Solar, Location-Variable Costs, and the 21st-Century Grid

California is pushing solar energy more aggressively than any other state in the union, and the debates unfolding there may presage controversies likely to occur in Virginia. One of those debates is the relative value of utility-scale versus community versus individual rooftop solar, which is intimately tied to the issue of location-variable costs.

Building so-called utility-scale solar — essentially vast solar farms — allows economies of scale in installation and generates electricity at the lowest cost per kilowatt. Power companies like them because they typically own the utility-scale production and get to generate a profit off them. One drawback is that utility-scale production consumes vast swaths of land. Another is that solar farms rely upon high-capacity transmission lines to move their electricity to distant locations, while local generation can feed into local distribution grids.

Utility-scale solar is going gangbusters in Virginia, but opportunities for communal and rooftop generation are limited by state law. A huge sticking point here, as it is in California, is how much utilities should reimburse small-scale producers for the electricity they generate. To what degree should small-scale producers share in the cost of maintaining the larger distribution and transmission grid they rely upon when the sun isn’t shining?

Steven Sexton, an associate professor of public policy and economics at Duke University, is skeptical of the numbers that California officials are using to justify a mandate requiring all new houses built in the state to be equipped with solar panels. In a Wall Street Journal column today, he introduces an element into the debate that I haven’t heard discussed here in the Old Dominion: grid congestion. He writes:

Regulators should tailor policy to reflect routine variation in the value of solar generation across the state’s congested electricity grid. Solar panels are most effective when installed where transmission constraints make supply relatively scarce — not on every roof in California.

In other words, the value of rooftop/community solar generation varies depending upon its location on the electric grid. The value is greater where grid congestion is worse and the alternative is spending tens of millions of dollars upgrading the transmission system — often adding to visual blight in the process. The value of rooftop/community solar is less where grid congestion is not an issue.

I would add a corollary to that observation: Rooftop/community solar generation has greater value in remote, hard-to-serve areas where new development would require the installation of additional sub-stations and distribution lines. Back in the days when I wrote about land use, I advocated the principle that all property owners should pay the location-variable costs of their decisions about where to build. The biggest of those location-variable costs is transportation infrastructure, but a not-insignificant one is the supply of electricity. Why should city dwellers, who require less electric infrastructure per-capita, pay extra to subsidize rural dwellers? Conversely, why shouldn’t rural dwellers who generate some or all of their own electricity, receive some benefit when they avoid some of the cost of building rural electric infrastructure?

The root of the problem is that electric utilities charge a flat rate for all customers within the same class (residential, commercial and industrial) regardless of the variations in cost of serving those customers. I haven’t heard anyone in Virginia challenge that premise, but charging location-variable rates may be a necessary step for building an electric grid for the 21st century.

Share this article


(comments below)


(comments below)


17 responses to “Solar, Location-Variable Costs, and the 21st-Century Grid”

  1. “The root of the problem is that electric utilities charge a flat rate for all customers within the same class (residential, commercial and industrial) regardless of the variations in cost of serving those customers. I haven’t heard anyone in Virginia challenge that premise, but charging location-variable rates may be a necessary step for building an electric grid for the 21st century.”

    Then you weren’t paying close attention to the Haymarket case as that was exactly the argument made to pass the cost of the new transmission line on to Amazon as it was for all intents and purposes the sole beneficiary. The SCC Judges and particularly the Hearing Examiner did more than balk at applying the line extension policy to a transmission line. If you build your home in an area without the necessary infrastructure and requires a line extension, the homeowner pays for it. If you are Amazon, not so much.

    1. Good point. Next step: Apply the same logic elsewhere, not just in an instance that generated massive pushback.

      1. What’s the point so long as local governments keep approving greenfield development without consideration of the infrastructure necessary to support it, and lets not get into the stupidity of proffers. Dominion (and other utilities) consistently assert that they are required to provide power and that the cost of any new infrastructure is the responsibility of rate payer base, unless of course you are that homeowner on the top of a mountain, then it is your sole responsibility.

        Don’t look for the cowards at the SCC (commissioners and hearing examiners not staff) to change that dynamic anytime in the near future, or at least until Wagner, Kilgore, Norment and Saslaw have left the building with Elvis.

  2. This is what I was getting at when I suggested a Value of Solar tariff as opposed to net metering. When done correctly, the costs and benefits of installing distributed solar are calculated and netted out for various segments of a utility’s grid. This avoids any issues about other ratepayers subsidizing the wire charges relating to solar installations.

    In places where this has been applied, the value of distributed solar nearly always has a net positive benefit to the grid for added resiliency, reliability, voltage regulation (with modern inverters), and reduced congestion. Some areas benefit more than others and that is reflected in the bonus paid for solar in those areas.

    This should put to rest all of the arguments about net metering resulting in a subsidy for solar installations to rest, but that hasn’t been the case. This is partly because utilities are trying to hold the line on customer sited generation cutting into their revenues. The added fixed charges for customer-installed solar gave the utilities more revenue and discouraged some installations.

    Adding only utility-scale solar can actually increase transmission congestion and do nothing to relieve distribution congestion. That is why modern grids in deregulated states have a significant amount of new solar additions as distributed solar units.

    Con Ed is avoiding an investment in a $1 billion substation because of local solar, storage, energy efficiency and demand response. But that only works in deregulated states. Virginia’s utilities want to put as much in the rate base as they can, even if it costs customers more.

    The congestion relieved by distributed solar is directly at the distribution level. But by increasing the amount of local generation, it reduces the amount of electricity that must be imported from outside the area, thus also indirectly reducing transmission congestion.

    In a sense, rural users do pay a higher cost for living in a less dense, more expensive to serve area. The big utilities cherry picked the urban centers and left the rest to the co-ops. That is one reason the co-op structure was selected to reduce the amount charged as profit, making rural electricity more affordable than it would be if it was served by an investor-owned utility.

    For solar to fully realize its potential in Virginia, we need to remove the incentive for utilities to own and control most of the solar development. I have spoken with a solar developer and we are trying to find a way to create a hybrid arrangement where the customers will be able to access lower cost solar than what the utility would provide (because of it being in the ratebase), but giving the utilities an extra revenue stream by being a financier or part-owner of solar facilities.

    The big obstacle is our continued reliance on 100-year old regulatory structures to create a modern grid. Whatever we do has to be good for the customers and the utilities. Our current energy policies are not.

  3. LarrytheG Avatar

    I dunno… you’re gonna take this approach with cable TV or phone service…libraries, water/sewer , fire/rescue? You’re gonna bill each new house or subdivision according to location-specific costs?

    And don’t get started on proffers and schools.. Since they changed the law -localities apparently cannot take proffers for schools… but they can turn down a proposal if the local school is at capacity..

  4. LarrytheG Avatar

    First time I’ve heard that location and “congestion” affects electricity. I thought the more informed here has said that where you put more power into the grid was not that dependent on location – and that – that’s how PJM “works”. Someone buys power at location X and someone provides it at Location Y within the PJM region.

    If “congestion” is a problem – that appears to give Dominion and ready-made way to bar 3rd party utility-scale solar, no?

    Unless I misunderstand, California law is with respect to NEW homes – which pretty much correlates with increased ‘congestion” – i.e. hooking up new subdivisions to existing wires…

    If the utility can be a constructor of solar – it would seem to be a good thing that allows them to expand their business model .

    And if they buy power from individual/community solar – it seems reasonable that they charge back a grid “availability” fee.

  5. PJM’s locational marginal pricing takes the transmission issues account.

    Regarding California – if new houses come equipped with solar, they provide part of the “congestion” solution to the problem that more new houses creates.

    It could create distribution congestion at night if the distribution connection to the new subdivision was not adequate to serve all of the new homes when the sun was not shining. It could also be a bottleneck (congestion) when the new homes produced much more solar than they consumed during maximum sunlight periods. The distribution system would have to upgraded to make use of the surplus solar. Once the upgrade was complete the local grid would be improved by the addition of more local generation.

    This complicates things for utilities because it requires more attention than plopping in a single utility-scale unit every six months or so. But the grid will be improved by having a higher amount of local generation.

    It would not be appropriate to charge a grid availability fee if that was already factored into the Value of Solar calculation.

    Having the utility build solar and putting it in the rate base makes it more expensive.

  6. LarrytheG Avatar

    re: ” Having the utility build solar and putting it in the rate base makes it more expensive.”

    Can a utility sell products to customers and the transaction is not in the rate base?

    Like solar, or backup generators, or smart thermostats, or Tesla Walls, etc?

    similar to how the cable or phone company can sell “add ons” to the basic services….

  7. TooManyTaxes Avatar

    Competition at the generation level will work if and only if consumers are given a choice of a supplier that charges less than Dominion. I don’t give a rat’s xxx about the source of my electricity, only the price. It’s a commodity and commodities trade on price.

    I’m not sure how to handle Dominion on the generation side. If there is enough capacity from other sources, I would think that the PJM process would control price. If there’s not, Dominion might be able to be a successful rent seeker.

    I fear Dominion being able to amortize excess investments (in non-transmission and distribution plant) and charge customers. Perhaps, one could make an argument that Dominion should not be permitted to amortize all of its investments because it could have obtained cheaper power elsewhere, including by financing demand-reducing measures for customers. Ergo, all of its investments were not used and useful but excessive.

    Decades ago, the FCC and many PUCs proposed downward adjustments to the Bell System’s rate base(s) because it could have purchased equipment from suppliers other than its affiliated manufacturer Western Electric. In response, the Bell System conducted life cycle cost studies that, over the life of the equipment at issue, demonstrated lower overall costs. My first exposure to rate cases. My witness was the cost study guy for WE purchase.

    I would think a similar effort could be made with Dominion’s investments especially as they are contrasted to other power companies nationwide. If some of the things I’ve read here are true, life cycle costs should be able to trip up Dominion.

  8. LarrytheG Avatar

    One question that I have is this – is the monopoly model for electric utilities obsolete or in it’s current form out of phase with how technology and regional entities like PJM have evolved?

    Tom and others have commented that until we find a way for Dominion to benefit from advancing technology – they will continue to preserve and protect regulatory rules and law that benefit them while harming competitors and customers.

    Is the monopoly type that Dominion currently operates – one that is sustainable? Do other states -other regions or countries “do” electricity differently such that it is “better” and by that I mean cheaper and less polluting… and gives customers more freedom and options in choosing how to meet their own needs?

    for myself – I DO CARE how electricity is generated if it is polluting… IF electricity can be generated with LESS pollution. I’d think the very same way about just about anything else… it’s a balance.. but we ought not be polluting just for convenience or low price… especially if in doing that – it actually does cause harm to people and critters..

    1. TooManyTaxes Avatar

      Electricity is a commodity and competition must be about price. Look at long distance competition back in the 1980s. MCI and Sprint competed against AT&T on the basis of price. They charged a lower per-minute rate than did AT&T, which, in turn, provided customers, both residential and business, an economic incentive to use the new entrants’ services and, when equal access became available, presubscribe their 1+ dialing to these other carriers. Why is telecom different from electricity?

      If green energy has become less expensive to generate than fossil fuel energy, it should be available at a lower price than the incumbent power companies charge for generation. But all I see are offers to pay more than I pay Dominion for a kwh. Why?

      Moreover, if green power companies can supply a significant amount of power at lower prices, not only will customers move from Dominion to the new entrants but it will also put downward pressure on Dominion’s rates. What is significant? I don’t know.

      What the public needs to worry about is the likely attempt by Dominion to shift costs from generation of power to the monopoly services of transmission and distribution.

      1. TMT you are correct, cost shifting is a risk, and this is one big reason we need the oversight of the FERC and the SCC to keep these costs in the right cubbyholes. It becomes especially difficult when formerly vertically-integrated utilities like Dominion use a complicated set of cubbyholes of their own, using separate corporate subsidiaries, to perform functions that do NOT align precisely with the regulatory jurisdictions involved, or that do NOT align with the separation of functions on the grid between generation transmission, distribution and retail sales (by continuing to “rate-base” dedicated generating units at ratepayer expense rather than sell to and buy from the wholesale regional energy markets.)

  9. Jane Twitmyer Avatar
    Jane Twitmyer

    Reading about the change that will come to how are electricity is generated and reaches us … the future is not really clear, but as Tom has told us, there is a remaining monopoly part and that is the grid. It was too expensive to have competing wires, which is actually what decided the monopoly regulation model in the first place. Since technology does not have to be centrally generated anymore, the grid can become a platform.

    In designing NY’s new model REV, “regulators aim to transform the traditional utilities into platform providers — entities that facilitate the deployment of distributed energy resources (DERs) and use them instead of traditional infrastructure to serve system needs. … Platform companies try to find ‘common interest’ with companies that use their platforms,” … Rather than thinking of DERs as a threat, [utilities should] think about them as your services and your partners.” So, the point is to base earning in a different way using the monopoly platform … differing rates can apply.

    This is all part of the argument about rural solar …which VA has a large opportunity based on space available vs. on-site and community solar. NREL in the GIS report said Virginia could fulfill 25% of their total demand and that was only looking at roofs available, not at the idea of community solar where the panels are sold to individual buyers and the panels do not have to be onsite but can be down the road. I would love to see that happen on Wintergreen where the trees block lots of roofs.

    Some more thoughts from Utility Dive articles … I have been quoting

    “New York’s Reforming the Energy Vision captured the nation’s imagination three years ago. An ambitious undertaking, regulators sought to remake the utility business model to incentivize deployment of DERs and demand management as an alternative to traditional infrastructure. To make this goal palatable to utilities—used to collecting a hefty rate of return on massive infrastructure projects—regulators are devising a framework allowing them to earn a rate of return tied to investing in DERs and achieving societal and energy goals.

    Some states, like NY and CA are tackling the whole structure, while others are going with pilots. I can’t see VA taking a big step but where are the pilots … for storage, for microgrids? MA’s incremental plans call for … ”Roughly $145 million of the modernization plan will go to storage and EV charging, and $254 million will be spent on distribution automation.”

    So the question is who will direct Dominion to release their grip on central generation and the “sell more, build more” regs?

  10. I do not agree with Mom and others that locational pricing for retail customers makes much sense on today’s grid. The costs of transmission congestion and losses are simply too subtle to matter much in the mid-Atlantic region, except to bulk power participants like utility-scale independent generators and large industrial and institutional customers. [Big caveat: those costs matter hugely if the transmission to a location is inadequate — like on the Peninsula — but the utilities in the PJM region have done a good job of building the transmission needed to avoid such “load pockets”.]

    Even “value of solar” pricing for distributed solar can be based on grid-wide, or at least utility-wide, avoided costs. That said, the real value added from distributed (“rooftop”) solar is grid resiliance and cost savings at the local distribution level of the grid, not at the bulk power, transmission level. And to get that savings, first, the local distribution utility (Dominion, not PJM) has to ensure that its distribution substations are built to handle two-way power flow, which means there’s an up front threshold cost before the savings. I’m all for eliminating barriers to distributed solar — but pricing distributed solar variably by grid location is just too complicated, not to mention capricious. First, the locational variations aren’t that great, and second, they vary with time of day and other local generation operating and even weather conditions, and third, locational differences would have to be updated near-constantly every time the local distribution grid was upgraded. Just avoid all this by pricing d.s. based on a utility-wide, averaged benefit.

    1. I am not familiar with the type of regional variations that might exist in Virginia. If they are not that great, I agree – do a utility wide net benefit. The simpler the better.

      The greater and more expensive congestion issues probably occur at the transmission level anyway (the Peninsula).

      All of this assumes that we find a way to remove the caps on distributed solar (net metering caps). And find a way to pay the utilities differently so that they realize the benefits from distributed energy and are not hurt by the revenue loss.

      1. TomH, your last paragraph is aspirational — “find a way.” That way, however, must be compatible with the existing regulatory regime. That regime is ponderous at best, overly accounting driven, opaque to those not steeped in it, with results that are even obtuse at times. But, cost of service, tariff-based utility regulation works fairly well at balancing shareholder and “the public” interests, all things considered, and it survives because we have no better alternatives.

        Consumers want regulation that reflects local conditions in real-time, but they just can’t have it. Real-time, local conditions change way too fast for tariffs that take months to process. We have to accept the use of averages over time and geography for the regulatory system to work. Either that, or incorporate formulas in the tariff that reflect in advance typical seasonal and time-of-day and customer-classification-based differences. Or flow-through someone else’s determination of real-time costs over a large region, such as a fuel adjustment clause, or retail industrial rates keyed to the PJM wholesale energy market price as it changes minute to minute, that the regulator approves in advance subject to special followup scrutiny of the results. I’m not against value-of-solar pricing in concept but to sell it to the regulator it has to be workable. And from what I’ve read, a Dominion-wide calculation, updated periodically in the course of base rate proceedings only every few years, would be workable.

        Getting any finer-grained than that imperils the concept. For example, it would be foolish to try to sell the regulator on a calculation that assigns each neighborhood a different value of solar according to its different grid infrastructure, during each different seasonal and daily time frame. A patchwork of different value-of-solar rates — even if cost justified at a single, historical moment in time — would create capricious winners and losers going forward and would drive the regulator crazy. I know you are looking for something workable and that patchwork approach, in my opinion, would not be.

      2. And as for those occasional “greater and more expensive congestion issues” — I think you will find that the Dominion and PJM planning processes have made the Peninsula situation the exception that proves the general rule: such differences at the transmission level are few, particularly within a well-run traditionally-integrated utility’s service territory. Indeed the Peninsula situation has been a cause of concern for years, and results both from the generation shutdowns there and extraordinary delays in the completion of transmission upgrades to fix the load pocket condition resulting from those. That is certainly not typical of the rest of DOM’s grid.

Leave a Reply