” problem–they take the AC unit that the repairman suggests with little consumer research. So its the HVAC contractors that have to be educated/persuaded in some fashion.
Lucas,
You say “The ideal would be to have all prices set equal to social marginal cost.” Agreed, but SMC is also geographically differentiated (e.g. marginal GHG and criteria emissions from electricity, congestion impacts of driving…) so I’m not sure that would help with this particular issue.
BTW we also live in an area where A/C is really only needed a handful of days a year. I got a micro swamp cooler that blows air over some kind of special ceramic (and water from a small reservoir) and gives a 10 degree F temperature reduction where you point it. It’s not enough to cool even a small room, but it also costs less than $100 and only draws 10W from a USB input, so you can even take it camping with a portable battery.