Duke Energy management recently highlighted the $16 billion increase to its five-year capital plan, bringing it to a record $103 billion, as it reaffirmed its 5% to 7% long-term EPS growth outlook through 2030. The consensus reading of that headline is bullish - the utility is on track for a multi-year earnings expansion fueled by AI data center demand, and the stock hasn't fully caught up.
I've been very surprised that investors accept that framing without asking what "capital upside" actually means for the shareholder sitting on the other side of $141.6 billion in total debt, $3.3 billion in negative free cash flow, and $10 billion in planned new equity issuance.
The false narrative here is that management's internal valuation estimate translates into share price upside for existing holders. It doesn't - not without understanding how the capital gets funded and what dilution price pays for.
Quick Backtesting Tool
Let me decompose the Duke EnergyDUK-- thesis the way an engineer would: pillar by pillar, with the numbers first.
The data center story is real, but the funding math is brutal
Duke Energy is no exaggeration when it calls its five-year $103 billion capital plan the largest ever for a regulated U.S. utility. The company boosted that plan by $16 billion in February and is already spending more than $1 billion per month. About 60% of that capital goes to new power generation - 14 gigawatts of new capacity plus 4.5 gigawatts of battery storage over five years. The remaining 40% funds grid expansion and hardening.
Duke has signed electric service agreements totaling 4.5 gigawatts of data center load, with Amazon, Microsoft, Google, and Meta among its customers. Duke counts Amazon, Microsoft, Google, and Meta as major data center customers. The company says its pipeline of potential data center demand has topped 9 gigawatts.
That's the bullish half of the story. These are contracted customers, not speculative growth. The load ramp is expected to begin in late 2027 and accelerate into 2028. The structural demand shift from AI compute is not a cyclical flicker - it's the real reason Duke's rate base is expanding.
However, the capital spending to serve that demand is consuming every dollar of operating cash flow and then some. DukeDUK-- generated $11.66 billion in operating cash flow over the trailing twelve months but spent $14.96 billion on capital expenditures. Free cash flow came in at negative $3.3 billion - a 1,646% decline year over year. That's not a rounding error. It means Duke is funding its own growth by borrowing more money and issuing equity.
The debt mountain and the dilution plan
Duke Energy carries $141.6 billion in total debt against $56.5 billion in equity, for a debt-to-equity ratio of 159.8%. That is the kind of leverage that works in a stable, regulated environment with predictable returns on invested capital - and breaks badly when rates stay elevated for longer than expected or a rate case goes wrong.
The company's net debt sits at $88.1 billion, and interest expense is one of the primary drivers eating into second-quarter earnings. Duke's return on invested capital stands at 5.77%, and return on equity is 9.85%. Those are acceptable returns for a regulated utility but not exceptional. The spread between cost of capital and return on invested capital is thin enough that any misstep in rate-setting or cost recovery narrows the margin quickly.
Now layer in the equity plan. Duke's CFO Brian Savoy told investors the company expects to issue approximately $10 billion in new equity between 2027 and 2030. That figure is a floor, not a ceiling, given CEO Harry Sideris's own admission that the capital plan "will probably go up as we move into the future."
A $96.6 billion market cap company raising $10 billion in equity is looking at roughly 10% dilution - more if the stock price declines, as it has over the past month, dropping 4.1% over the last five days and 3.4% over the last twenty. The "capital upside" management references is the value created in the rate base, not the per-share value accruing to today's holders after dilution.
Q2 2026: The beat that doesn't change the structural picture
Duke reported second-quarter 2026 adjusted EPS of $1.43, beating estimates. GAAP EPS was $1.38, up from $1.25 a year earlier. Operating revenues grew to $7.59 billion from $7.51 billion. The company reaffirmed full-year 2026 adjusted EPS guidance of $6.55 to $6.80.
Those numbers are fine for a utility. Revenue growth of 7.2% year over year and gross profit growth of 11.3% reflect the expanding rate base and infrastructure investment recovery. But the Q2 beat sits on the same structural backdrop: capex of $15 billion annually, growing debt service costs, and an earnings profile that depends entirely on regulators approving the returns Duke needs to make this capital program worthwhile.
Duke's rate case in North Carolina - its largest and most important territory - is still pending. South Carolina recently approved rate increases, which is a positive. The Carolinas utility combination, targeted for January 2027, is projected to deliver $2.3 billion in customer savings through 2040, but that savings language is customer-facing, not shareholder-facing.
Where Duke ranks among utility peers
Of the major utilities I track, Duke Energy trades at 19 times trailing earnings and 20.7 times forward earnings, with a 3.43% dividend yield and 64.9% payout ratio. That valuation sits near the middle of its peer set - below Southern Company at 22.9x trailing PE but above NextEra at 19.5x. Dominion Energy trades at 23.9x with a higher 3.82% yield.
Duke's dividend track record is solid: 20 consecutive years of dividend growth over 22 years of payments. The payout ratio, while elevated, is still within a sustainable range for a regulated utility - assuming returns on that expanded rate base materialize as planned. The dividend per share stands at $4.25 trailing twelve months.
But here's the distinction that matters. Southern Company carries a lower EV/EBITDA of 13.1x despite a higher PE, suggesting its earnings power relative to enterprise value is cheaper. NextEra has a larger market cap at $181 billion and a higher EV/EBITDA of 18.3x, reflecting its renewable scale advantage. Duke sits in an awkward middle - too expensive to be a deep value utility, not differentiated enough to command a premium multiple.
The investor's calculation
Duke Energy is positioned to benefit from the AI data center boom combined with population growth across the Carolinas and Florida. That's the structural thesis, and I believe it's genuine. The company serves 8.7 million electric customers and 1.6 million natural gas customers across six states, and it's vertically integrated in a way hyperscalers value - one point of contact from grid planning to generation.
That being the case, the investment question isn't whether Duke's capital plan makes operational sense. It makes sense. The question is whether existing shareholders capture the upside or subsidize it.
My calculation is this: with $3.3 billion in negative free cash flow, $141.6 billion in debt, a pending North Carolina rate case, and $10 billion in planned equity dilution, the capital upside management describes is the utility's upside, not yours. The rate base grows, the earnings base expands, and the EPS growth target of 5% to 7% is achievable - but per-share earnings growth after dilution looks considerably lower.
At 19 times trailing earnings and a 3.43% yield, Duke Energy isn't cheap. The stock is up 5.8% year-to-date but down 1.4% on a rolling annual basis, trading roughly 8% below its 52-week high of $134.49. That gap represents regulatory uncertainty and rate pressure, not a dislocation.
I rate Duke Energy as a Hold. The 3.43% dividend yield and 20-year streak of consecutive increases make it a defensible income holding, and the data center demand story provides a real structural tailwind. But the combination of massive dilution, heavy leverage, and thin returns on invested capital means the capital upside management describes is a management valuation exercise, not a shareholder return promise. In my opinion, the stock needs to either fall materially to offer entry value or demonstrate that per-share earnings - not just aggregate earnings - are growing fast enough to justify the capital it's raising to fund the buildout.
For investors seeking utility exposure with a better risk-reward profile at these levels, I favor Southern Company for its cheaper enterprise valuation relative to earnings power, and NextEra for its renewable generation scale, which positions it ahead of the gas-heavy buildout Duke is executing. Of those two, I lean Southern Company for the combination of lower EV/EBITDA, similar yield, and a less dilutive capital trajectory.
