Roger Caiazza
I have spent this year documenting how New York regulators sell the Regional Greenhouse Gas Initiative (RGGI) as a successful cap-and-invest program while allowance prices climb. Governor Abigail Spanberger made affordability the cornerstone word of her campaign, and her energy plan promised to “negotiate the best ratepayer deal to rejoin RGGI.” For a Governor who ran on affordability, the entire Virginia RGGI re-entry process has been shocking.
Virginia’s first auction back in the program cleared at $37.65 per allowance, raising $259 million for the Commonwealth. The last auction before Virginia left cleared at $14.88. Dominion Energy has asked the State Corporation Commission (SCC) to reinstate its RGGI rider at $13.00 per month for a typical residential customer, or $10.36 under its mitigation proposal. It was $4.43 per month when removed in June 2024.
This post draws on the 2026 Virginia Energy Plan (VEP) and a presentation by William Shobe of the University of Virginia and member of the Virginia Air Pollution Control Board, “Explaining the RGGI price path.” Shobe helped design RGGI and is no program critic.
Rushing in six months early
The General Assembly did not have to force a mid-year return. The budget language itself required DEQ to align Virginia’s regulation with the new Model Rule “no later than January 1, 2027.” Instead, the February “caboose” budget, HB 29, directed agencies to repeal the 2023 exit regulation and reinstate the old one within 90 days. Virginia resumed participation on July 1, 2026.
The market noticed. Argus Media reported that the RGGI price rally was “likely kickstarted, at least in part, by news that Virginia would rejoin RGGI as soon as 1 July, six months earlier than what most market participants had anticipated.” The early start also meant that Dominion had to buy allowances in the September auction before its cost-recovery mechanism had been reinstated. Customers will pay for those eight months plus financing, which Appalachian Voices estimated could reach $17 million.
Cutting the 2026 budget in half instead of adjusting for the season
Virginia’s 2026 base budget was simply cut in half, to 11.48 million tons of the 22.96 million tons a full year would have received. The July–December period includes the summer peak, when fossil units run hardest. Shobe’s presentation notes that Virginia’s covered sources emitted about 17 million tons in the second half of 2025, roughly 50% more than the six-month budget they were given (Figure 1, Panel A). A seasonally adjusted allocation would have reduced the impact; without one, Virginia came back shorter on day one.
Leakage
Virginia is in the PJM RTO. When Virginia units add RGGI allowance costs to their bids, they become less competitive than units in states like West Virginia, which shifts emissions out of state, or leakage. Table 1 compares statewide emissions for facilities that report gross load and had data available for the third quarter of 2026. Three facilities lacking data and 12 that report steam load were excluded. If the RGGI cost differential caused all the third-quarter changes between 2025 and 2026, the upper-bound leakage would be 1.9 million tons of CO2, 2,683 tons of SO2 and 1,433 tons of NOx. This is a rough estimate because I do not have all the necessary data.
Table 1: Comparison of WV and VA facilities in the third quarter of 2025 and 2026
No stakeholder engagement
Every step bypassed Virginia’s Administrative Process Act (APA). The HB 29 language says that any regulatory action needed to rejoin is “hereby exempted from the provisions of the Administrative Process Act” and is to be completed “without further action by the State Air Pollution Control Board.”
Revision D26, which sets Virginia’s budgets through 2037, worked the same way. DEQ said the action “will not have an associated public comment period.” The DEQ Director adopted it on September 24, 2026, again without Board action. A decade of compliance costs got no public comment, no APA economic analysis, and no citizen-board vote.
“Aligned” to modeling that never included Virginia
The legislation directs DEQ to “align” Virginia’s regulation with the Third Program Review. That review, however, was carried out by the ten states excluding Virginia then participating, and the electricity-sector modeling behind the cap was a ten-state exercise. Even the Virginia Energy Plan labels its cap table “Summary of RGGI Regional Allowance Cap Updates, not including Virginia.”
DEQ and RGGI’s consultant ICF did not model Virginia. DEQ used a formula, taking Virginia’s 2026 budget of 22.96 million tons, which dates to the 2020 regulation, and applied the RGGI trajectory reductions of 2.55 million tons a year from 2027 to 2033 and 0.73 million tons a year from 2033 to 2037. That leaves 2.19 million tons in 2037. No analysis tested whether Virginia’s fleet can meet that path, or at what cost.
The world has also changed since that modeling. Dominion’s ten-year growth forecast has gone from 0.4% to 2.2% to 5.0% a year. On the supply side, the VEP admits that “Virginia’s investor-owned utilities resource additions are not on track to meet the VCEA’s net zero goals.” Higher demand and slower renewable buildout both push allowance demand up.
The final budget is also tighter than the one the Energy Plan used. The VEP modeled 2.64 million base allowances in 2037, which it said was 99.87% of DEQ’s preliminary revised trajectory, “suggesting a high amount of certainty.” The final D26 rule sets the 2037 base budget at 2,186,667 tons, about 17% lower.
The Energy Plan still depends on natural gas
The cap conflicts with the state’s own plan. In the VEP’s Current Policies, moderate load case, Virginia adds about 6.7 GW of new natural gas capacity between 2029 and 2034 (Figure 56), and the plan’s projections keep roughly 13 to 15 GW of natural gas capacity serving the Commonwealth through 2045. According to the VEP, Dominion’s 2025 Integrated Resource Plan calls for roughly 8 GW of new gas, with no fossil retirements. Meanwhile, Virginia’s base budget drops from 20.4 million tons in 2027 to 5.1 million tons in 2033. A gas fleet that grows while the cap falls 75% in six years has only one outcome: Virginia buys allowances from everyone else at whatever price the market sets.
No in-state projection of regulated-source emissions
Neither Virginia nor the VEP projected what RGGI-covered sources will emit compared with the budget. The VEP’s capacity expansion model forced compliance by assuming “Virginia electric utilities will purchase and retire at least all the base and CCR allowances prorated for Virginia.” Yet the plan concedes that “Virginia entities could purchase allowances that are allocated to other states.” How many allowances Virginia sources will need beyond Virginia’s allocation, the question that decides ratepayer costs, was never studied.
Net buyers cannot be sheltered by their own auction revenue
The Governor’s affordability argument depends on the bill credit. The budget sends 45% of RGGI revenue back as bill credits to residential, small business, and church customers. The VEP says these classes “were assigned 41% of the costs related to RGGI compliance,” so they “should expect a net credit.” The Governor said “average monthly bills could drop by as much as $36 annually.”
The arithmetic does not work. Shobe’s slide is titled plainly: “Virginia is likely a net buyer.” Dominion alone expects to buy 51 million allowances from July 2026 through December 2028 at a projected cost of roughly $1.2 billion. Virginia’s base budget for that same period adds up to only 49.75 million tons (Figure 1, Panel B). More importantly, RGGI costs land on the retail side. Imported and clean electricity serving Virginia load is still sold at a marginal price set by a RGGI fossil generator, so the entire load bears RGGI costs while the State collects only from in-state fossil generation, which is just a share of the total.
Large commercial, industrial customers, and ratepayers from municipal utilities who buy power from independent power producers like Potomac Energy Center are not among the classes receiving the credit, and their higher costs show up in everything Virginians buy. The Virginia Mercury notes the credit means less money for low-income energy efficiency and flood preparedness, which previously received 50% and 45% of proceeds.
Scarcity, not speculation
RGGI supporters often blame speculators for high prices. Shobe’s slide is titled “Scarcity, not speculation.” Investor-held surplus fell from 60 million to 37 million allowances between the second quarters of 2025 and 2026, and 71% of allowances in circulation are now held by compliance-oriented entities. “A speculative bubble would show investors accumulating. The data show the opposite.”
The Cost Containment Reserve “sold out in the first auction of every year since 2024, so the CCR no longer caps the price,” and futures above the $29.25 Tier 2 trigger mean “the market expects both tiers to be exhausted in 2027.”
Scarcity prices are structural, not temporary. The cap falls faster after 2027, data center load keeps growing, and the Energy Plan builds more gas.
Is Dominion collecting RGGI twice?
The SCC needs to examine this before February 5, 2027. RGGI costs reach customers through the rider, which recovers Dominion’s allowance purchases, and through the wholesale market. As Shobe explained to VPM, in PJM “any generator that has carbon-intensive generation has to add that RGGI price” to its offer. When an emitting unit sets the locational marginal price (LMP), every megawatt-hour clearing in that interval, including nuclear, solar, and imports, gets paid the higher price.
The SCC should ask:
- Do the extra PJM revenues Dominion’s fossil units earn at RGGI-inflated LMPs fully offset Rider RGGI, or does the rider recover costs the market already paid?
- How much RGGI cost reaches Dominion customers through merchant Virginia units and the roughly 30% of Virginia’s electricity that is imported?
- How much is a windfall for non-emitting generators that never buy an allowance?
In my New York work, the wholesale-market effect roughly doubles the consumer cost of direct allowance purchases. Virginia’s rider debate leaves that half out.
What an affordability Governor should have done
A Governor who meant what she said about affordability would have waited until January 1, 2027, when the new Model Rule takes effect. She would have followed the Administrative Process Act with a real economic analysis, published a projection of Virginia’s covered-source emissions against the budget before accepting a trajectory built for other states, and required the SCC to account for the wholesale-price effect. None of that happened. Virginia rushed in at the worst possible time, accepted a cap built for other states, skipped every public check, shifted generation to dirtier West Virginia units, and is presenting a partial refund of residents’ own money as a savings program.
The Governor promised to negotiate “the best ratepayer deal.” As far as I can tell, there was no negotiation and there is no deal. Virginians are left with a higher bill and a credit that can’t cover it.
——————————————————————————————————————
Roger Caiazza blogs on New York energy and environmental issues at Pragmatic Environmentalist of New York. Dealing with the RGGI regulatory and political landscapes is challenging enough that affected entities seldom see value in speaking out about fundamental issues associated with the program. He has been involved in the RGGI program process since its inception and has no such restrictions when writing about the details of the RGGI program. This post is based on an outline from an anonymous colleague from the electric utility industry. This represents his opinion and not the opinion of any of his previous employers or any other company with which he has been associated. He acknowledges the use of Perplexity AI to help research and draft this post.
