Duke Energy grew Q2 2026 adjusted EPS 14% to $1.43 almost entirely on approved rate increases and riders rather than demand, while six-month capital spending jumped 28% against a 15% drop in operating cash flow, funded by $5.3bn of asset sales.
Revenue
$7.6B
+1.1% YoY
Net income
$1.1B
+10.9% YoY
Diluted EPS
$1.38
+10.4% YoY
Operating margin
27.0%
Rate increases, not demand, carried the quarter
Duke Energy's second quarter of 2026 produced reported earnings per share of $1.38, up from $1.25 a year earlier, and adjusted EPS (management's preferred measure, which strips out items it considers one-off) of $1.43. Revenue barely moved — up 1.1% to $7,592 million — while operating income rose 12.0%. That gap is the most important thing in the filing, and it is not a story about selling more electricity.
Two mechanical effects explain most of it. First, $278 million of storm-recovery revenue at Duke Energy Florida rolled off compared with last year. That is money the utility collects from customers to pay back the cost of past hurricane restoration; because the collection is matched by an expense that writes down the deferred storm cost, it is close to earnings-neutral. Sure enough, operation and maintenance expense fell $255 million, "primarily due to lower storm amortization at Duke Energy Florida." Revenue and cost dropped together, so the headline revenue line understates the underlying business and the margin line overstates the improvement. Excluding that swing, revenue rose roughly $362 million, or about 5% measured against the reported prior-year base.
Second, what actually moved earnings was regulated cost recovery: higher base rates approved by state commissions, and riders (surcharges that recover specific investments outside a full rate case).
Metric
Q2 2026
Q2 2025
YoY Change
Total operating revenues
$7,592M
$7,508M
+1.1%
Operating income
$2,049M
$1,830M
+12.0%
Operating margin
Read 0 community reports on Duke Energy, or write your own.Write a report
Operating margin — the share of revenue left after the costs of running the business, before interest and tax — is flattered by the storm mechanism described above. Holding that flat year over year (adding the $278 million of revenue and $255 million of matching expense back into this quarter) puts the margin at roughly 26% rather than 27%. The improvement is real; it is just about a point smaller than the reported figures suggest.
The earnings bridge: recovery in, asset costs out
The company's own quarter-over-quarter EPS walk is unusually clean, and it is worth reading line by line (all figures per share, on the adjusted basis):
Driver
Impact on EPS
Rate case impacts, net
+$0.10
Riders and other retail margin
+$0.08
Volume
+$0.08
Wholesale
+$0.04
AFUDC equity
+$0.02
Depreciation and amortization
-$0.09
Interest expense
-$0.04
Weather
-$0.02
Other
+$0.01
Total
+$0.18
Roughly $0.22 of the gain comes from being allowed to charge more (rate cases, riders, wholesale contracts), against $0.13 of cost from the assets and debt that earn those charges. That is the regulated-utility treadmill working as designed: Duke spends capital, commissions approve recovery of it, and the spread between the approved return and the depreciation-plus-interest drag is the earnings growth. It also means the growth rate is a function of regulatory outcomes and capital deployment, not of customer demand, which is why the volume numbers below matter less to this quarter than they would at an industrial company.
On the revenue side the filing is specific: a $145 million increase in fuel revenues (a pass-through of fuel costs, not margin), $117 million from higher pricing from rate cases, $84 million from higher weather-normal retail sales volumes, $74 million from riders — chiefly Duke Energy Florida's storm protection plan and Duke Energy Ohio's distribution capital investment rider — and $52 million from higher wholesale capacity volumes and rates at Duke Energy Progress.
Demand: weather hurt, underlying sales grew about 1.3%
Reported retail sales rose only 0.3%, and the filing attributes a $20 million revenue decline to "less favorable weather in the current year." Weather-normalized — that is, adjusted for what sales would have been in typical temperatures, which for a utility is the cleaner read on underlying demand — total retail sales rose 1.3%. The composition:
Customer class
Reported GWh change
Weather-normal change
Residential
+1.7%
+2.1%
Commercial
+1.3%
+1.7%
Industrial
-2.7%
-0.5%
Total retail
+0.3%
+1.3%
Residential and commercial growth tracks customer count, which rose 1.4% to 8.73 million — Duke's Southeast service territories are still gaining population. Industrial volumes shrank even after normalizing for weather, and the industrial customer count fell 2.1%. Management's growth case rests heavily on data-center load, and the filing says contracted data-center capacity continues to expand with "financial protections designed to support system reliability" and to "align the costs of serving new large-load customers with the customers driving those investments" — but it puts no figure on contracted capacity, so this quarter's numbers provide no way to size that pipeline.
By jurisdiction, four of five electric utilities grew volumes; Duke Energy Indiana was the exception, at 7,022 GWh versus 7,538 a year ago.
Nuclear capacity factor — the share of theoretical maximum output the fleet actually produced — was 93%, down from 99%. Less of Duke's cheapest generation showed up, which is consistent with fuel and purchased power costs rising $35 million "primarily due to higher purchased power costs." In a regulated utility most of that is recovered from customers rather than absorbed, so the earnings effect is small, but it does push customer bills up at a time when affordability is an explicit regulatory theme.
GAAP versus adjusted, and what the $0.05 gap is
The $0.05 difference between reported EPS of $1.38 and adjusted EPS of $1.43 is a $39 million after-tax charge (a $51 million pretax charge net of a $12 million tax benefit) for regulatory settlements tied to Duke Energy Carolinas' 2025 North Carolina rate case and an energy-efficiency program liability. That shows up in the consolidated impairment line, which jumped to $49 million from $3 million.
The six-month picture is where the adjustments get more interesting. Year-to-date reported EPS is $3.35 against adjusted EPS of $3.36 — nearly identical, but only because two large items of opposite sign cancel: a $368 million pretax gain on the sale of Piedmont's Tennessee natural gas business, offset by legal and regulatory settlement charges. Investors reading the near-match as "no unusual items" would be reading it wrong.
The same gain distorts the Gas Utilities and Infrastructure segment. Its six-month segment income of $542 million versus $355 million looks like a 53% jump, but $374 million of gains on asset sales sits inside that operating income; the underlying gas business did not improve on anything like that scale. The segment's effective tax rate for the half rose to 35.4% from 19.7% because goodwill allocated to the sold business was not tax-deductible. In the second quarter alone, with the gain behind it, gas segment income was $10 million against $6 million — the second quarter is seasonally irrelevant for gas distribution.
Funding: capital spending is outrunning cash generation
This is the part of the filing that deserves more attention than the EPS beat.
Capital expenditures for the six months were $8,240 million, up 28.2% from $6,428 million. Over the same period, cash from operations fell to $4,272 million from $5,040 million. Duke closed the gap with one-time proceeds of roughly $5.3 billion: about $2.8 billion from selling a 9.19% stake in Florida Progress (the holding company of Duke Energy Florida) to an affiliate of Brookfield Super-Core Infrastructure Partners in March, and about $2.5 billion from selling Piedmont's Tennessee business to Spire. The filing states the Piedmont proceeds were used partly to repay debt and partly to fund the capital plan "primarily by displacing the issuance of common equity in the near term" — in plain terms, Duke sold assets instead of issuing shares, avoiding dilution of per-share earnings.
The Brookfield deal is dilution in a different form. Net income attributable to non-controlling interests rose to $53 million in the quarter from $23 million: that is the share of Duke Energy Florida's profit that now belongs to Brookfield rather than to Duke shareholders. Further staged investments are anticipated through 2028, so that leakage grows.
Interest expense rose 6.7% to $957 million on higher debt balances across jurisdictions, plus lower returns on deferred storm balances and accrued financing costs on deferred nuclear production tax credit liabilities at Duke Energy Carolinas.
Takeaway: The 14% adjusted EPS increase came almost entirely from regulatory recovery of capital Duke has already spent, not from demand — and with six-month capital spending up 28% while operating cash flow fell 15%, the funding of the next round came from selling a stake in the Florida utility and the Tennessee gas business. The earnings model is working; the question the filing raises is what funds it in 2027 and beyond, once the one-time proceeds are consumed and the Brookfield tranches are exhausted.
Regulatory scorecard
Recovery outcomes in the half were mostly favorable, with one exception worth naming:
Revised base rates took effect in the first quarter for Duke Energy Carolinas' and Duke Energy Progress' South Carolina territories and Duke Energy Kentucky's gas business.
Settlements were reached in July in Duke Energy Carolinas' 2025 North Carolina rate case and in proceedings on Winter Storm Fern, the late-January storm that drove record winter peak demand across the Carolinas.
The South Carolina commission approved a new combined-cycle unit in Anderson County and out-of-state certificates for combustion turbines at Marshall Steam Station and combined-cycle units in Person County, North Carolina — the generation build behind the capital plan.
FERC authorized the planned combination of the two Carolinas electric utilities, with settlements reached in both states and approvals from the North Carolina and South Carolina commissions; the targeted effective date remains January 1, 2027.
Against that: on June 25, FERC ordered the removal of the return-on-equity adder from Duke Energy Ohio's transmission formula rate and required refunds back to February 24, 2022. Duke Energy Ohio cut operating revenues by $15 million in the quarter and carries a $20 million regulatory liability, with refunds expected by the first quarter of 2027. A rehearing request was filed on July 24. The amount is immaterial to consolidated results; the precedent — a regulator clawing back an authorized return — is the kind of item that matters more than its size.
Two open risks the filing flags without quantifying: the EPA's 2024 coal-ash rule, which Duke says could have a material impact and which the EPA has since proposed to partly rescind, and goodwill at Duke Energy Ohio's gas reporting unit, the only reporting unit whose fair value did not materially exceed carrying value at the August 2025 annual test.
Outlook
Management reaffirmed 2026 adjusted EPS guidance of $6.55 to $6.80 and a long-term adjusted EPS growth rate of 5% to 7% through 2030 off the 2025 midpoint of $6.30, adding that it has "confidence to earn in the top half of the range beginning in 2028." The company does not forecast GAAP EPS.
Half-year adjusted EPS of $3.36 is just over half the $6.68 guidance midpoint, and the third quarter is seasonally the strongest for a summer-peaking Southeastern utility, so the 2026 range looks reachable without unusual help. The year-to-date drivers — rate case pricing, riders, wholesale, weather-normal volume growth — are recurring rather than one-off, which is what a guidance reaffirmation should rest on.
Our read on trajectory: the earnings guide is the easy part. The 5% to 7% growth rate is arithmetically a function of rate base expanding faster than shares outstanding and interest costs, and Duke has bought itself two years of equity-free funding through asset sales. What that buys is time, not a permanent solution — capital expenditures running 28% above last year against falling operating cash flow is a gap that eventually gets closed by issuing equity, issuing more debt, or moderating the capital plan. The Carolinas utility combination targeted for January 1, 2027 and the staged Brookfield tranches through 2028 are the two levers management has already identified; the Ohio FERC decision and the pace of industrial and data-center load are the two things most likely to move the answer in the other direction. Nothing in this quarter changes the growth algorithm, but the funding side of it is where the next surprise is more likely to come from than the earnings side.