Duke Energy Progress and North Carolina Public Staff Sign Settlement Cutting Requested Rate Increase by More Than Half
The August 5 agreement follows a similar July deal covering Duke Energy Carolinas customers; the North Carolina Utilities Commission has not yet ruled, and the state attorney general has declined to sign either settlement.
A price cut that's still a price hike
Duke Energy Progress and North Carolina's Public Staff signed a settlement on August 5, 2026, that trims a requested electricity rate increase by more than half[1]. That is true. It is also true that under the parallel deal struck in July for its sister utility, Duke Energy Carolinas, a typical household will still pay 5.9% more starting January 1, 2027, and 3.6% more on top of that the year after — 9.5% in total[3]. Both numbers are accurate. Neither one is the whole story.
Duke serves the eastern half of North Carolina through Duke Energy Progress and the western half through Duke Energy Carolinas[2]. The Progress settlement, signed by the Public Staff along with the Carolina Industrial Group for Fair Utility Rates, the Carolina Utility Customers Association, the North Carolina Sustainable Energy Association and Walmart, mirrors the Carolinas deal from a month earlier[1]. The state utilities commission hasn't ruled on either one. It can approve, shrink or reject them, and new rates would take effect at the start of 2027 if approved[3][4].
One person notably didn't sign either settlement: North Carolina Attorney General Jeff Jackson. He said the deal moved in the right direction but is "still too high"[4]. His objection, and the fight over who actually pays for North Carolina's data-center boom, is where this story gets interesting.
The number both sides use, and why it means different things to each
Duke originally asked for something close to an 18% residential rate increase in the Carolinas case, later lowered its own ask to about 11.6%, and settled at 9.5%[3][4]. Seen one way, that's a company cutting its request nearly in half after negotiation — which is exactly how Duke's press release frames it, under the headline "a lower-cost path to power North Carolina's future"[1]. Seen the other way, it's still a bill going up almost 10% over two years, which is how Attorney General Jackson frames it[4].
Both framings use the same figure. The difference is the baseline each side picks: Duke measures against its own opening ask, Jackson measures against what people pay today.
Buried in the same settlement is a second number that took more work to negotiate: the return Duke is allowed to earn on its investment, known as return on equity, set at 9.8%[3][6]. Duke had asked to raise that figure from its previous authorized level of 10.1% up to 10.95%; the settlement brought it down to 9.8% instead[3][6].
That number needs unpacking, because it's the real argument underneath the headline percentage. Duke is a regulated monopoly — it has no competitors, and it can't just set its own prices[3][6]. Instead, regulators decide how much profit Duke is allowed to earn on the money it spends building power plants, substations and transmission lines. That allowed profit rate exists for a reason: it's what Duke has to offer investors to get them to lend it money for a $103 billion construction plan[6][7]. Set the number too low, Duke's argument goes, and its borrowing gets more expensive — a cost that eventually lands on customers anyway, through higher financing charges baked into future rates. Set it too high, critics argue, and shareholders are pocketing money that should have stayed in customers' pockets. Every tenth of a percentage point is real money spread across a multibillion-dollar base of assets, which is why Jackson fought over 9.8% instead of just the headline rate increase[4].
What's driving the building boom nobody can quite agree on
Here's the tension that doesn't resolve: North Carolina's electricity demand sat roughly flat for about twenty years, and now it's rising fast, largely because of data centers[12]. Duke has signed 7.8 gigawatts worth of service agreements with data-center customers, including Microsoft, Amazon, Digital Realty and QTS, and has told investors that demand could add $5 billion to $10 billion more to its already-huge capital plan[6][7].
Utilities Commissioner Tommy Tucker raised the sharpest version of the question in hearings: roughly 70% of Duke's queue of large new customers is data centers, he said, and asked how everyone else is supposed to help pay for close to $100 billion in investment by 2035[5]. Duke disputes that data centers are a major driver of this particular rate request, and says safeguards already exist to keep households from subsidizing infrastructure built for big new customers[5].
That disagreement is why the settlement includes something easy to miss in the headline numbers: a separate, faster proceeding to build a "large load tariff" — a special rate class that would make very large customers like data centers pay directly for the specific grid capacity built to serve them, instead of spreading that cost across everyone's bill[3][10]. The North Carolina Sustainable Energy Association signed onto the broader settlement partly because that separate process now exists[10]. Jackson wants to go further — he argues Duke Energy Progress should let its biggest customers build their own power generation, so Duke isn't building it and billing everyone else[4][5].
A company that just posted record profit is also asking for more money
Duke reported adjusted second-quarter 2026 profit of $1.11 billion, or $1.43 per share, up 14% from a year earlier, on revenue of $7.59 billion[6][7]. That earnings call happened on August 4 — one day before the Progress settlement was announced[6].
Critics see a profitable company asking households for more money at the same time. Supporters see a capital-intensive company raising the money a fast-growing grid actually requires, where quarterly profit and rate requests are two separate mechanisms that both flow from the same regulatory formula. Both groups are looking at the same numbers. Neither is wrong about what the numbers say — they disagree about what the numbers mean, and that disagreement isn't going to be settled by finding one more fact.
How the same deal became a different headline everywhere
Coverage of this settlement split less over the facts than over which fact came first. Duke's own release led with "lower-cost path," measuring the deal against its own original ask and not mentioning that the attorney general refused to sign[1]. Carolina Journal, a North Carolina outlet aligned with free-market advocacy, conceded the cut in its headline but pivoted immediately to "NC households still face 9.5% increase" — notably, without leading on Duke's profit figures, since its critique is about cost of living rather than corporate earnings[8].
NC Newsline, a left-leaning nonprofit outlet, led instead with Jackson's refusal to sign, running the headline "NC AG Jackson rejects Duke Energy's proposed rate increase" — though Jackson has no power to reject a settlement, only to withhold his own signature from it[4]. Canary Media, a clean-energy nonprofit outlet, opened by naming data centers as the "key" driver of the fight, a conclusion Duke itself disputes[5]. Trade outlet Utility Dive surfaced Commissioner Tucker's sharpest question about the 70% data-center queue, but framed it as a regulatory risk to Duke rather than a cost question for households[5]. Axios offered the plainest version — "Duke Energy reaches settlement to lower its proposed rate increase" — though even that can read, to a fast skimmer, like an actual rate cut rather than a smaller increase[8].
What the settlement doesn't decide
International coverage barely mentions North Carolina specifically. Instead, wire reporting folds this into a bigger story: AI data centers pushing electricity costs onto ordinary households across the U.S., with Reuters reporting that households in PJM's grid territory could see rate hikes up to 60% over five years, and PJM attributing $6.3 billion of added consumer cost largely to data-center demand[11]. A Consumer Reports survey from November 2025 found 78% of respondents were concerned data centers would raise their own energy bills[11].
The North Carolina Utilities Commission still hasn't ruled on either settlement, and it's free to approve, modify or reject what's in front of it[4]. Even if it approves both deals as written, the bigger question — how much of the state's data-center building boom gets billed to data centers themselves, versus spread across every household's monthly bill — won't be settled here. That fight moves next into the separate large-load tariff proceeding the settlement created, where the dollar amounts at stake are larger than anything decided so far[10].
Summary
On August 5, 2026, Duke Energy Progress announced a settlement with the North Carolina Public Staff and several other parties that would cut its requested electricity rate increase by more than half[1]. Duke Energy Progress serves the eastern half of North Carolina. The deal follows a similar agreement reached in July for Duke Energy Carolinas, which covers the western half[2][3]. Under the Carolinas deal, typical residential rates would rise 5.9% in the first year and 3.6% in the second — 9.5% in total — instead of the roughly 18% Duke first sought[3][4]. Neither settlement is final. The North Carolina Utilities Commission must approve, change, or reject them, with new rates set to start January 1, 2027 if approved[3].
The Public Staff is a state agency created by law to represent utility customers in rate cases[3]. Its decision to sign is the settlements' main selling point. But the state's other consumer advocate did not sign. Attorney General Jeff Jackson, a Democrat, refused to join, saying the increase 'is still too high'[4]. He also argued Duke's allowed profit rate should be lower than the settled 9.8%, and that Duke Energy Progress should let its largest customers, including data centers, generate their own power[4][5].
The fight underneath the numbers is about data centers. North Carolina is absorbing a wave of them. Duke told investors it has signed 7.8 gigawatts of electric service agreements with data-center customers including Microsoft, Amazon, Digital Realty and QTS[6]. Utilities Commissioner Tommy Tucker noted in hearings that roughly 70% of Duke's queue of large new customers is data centers, and asked how everyone else is supposed to pay for roughly $100 billion in investment by 2035[5]. Duke says data centers are not a major driver of this particular rate request and that safeguards already exist to stop households from subsidizing them[5].
The timing is what makes the story contested. Duke reported adjusted second-quarter profit of $1.11 billion, or $1.43 a share, up 14% from a year earlier[6][7]. Critics read a profitable company asking households for more. Duke and its supporters read a company raising the enormous capital a growing grid requires. Both readings use the same numbers.
The Event
On August 5, 2026, Duke Energy announced that its Duke Energy Progress subsidiary had reached an agreement with the North Carolina Public Staff and other parties that reduces its proposed rate increase by more than half[1]. Signing parties include the Carolina Industrial Group for Fair Utility Rates, the Carolina Utility Customers Association, the North Carolina Sustainable Energy Association and Walmart[1]. The announcement follows a comparable settlement reached in July in the Duke Energy Carolinas rate case, which set a 9.8% return on equity and a 53% equity capital structure[3][6]. The North Carolina Utilities Commission has scheduled hearings and has not issued a decision; if approved, new rates take effect January 1, 2027[3][4].
Undisputed Facts
- Duke Energy Progress announced a settlement with the North Carolina Public Staff and other parties on August 5, 2026, reducing its requested rate increase by more than half[1].
- Duke Energy Carolinas reached a similar settlement in July 2026 that would raise typical residential rates 5.9% in year one and 3.6% in year two, for a cumulative 9.5%[3].
- Duke Energy Carolinas originally requested a residential increase of about 18% and later lowered the request to about 11.6% before settling[4].
- The Duke Energy Carolinas settlement sets an authorized return on equity of 9.8%, down from the 10.95% Duke requested in this case (its previously authorized return was 10.1%), with a capital structure of 53% equity and 47% debt[3][6].
- North Carolina Attorney General Jeff Jackson declined to sign the settlement, stating the increase 'is still too high'[4].
- The North Carolina Utilities Commission has not ruled; it may approve, reduce or reject the request, and new rates would begin January 1, 2027 if approved[3][4].
- Duke Energy reported adjusted second-quarter 2026 profit of $1.11 billion, or $1.43 per adjusted share, a 14% increase from a year earlier, on revenue of $7.59 billion[6][7].
- Duke Energy has signed electric service agreements totaling 7.8 gigawatts with data-center customers, including Microsoft, Amazon, Digital Realty and QTS[6].
- The settlement provides for a separate, expedited proceeding to create a 'large load tariff' governing how very large customers such as data centers pay for grid costs[3][10].
The Pressure
Strip away the moralizing and blame. What structural realities persist regardless of which narrative wins?
- Regulated monopoly economics
- Duke cannot set its own prices and has no competitors. Its earnings come almost entirely from a regulator-approved return on the capital it invests in the grid. That structure creates a permanent, built-in incentive to build: more approved investment means more earnings. It is not a scandal, it is the design of the system — and it is why an independent regulator setting the return on equity exists at all[3][6].
- Load growth after two flat decades
- U.S. electricity demand was roughly flat for about twenty years. It is now rising fast, and AI data centers are a large part of why. Utilities across the country requested more than $29 billion in rate increases in the first half of 2025, double the first half of 2024[12]. North Carolina is not an outlier; it is a case study.
- The settlement machine
- Most large rate cases end in settlement, not litigation. Utilities open high, staff negotiates down, and both sides present the gap as a win. This makes the opening ask a strategic number rather than a neutral one — which is why 'cut by more than half' is a real concession and a favorable baseline at the same time.
- Cost allocation is the actual fight
- The total dollars Duke will spend are largely determined by demand and by the commission's resource plans. The open question is who pays which share. That is why the large load tariff proceeding, not the headline percentage, is where the durable money is decided[10].
Material realityNorth Carolina's electricity demand is rising and Duke is building to meet it. Duke has signed 7.8 gigawatts of service agreements with data-center customers and told investors data-center demand could add $5 billion to $10 billion to a capital plan already around $103 billion[6][7]. That construction gets paid for by ratepayers, by large customers, or by shareholders accepting a lower return — there is no fourth option. Duke's adjusted second-quarter profit was $1.11 billion, up 14%[6][7]. Under the settled terms a typical residential customer pays 5.9% more starting January 1, 2027 and 3.6% more the year after[3]. The North Carolina Utilities Commission has not ruled, and it can approve, modify or reject the deal[4]. Nationally, a 2026 modeling study from NC State, Carnegie Mellon and others projected demand-weighted wholesale power prices could rise 6% to 29% on average by 2030 from data-center and cryptocurrency demand, and as much as 57% in the hardest-hit regions[12]. Those are projections, not measurements.
Narrative as a weaponThree parties are actively shaping how this reads. Duke wants you to measure the outcome against its own opening request, so that a 9.5% increase registers as a savings event — hence a press release headlined 'lower-cost path.' The Public Staff wants credit for cutting the ask roughly in half and needs the settlement to look like a hard-won bargain, not a handshake. Attorney General Jackson wants the residual increase to be the story and the settlement to look premature, which keeps him positioned as the consumer's last advocate. A fourth interest is quieter: data-center operators are largely absent from the public argument while the large load tariff proceeding — the thing that will actually set their share of the grid bill — is negotiated. Note what almost no one in this coverage does: explain how a regulator-set return on equity works. Without that mechanism, 'Duke made $1.1 billion and wants more' is intuitive and complete, while Duke's capital-cost argument sounds like an excuse. That gap is doing more framing work than any single adjective in any headline.
How Each Side Sees It
Each major actor’s view — how it frames things, its underlying incentive, and how it’s materially affected. Tap a side to read it.
Frames it asDuke's core argument is that it is not choosing to spend money — demand is arriving whether it builds or not. North Carolina's population and industry are growing, and Duke says it must add generation, transmission and substations to keep the lights on. It argues the settlement proves good-faith negotiation: it cut its own ask by more than half and accepted a lower profit rate than it requested[1][3]. On the profit question, Duke's best case rests on how regulated monopolies are financed. A utility cannot set its own prices, so regulators set an allowed return on equity — the profit rate on shareholder money invested in the grid. That number is not a windfall; it is the price Duke must offer to attract the investors who fund construction. Set it too low and investors go elsewhere, Duke's credit rating slips, its borrowing costs rise, and customers pay for that more expensive debt in later bills. Duke also says data centers are not a major driver of this rate request, and that safeguards already exist so residential customers do not subsidize infrastructure built for large new users[5].
WhyDuke needs a return on equity high enough to raise roughly $103 billion in planned capital spending, and it has told investors data-center demand could add $5 billion to $10 billion more[6][7]. A regulator-approved settlement is far more predictable than a contested ruling, and predictability is itself worth money to a capital-intensive company.
Impact on themDuke reported adjusted second-quarter profit of $1.11 billion, up 14% year over year[6][7]. The settled 9.8% return on equity is below the 10.1% it asked for, so the deal costs it real revenue[3][6]. In exchange it gets a faster, more certain path to recovering what it has already built.
Frames it asThe Public Staff is the state agency assigned by law to represent utility customers in rate cases[3]. Its argument for signing is that a negotiated cut you can bank beats a litigated outcome you cannot predict. It pushed the residential increase down from roughly 18% to 9.5% and the profit rate from 10.1% to 9.8%[3][4]. Industrial and business signers — the Carolina Industrial Group for Fair Utility Rates, the Carolina Utility Customers Association and Walmart — want cost certainty and reliable power for their own operations[1]. The North Carolina Sustainable Energy Association signed in part because the deal locks in coal-plant retirement dates used to set depreciation and opens a separate proceeding on a large load tariff[10]. That tariff is the real prize: it is a special rate class that would make very large customers like data centers pay for the specific grid capacity built for them, rather than spreading those costs across everyone's bill.
WhyThe Public Staff is measured by concessions won, not by cases fought. Settling early locks in reductions and moves the data-center cost question into its own proceeding, where it can be handled directly instead of buried inside a general rate case.
Impact on themIf the commission approves, the Public Staff will have cut the typical household increase roughly in half. If the commission later approves a weak large load tariff, that win narrows considerably.
Frames it asJackson's position is that the settlement improved the deal but did not fix it. He acknowledged Duke moved in the right direction, then said 9.5% 'is still too high' and refused to sign[4]. His two specific objections are concrete, not rhetorical. First, he argues 9.8% is a higher profit rate than Duke needs to attract capital, and every tenth of a percentage point is real money on a multibillion-dollar rate base[4]. Second, he argues Duke Energy Progress should give its largest users, including data centers, a path to generate their own power — if a hyperscaler builds its own generation, Duke does not have to build it and bill everyone else for it[4]. The underlying principle: a legal monopoly's customers cannot shop elsewhere, so the state must bargain on their behalf as hard as a competitive market would.
WhyJackson is an elected Democrat in a state where electricity bills are a live pocketbook issue, and utility rate cases are a traditional platform for a North Carolina attorney general. Staying out of the settlement preserves his ability to litigate and to campaign on it.
Impact on themHis refusal does not block the deal — the commission can approve a settlement without him. But it removes the appearance of unanimous consumer-side consent and keeps an active adversary in the record.
Frames it asThe hyperscalers' case is that they are the best customers a utility can have. They use power steadily around the clock, sign long-term contracts, and pay for capacity they may not fully use — which spreads the grid's fixed costs over more kilowatt-hours and can hold rates down for everyone else. They point to the tax base and construction jobs that follow. They also argue that if a state makes large-load tariffs punitive, the projects simply move to Georgia, Virginia or Texas, and North Carolina gets the higher bills without the investment. Duke's own position aligns here: it says data centers are not a major driver of this rate request[5].
WhyPredictable, cheap, fast power is the binding constraint on AI expansion. Every month of interconnection delay is capacity that competitors get first. Their interest is in a large load tariff that is workable, not one that front-loads the entire grid cost onto them.
Impact on themDuke has already signed 7.8 gigawatts of service agreements with data-center customers[6]. The separate large load tariff proceeding will set how much of the grid buildout they pay for directly[10].
Frames it asThe advocates' argument is about who bears risk. A household cannot switch electric companies, cannot negotiate, and cannot walk away. Duke can raise capital; a family on a fixed income cannot raise income. Protesters rallied against the increase in June 2026[9]. Their strongest evidence is not the profit figure but Commissioner Tommy Tucker's own question from the bench: Duke's large-load queue is roughly 70% data centers, and he asked how other customers are expected to pay for roughly $100 billion of investment by 2035[5]. They also point to national data — Reuters has reported PJM-territory households could face rate hikes of up to 60% over five years, and PJM attributed a $6.3 billion consumer cost increase largely to data-center demand[11]. A Consumer Reports survey of 2,146 U.S. adults in November 2025 found 78% were somewhat or very concerned data centers would raise their energy bills[11].
WhyConsumer groups want the lowest defensible rate. Environmental groups want a different mix of what gets built — and some, like the Sustainable Energy Association, judged the coal retirement dates and the tariff proceeding worth signing for[1][10].
Impact on themUnder the Carolinas terms, a typical residential customer's rate rises 5.9% on January 1, 2027 and another 3.6% the following year[3]. On a $150 monthly bill, that is roughly $9 more per month in year one, then roughly $6 more on top of that.
Like this article?
The Bias Ledger average rating 4.5
The same story, as framed by outlets across the spectrum, ordered least to most biased. The bias score (1 = straight, 10 = heavily spun) is an AI assessment of that framing — click an outlet to see its track record. The tell is the word choice or omission that reveals the angle.
| Outlet | Vantage | Bias | How they frame it | The tell |
|---|---|---|---|---|
| Axios | U.S. center | 2 | 'Duke Energy reaches settlement to lower its proposed rate increase.' | Plainest available framing — states the action, no adjectives. The compression Axios is built on does cost something: 'lower its proposed rate increase' can read to a skimming reader as a rate cut rather than a smaller rate rise. |
| Utility Dive | U.S. trade press (utility-industry audience, advertiser-supported) | 3 | 'Duke reduces rate hike request, still faces regulator pushback.' | Trade framing centers the regulatory process, not the household bill. Surfaces the sharpest primary-source detail in the coverage — Commissioner Tucker's question about the 70% data-center queue and $100 billion — but presents it as a business risk to Duke rather than a cost question for customers. |
| Carolina Journal | U.S. right (published by the John Locke Foundation, a North Carolina free-market advocacy nonprofit) | 4 | 'Duke settlement lowers rate hike, NC households still face 9.5% increase.' | Concedes the cut in the first clause, then pivots to the residual burden on households. Uses 'still face,' which frames the outcome as something done to households rather than negotiated for them. Notably does not foreground the profit figure — the criticism is cost-of-living, not corporate earnings. |
| NC Newsline | U.S. left (nonprofit outlet in the States Newsroom network, funded largely by progressive-aligned donors) | 5 | 'NC AG Jackson rejects Duke Energy's proposed rate increase.' | Leads with the Democratic attorney general's refusal rather than with the settlement itself, making opposition the news and the agreement the backdrop. 'Rejects' is stronger than the record supports — Jackson declined to sign a settlement he cannot veto. |
| Canary Media | U.S. left-of-center clean-energy nonprofit (funded by climate philanthropy; affiliated with RMI) | 5 | 'Data centers are key to fight over Duke electric rates in North Carolina.' | Names data centers as the story before the reporting establishes them as the driver — Duke disputes exactly that. The framing is a conclusion in headline form. Its underlying reporting on cost allocation is specific and well-sourced. |
| Duke Energy | Corporate (company press release) | 8 | 'Duke Energy Progress reaches agreement with North Carolina Public Staff and other stakeholders to deliver a lower-cost path to power North Carolina's future.' | The headline describes a rate increase as 'lower-cost' and 'a path to power North Carolina's future.' The reduction from Duke's own opening ask is the only baseline offered; the increase over what customers pay today is not in the headline. The word 'stakeholders' also quietly papers over the fact that the state attorney general refused to sign. |
References
- Duke Energy Progress reaches agreement with North Carolina Public Staff and other stakeholders to deliver a lower-cost path to power North Carolina's future — Duke Energy · Corporate press release from the company that is party to the settlement
- Duke Energy Carolinas reaches agreement with North Carolina Public Staff and other stakeholders — Duke Energy · Corporate press release
- Duke Energy-stakeholder agreement pares down residential rate increase to 9.5% over 2 years — Port City Daily · North Carolina local news, center; advertiser- and subscription-funded
- NC AG Jackson rejects Duke Energy's proposed rate increase — NC Newsline · U.S. left; nonprofit in the States Newsroom network, funded largely by progressive-aligned donors
- Duke reduces rate hike request, still faces regulator pushback — Utility Dive · U.S. utility-industry trade press; advertiser-supported
- Duke Energy (DUK) Q2 2026 Earnings Call Transcript — The Motley Fool · U.S. investor-focused financial media; subscription-funded
- Duke Energy posts $1.1B second-quarter profit, big data center demand — Business North Carolina · North Carolina business trade publication; business-audience orientation
- Duke settlement lowers rate hike, NC households still face 9.5% increase — Carolina Journal · U.S. right; published by the John Locke Foundation, a free-market advocacy nonprofit
- 'No rate hikes.' Protestors rally against proposed Duke Energy rate increases — WUNC · NPR member station; public and listener funded, center-left audience
- The Duke Energy settlement is just the beginning for North Carolina's data center energy policy — Environmental Defense Fund · U.S. environmental advocacy nonprofit; party-adjacent to the clean-energy intervenors in this case
- AI Data Centers: Big Tech's Impact on Electric Bills, Water, and More — Consumer Reports · U.S. consumer advocacy nonprofit; member-funded, takes no advertising
- State Regulation of Data Centers in 2026 – A Shifting Landscape — ArentFox Schiff · U.S. corporate law firm client alert; written for clients on the industry side