Skip to content

The Decline of Sloppy Electricity Rate Making

Back in the “good old days” most customers had no choice about how to buy electricity and a regulator’s life was pretty easy.  The utility needed sufficient revenue to cover its costs, but the regulator approving rates was mostly just deciding whose ox to gore.    How much should industrial customers cover versus commercial or residential customers?  Is a fixed charge fair to those who don’t consume much?  That sort of thing.

Of course, even back in the old days some customers had choices, particularly large industrial firms.  They could self-generate if the utility tried to charge them a price that was too high.  And if they hadn’t already set up shop in the utility’s territory or weren’t too invested in the area, they could take their demand elsewhere.  Regulators were pressured not to foist too much cost on the large customers who had an option to bypass the utility in whole or in part.  That showed up in rate design and, sometimes, in customer-specific arrangements.


For those customers, rates were set to reduce so-called “inefficient bypass,” which described when a customer would find an alternative supplier (or self-generation) that wasn’t actually lower cost than the utility, but offered a lower price.  Avoiding inefficient bypass meant the utility tried to keep customers for which it was the lowest-cost supplier by setting price close to that cost.

Luckily for utilities — and for the stress level of regulators — few customers had real bypass opportunities in those days, certainly not residential or small commercial customers.  But that luck has run out; technology is now making every customer a potential bypasser.

Rooftop solar panels are the leading bypass mechanism for small customers.  They make economic sense for the customer so long as the retail price of the kilowatt-hours crowded out by the solar generation is greater than the cost of solar electricity.  But, as I’ve discussed previously, they are only efficient for society if they lower the overall cost of supplying the electricity needed on the grid.  The gap between retail price and avoided cost opens the door for inefficient bypass.

And solar panels aren’t the only bypass news.   With low natural gas prices, combined heat and power (CHP) installations onsite can lower bills for some customers.   Fuel cell technology continues to advance, pushing closer to the retail cost of electricity.  Batteries can store power from the grid or from onsite generation at lower and lower costs, making it easier for customers to rely less on the grid or to choose when they want to rely on the grid.


Except for some very particular narrow applications, none of these technologies lowers total grid costs by as much as it lowers customer bills, which means bypass leaves the utility with a revenue shortfall.  The potential shortfall, and the need to then raise prices, and the resulting incentive for more bypass, has been dubbed “the utility death spiral.”

The drama, and implied visuals, of a utility spiraling into the abyss (exactly what? a power plant?) creates lots of excitement, but the phrase leads us away from the real issue.  No technology available today — or likely to be available for years to come — will lead more than a fraction of the customer base to fully cut the cord, and operate without the utility.   Decades from now, most customers will still want access to the grid, and will still need the utility.

Something is dying alright, just not the utility.  It’s the ability of regulators, utilities, and interest groups to push around revenue collection among customers without the customers pushing back.

  • Try to punish high-consuming households by raising their price many times above cost – as has been done in California for the last 15 years – and they will now install solar to reduce their grid purchases, undermining revenue collection.
  • Try to use “demand charges” that are based on a customer’s peak usage — regardless of whether its peak coincides with system peak — and soon they will be installing batteries to smooth their peak, but in many cases without helping to lower grid costs.
  • Try to raise retail rates for most customers in order to offer discount electricity to low-income households and the high-price customers will turn to all forms of distributed generation instead of subsidizing the poor.
  • Try to stick commercial and industrial customers with more of the utility costs and they will invest in CHP and other onsite technologies.
  • Try to encourage demand shifting to off peak with exaggerated peak-period prices during all summer weekdays and the customer will use batteries to shift not just on the hottest high-demand days, but also on days when there is no benefit to society, though still an arbitrage play for the customer.

Smart Meter

You may agree with the equity goals behind some rate design choices and may disagree with others.  That’s not the point.  The point is that technology is making it ever easier for customers to respond to prices, and to arbitrage between price differences.  That’s great news when those prices and price differences reflect real cost impacts, because customers can respond to efficient cost-based prices with efficient actions.  But when the prices don’t reflect costs, customers are still going to respond, and that will undermine system efficiency.

That means that the flexibility regulators have had in designing retail rates to pursue other goals – whether helping the poor, subsidizing grid-scale renewables, paying for energy efficiency programs, or just keeping rate design “simple” – is going to come under increasing pressure by market participants ready to exploit any price wedge, whether it is based on a real cost differential or not.

Economists have for years argued that utility rate design should follow cost causation principles, because departures from cost will lead to inefficient customer response.  Regulators have often paid little heed largely because the inefficiency was small when customer ability to respond was limited.  That left regulators a free hand to harness rates for pursuit of other policy agendas.


Distributed generation, storage, electric vehicle charging, and smart customer-side usage technologies (think controllable communicating thermostats) mean that the inefficiencies from sloppy rate design – prices that depart substantially from cost – will be magnified.

But the flip side is that the opportunity to incent efficient customer-side participation in the market with smart rate design is greater than ever.  And that opportunity will grow exponentially in the next few years as we see continued improvement in generation technologies, batteries, and sensors that can control a panoply of household activities.  Accurate cost-based pricing can not only lower costs, but can also use customer-side participation to gain the flexibility that will be required to integrate more wind and solar power.

The pressures to align utility rates with costs are only going to increase.  Here’s hoping regulators will harness these changes to reduce total system costs and smooth the integration of intermittent generation.



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.

33 thoughts on “The Decline of Sloppy Electricity Rate Making Leave a comment

  1. Severin, your article might have as aptly been titled, “The Dismemberment of Cohesive Enviornmental Policy”. You note that

    “Something is dying alright, just not the utility. It’s the ability of regulators, utilities, and interest groups to push around revenue collection among customers without the customers pushing back.”

    Something else is dying – the ability of regulators to have any control whatsoever over emissions. A bill in Congress, introduced by Angus King (I-ME), would allow any individual to contribute their own fossil-generated electricity – and get paid retail rates for it.

    Do you think it will be easier to account for carbon generated by millions of individual consumers than that coming out of a point source like a power plant?

    Do you think generating electricity, as needed, from a single point source like a power plant is less efficient than generating electricity at millions of consumer locations, which is then stored in batteries?

    Do you think allowing everyone to generate their own “fuel cell” (i.e., natural gas) electricity will lessen society’s dependence on fossil fuels?

    Think again. This push towards individual empowerment, like the NRA’s gun rights juggernaut, is positively medieval in its sophistication. It will likely have similar ramifications for society as a whole.

  2. Our utility bills do contain a lot of taxes; but they were legislated by our state legislature. The utilities didn’t ask for all those taxes to appear on bills.

    • That statement is not entirely true–there are many examples of public purpose programs promoted by the utilities, Regardless though, that the taxes are legislated doesn’t mean that they are not also at risk. The decline in sales tax revenues as purchase of services and online sales increase is an example.

  3. Being technology the greater equalizer, the question becomes now how is this utility industry going to be regulated moving forward. This is one of the areas where common sense needs to kick in. Is the consumer paying the real price of this new technological development or is the government helping to subsidize? If the latter, we are not truly yet living the world of open market forces.

    • Determining what “real price”, “subsidy” and “open market forces” mean is much more difficult than it first appears. Setting the right frame of analysis is important; in general I’ve found that engineers typically get this wrong. Throw in uncertainty and risk, and it gets even more complex.

      • Ha, well, I think just about everyone typically “gets this wrong,” economist included, who sometimes forget, for example, that all the definitions you mention have a normative component. These discussions often lead to impasse because the participants are holding unspoken and differing underlying goals.

        • You’re right that often the premises are unstated, but there are in fact positive definitions of at least the first 2; “open” is impossible to achieve so that is probably left to normative description. And given that there are actual definitions, analysis can make these transparent and the parties can then understand where each stands, which can then lead to negotiation. I’ve been part of two substantial utility resource transfers that were in large part facilitated by improved transparency in these analyses. I’ve found that the parties that are either most interested in protecting the status quo or in using outside popular political forces to sway the decision are the most likely to keep the process as opaque as possible.

      • I think anybody who thinks he knows the “real price” of electric service is kidding himself. Would that be the real price with or without 100+ years of utility regulation and its impacts on investment, not to mention the incentives for innovation over that period. Would that be the real price with or without all externalities figured in? Or the real price including or not including subsidies, foreign and domestic, for the R&D and manufacture of generation and transmission technologies, many of which are implicit in nature?

        There is no “natural state” of the electric power system, not in the short term, and most assuredly not in the long term. People made and continue to make decisions, and we got the system that “we” collectively wanted, through the traditional bizarre uber-complex politico-economic weighing of preferences. Moreover, we are still make such decisions, and always will.

        I know I’m being somewhat eccentric, talking at a very abstract level when most of us in the power business have more concrete problems: determining the price of an incremental kWh tomorrow at 7pm, or the cost to build an incremental kW of capacity with characteristics x,y,z.

        But Severin’s original post is not about a specific rate design question, but about process rate design generally, so I believe my rambling is actually on-topic. The choices about how we think about rates will affect how and how much money flows into the system, paid for by who and to who, and that will in-effect affect investment and operational decisions in the electric system and also on the demand side, and that will affect the price of the service.

        Example: In CA, we have tiered retail rates because … reasons, and we now have a sizable industry that is essentially powered by that little ratemaking “quirk.” And that industry has moved the needle on the price of energy and will certainly move it further.

        So, back to my original point. Power system analyses that don’t even consider “well, what do we actually want?” are naive, and will almost certainly get you, in the end, the power system that somebody a bit more cunning than you wanted.

        • Yes, the “real” price is context specific, and it reflects a cumulative set of societal choices made over history. But that still can be described in a positive state. Yes, there is a need to focus on the normative aspect, and in no way am I advocating an analysis that does not consider our objectives. But there are underlying calculations that can be done without relying on a policy goal. An analysis need not start with a policy objective in mind. And transparency, as I discussed, can even mitigate that bias to a large degree.

  4. Severin, I always enjoy your missives–and you’re always right! I’d add the expansion of another customer option to the mix that is gaining some momentum; Community Choice Aggregation (Energy).

  5. I presented in a workshop at the CPUC on ratesetting methods at the CPUC as part of the SCE GRC. (We had just settled the revenue allocation portion of the case.) I posted the presentation here:

    Along with the issue of collecting “public purpose” revenues, the other big issue yet to be addressed is how to incorporate shareholder risk into rates. Assuring shareholder earnings in the face of declining market share is not a sustainable model. I can think of several ways to address this, with the worst one allowing regulated utilities to compete directly with others on the customer side of the meter, or even handing that market to the utilities.

  6. The blog is correct that utility demand charges and excessive on-peak / off-peak differentials will drive customers to use storage and shaping options (including thermal storage by pre-cooling or putting timer controls on water heaters) to avoid sloppy pricing.

    But it is missing the most important mistake being made by utilities and regulators — attempting to charge every customer (regardless of usage) a sizable fixed charge to access the electricity grid. These will push small users to uneconomic bypass by improving efficiency, installing batteries, and implementing informal grid connections (an extension cord to the next-door-neighbor), once again reducing the customer’s bill, but possibly not the total societal cost.

    We spent a decade trying to get rid of master-metering in order to give every customer responsibility for their own usage decisions. But in states with high fixed charges, we are seeing accessory dwelling units (ADUs, or “mother-in-law apartments) converting back to electricity service through a single meter at the property. This is a rational response to irrational pricing.

    We are seeing bypass by the thousand across the country; most new pedestrian crossing signals that are installed near schools now use solar, battery, and no grid connection. High-efficiency LED lighting helps make this cost-effective. They are doing this even though the utility grid is typically immediately adjacent to the location, and the lights operate at completely off-peak periods (shut off by 4:00 PM). Because utilities are charging $10 or $20/month for a grid connection, these schools find it cost-effective to go solar for an extra $800 per installation.

    No competitive industry charges for a grid connection. They cannot. The market won’t let them. When you go to Home Depot (and connect to the hardware grid), or to the supermarket (to connect to the grocery grid) or to the mini-mart (to connect to the gasoline grid) you pay for exactly what you buy; all of the costs of the grid — stretching around the world for these products — is built into the volumetric prices of the products they sell. The customer bears the cost of connecting to the grid — but all the other costs of the grid are built into the unit prices.

    The purpose for which regulation was created was to prevent the exercise of monopoly pricing powers by entities with a de-facto monopoly (see Munn v. Illinois, 1877). At that time, it was grain elevators and railroads, with exclusive access to groups of farmers. Today it is electric and natural gas utilities.

    The customer should bear the customer-specific cost of connecting to the grid — the service drop from the pole to the house — but all other distribution system costs should be built into the unit price of electricity, just as the market does for cookies, plumbing supplies, and gasoline. And utilities are moving away from this long-recognized standard — and quite possibly committing financial suicide in the process.

    40 years ago, the only off-grid options were diesel generators; today there are more and cheaper options. RMI’s recent paper on Grid Defection shows that utilities may become uneconomic in the future. With poor rate design, they will chase away the 2 kWh/month customers now (the pedestrian crossing signals), the 20 kWh/month customers next (the outdoor advertising sights) and the 200 kWh/month customers after that (every small apartment).

    They can retain these markets by pricing their products more like competitive businesses do — on a volumetric basis, differentiated by season, time-of-day, and location (Home Depot costs more in Hawaii than in Indiana; groceries cost more in Juneau than in Jacksonville — but it’s all volumetric in all of these places).

    Finally, in response to those who raise the “Costco” issue of paying an annual fee for membership stores:

    a) Costco gets only about 2% of its revenues from membership fees; if we were talking about $2/month utility fixed charges (of the average utility bill of about $100), this would not be an issue; utilities are seeking ten times that amount.
    b) Costco does this specifically to keep lookie-loo SHOPPERS out of the stores, so that people can efficiently maneuver around the store to fill their carts and BUY stuff; you NEVER see someone leaving Costco with an empty car. You go to the mall to SHOP. You go to Costco to BUY.
    c) Going to Costco is the equivalent of being a primary-voltage utility customer: you pay a little more to “connect to the grid” at the primary voltage level, but you pay less for your electricity; and
    d) Costco offers an Executive level of membership for $110/year; those customers get a 2% rebate back on all merchandise purchases; my own rebate every year approximates my $110 membership, so it’s really a “minimum bill” form of rate design, somewhat similar in character to that which the California PUC just approved for the California utilities.

    See our new publication, Smart Rate Design for a Smart Future, available at

    • Jim, the work of RAP is extremely helpful and thought provoking. I was just speaking with Rich Sedano this morning. However, I find some of the comparisons that are made here to be stretched beyond value.

      Electricity is truly a unique product with production and use balanced in real time. Inventory, while possible through storage, is expensive and is not a significant part of the system. The analogies you make with hardware, groceries, and gasoline all have significant inventory in the supply chain, probably on the order of weeks for hardware and days for gasoline and groceries depending on what type of groceries you are talking about. We are nowhere near having that kind of storage in the electric system. Stockouts are not the end of the world in many of the industries you mention, but when the electricity goes out, everyone knows about it instantly and they get quite upset if it is not restored quickly. Inventory is exactly the reason that rates for natural gas service are not receiving nearly the attention and feedback that they are for electric service.

      This discourse will continue to be important, and the contribution from RAP is appreciated.

      • Kevin’s reply — that electricity is “truly unique” — is commonly uttered, but not easily supported.

        My supermarket has no “obligation to serve” like my utility assets that it has — but my utility has twice in the past year failed to be there when I needed it, and my supermarket (to date) has 100% availability for the year. The last time my supermarket went down was when their electricity was out for four days. They’ve never failed to have tomatoes when I wanted them; once they did not have ORGANIC tomatoes, but my power company does not let me even have the option of choosing where my electricity comes from; I always get generic electricity from them. Even their “green power” program consists entirely of buying RECs, which I can do on my own.

        The point is that the “obligation to serve” that my utility assets may not be as compelling as the “desire to serve” that affects my supermarket. The purpose of regulation is to enforce on monopolies the pricing and quality of service discipline that markets enforce on competitive industries.

        Now, the storage issue is real, and the utility might address that by starting with something Sev and I have both advocated — more granular time-differentiated pricing. Let me connect to the grid for the cost of connecting to the grid, but when I deliver power TO the grid, credit me based on its value when supplied, and when I receive power FROM the grid, charge me based on its value when I take it.

        If that pattern requires storage, they the power supplier (utility or competitive supplier) will acquire storage to serve us, or we will acquire our own storage. I live in the Pacific Northwest, where there is approximately 70 million acre-feet of storage in hydroelectric reservoirs. That’s cheap storage.

        Controls on water heaters, and ice-storage for air-conditioners are also cheap storage. Give us time-differentiated prices, and those industries will respond. And other forms of storage can also evolve, including water and wastewater system pump and tank regulation, frozen food warehouse regulation, and pre-cool and coast-through refrigerators will find a place in the market.

        But exercising monopoly power in the pricing of grid access is not acceptable; that’s what regulation was created to prevent.

        • Jim, I fully agree with your response, and would go one step further: exercising RISKLESS monopoly power in grid pricing is unacceptable.

    • I am not buying the fixed-charge-is-crazy concept.

      I do not endorsing fixed charges necessarily, but there are plenty of examples of them in the non-electric world, including regulated markets like water and traditional phone service, and cable TV, to semi-regulated markets like mobile telephony, to totally unregulated markets like cable internet, Netflix, Amazon Prime, country clubs (well, all clubs, really)

      I’d like to see a bit more rigor in these assertions assertions that x-and-such activity is socially-inefficient grid bypass. Rooftop net-metering is well-studied, but others are less understood. For example, Are you certain that outdoor lighting applications with solar+battery+LED are more expensive (private- or social-perspective) than grid-connected? If you have ever cut, dug, and repaired a 20 foot long 4 foot deep ditch to connect to electric service, you might suspect otherwise.

      • Interesting that you raise the idea of digging a ditch: my very first encounter with utility rate design was digging a ditch to combine the shop service to the house service at a farm belonging to my college girfriend’s grandparents — because the Coop had adopted a $10/month “meter charge” and her grandpa did not want to pay double.

        Traditional phone service adopted high fixed charges in teh 1990’s — and has since lost 60% of their access lines. If I could still have a “home phone” for $6.35/month, I would, but at $40/month I have cut the cord. Cable TV is losing 2% of their subscribers a year, and many of the cable channels are moving to a la carte pricing.

        Amazon offers free shipping (free grid services) with purchases over $35, or unlimited grid services (AmazonPrime) for $100/year — and relatives in Alaska and friends in Hawaii are taking big advantage of that.

        Country clubs are a subject I know little about; and I don’t really consider relevant to a discussion of essentials like electricity.

    • “you NEVER see someone leaving Costco with an empty car”

      I’ll admit, I haven’t seen it, but that’s only because when I’m “buying” at Costco, I don’t care to pay attention to what other people are doing. There are lookie-loo teenagers / young adults who go to Costco using their parents’ memberships to go try all the free samples and maybe eat at the food court (access to which may not require a membership in some cases, if it is located outside the store entrance). How prevalent this subset of the Costco population is; I don’t know. But I know some of them exist: my friends post about doing this kind of “shopping” on social media!

      This is isn’t to discount the argument that the membership fee screens out casual shoppers, by and large. I just thought it was an interesting sociological observation.

  7. With the demise of the solar federal tax incentive for homes and small consumers while leaving that incentive with large corporations, some of which lease solar power to the consumer, is a disincentive for the small consumer to own his or her solar system. The same can be said for changes in pricing that are coming from the utilities as well. I suspect that this will decrease the rate of growth of the consumer solar industry. Then there are community regulations that prohibit the installation of solar panel or tax them through fees.

  8. In what ways does more efficient pricing reduce the ability of utilities to invest in new generation, T&D?

  9. For years Californian’s have witnessed the growth of the practice of taxation via utility bill. Perhaps the growth of customer alternatives will start a movement toward efficient utility pricing.

%d bloggers like this: