Holtec Prices Tonight at Up to $10.2 Billion. Its Operations Made Under a Tenth of Last Year's Reported Profit.

Holtec Nuclear Corporation is expected to price its initial public offering on the evening of Thursday 17 September 2026 and to begin trading on Nasdaq the following day under the ticker HNUC. The deal is 50,000,000 Class A shares in a $15.00 to $18.00 range, listing on the Nasdaq Global Select Market and Nasdaq Texas, with a 30-day underwriter option on a further 7.5 million shares. That implies equity value of roughly $8.5 billion at the bottom and $10.2 billion at the top, and gross proceeds of about $825 million at the $16.50 midpoint.

A note on a discrepancy readers will hit immediately: Renaissance Capital and 24/7 Wall St. describe this as an $825 million IPO, while Yahoo Finance and Qz headline $900 million. These are not competing facts. Both describe the same 50-million-share offering. $825 million is the midpoint and roughly $900 million is the top of the range.

The story being sold is small modular reactors and the Palisades restart. The prospectus describes a different company. Last year Holtec was, by revenue, a nuclear construction and site-services contractor with shrinking sales, and by reported profit, something closer to an investment portfolio.

The number on the cover and the number underneath

FY2025 revenue was $576.6 million, reported in some coverage as $577 million, down from $766 million the prior year. That is a decline of about 25% in a year when the entire nuclear sector was being repriced upward on data-centre power demand. Of that $576.6 million, $529.0 million was construction and site services, or roughly 92% of the top line.

Operating income on that revenue was $37.1 million. Net other income was approximately $433.9 million, driven by realised and unrealised investment gains associated with nuclear decommissioning trusts. Those two figures are consistent across every secondary source we checked, and they are the ones this article relies on.

The net income line is not consistent. Engineering News-Record reports 2025 net income of $434 million. Capital.com and Green Stocks Research both report consolidated net income of $386.6 million. We have not resolved which is the correct characterisation, and readers should not treat either as settled. It is worth noting that $37.1 million of operating income plus $433.9 million of net other income gives $471.0 million of pre-tax income, and $386.6 million reconciles to that at an effective tax rate of about 18%, while $434 million would require a tax charge close to zero.

Either way the ratio holds. Operating income was 8.5% of reported profit on the higher figure and 9.6% on the lower. Under a tenth of what Holtec reported as profit in 2025 came from operating the business.

Why a decommissioning trust shows up in a manufacturer's profit

When a US utility retires a reactor, federal rules require that the cost of dismantling it has been pre-funded through a decommissioning trust, a pool of invested capital that sits waiting to pay for a teardown that can take decades. Holtec's decommissioning business does not merely bid for that work. It acquires the retired plant itself, and the trust travels with it.

Because Holtec consolidates those entities into its financial statements, the trust assets appear on Holtec's balance sheet and the trust's investment returns run through Holtec's income statement. Not just the realised gains, when securities are actually sold, but the unrealised ones too, which are simply the change in market value of things still held.

The composition reported from the filing makes the point. Roughly $58.2 million came from asset-retirement-obligation settlement gains, $118.5 million from realised investment gains, and $230.0 million from unrealised gains. About $406.7 million of the total is those three lines.

Two consequences follow, and neither is usually stated in the coverage. The first is directional: unrealised gains reverse. A $230 million unrealised gain in a rising market becomes an unrealised loss of comparable scale in a falling one, with nothing about the nuclear business having changed. An investor pricing HNUC off a price-to-earnings multiple is, to a first approximation, paying a nuclear-services multiple for an equity portfolio.

The second is about ownership. The money in a decommissioning trust is earmarked. It exists to pay for dismantling a specific plant, and that obligation does not go away because the trust had a good year in equities. Holtec earns the decommissioning work, and to the extent trust returns exceed the cost of the teardown there is genuine surplus, but the gross investment gain is not distributable profit in the way the income statement makes it look.

What none of this establishes is wrongdoing. Consolidation is what the accounting standards require given the ownership structure, and Holtec discloses the components. The claim here is narrower and entirely about usefulness: for this issuer, in this year, net income does not tell you how the business performed.

What the money is for

Operating cash flow was negative $248.5 million in 2025, after negative $150.2 million in 2024. Capital expenditure rose about 80% year over year to $983.6 million, the bulk of it Palisades. Free cash flow was negative $1.23 billion.

Set the raise against that. An $825 million midpoint raise measured against $1.23 billion of annual free cash burn funds roughly eight months at last year's rate. Even at the top of the range, roughly $900 million buys under nine months. That is not a criticism of the deal so much as a statement of what it is: this is a funding round, not a liquidity event for a self-sustaining business.

The balance sheet moved accordingly. Holtec carried no debt at the end of 2024 and $958.6 million of total debt by 31 March 2026. Cash and equivalents fell from $106.3 million at the end of 2024 to $10.0 million at the end of Q1 2026, and working capital turned negative, with a current ratio of 0.76x.

Two things offset that picture. The Department of Energy has disbursed approximately $491.1 million of guaranteed loan funds toward the Palisades restart, so a material share of the capex is externally financed rather than equity-funded. And in December 2025 Holtec was selected as one of two recipients of a DOE Tier 1 First Mover Award, expected to provide $400 million in grant funding for the dual-unit SMR-300 plant at Palisades.

The pro forma basis, and why it is not comparable

The filing also presents a pro forma 2025: revenue of $893.4 million, operating income of $140.2 million, net income of $217.0 million, and pro forma operating cash flow of positive $158.2 million for 2025 and positive $172.8 million in the first quarter of 2026. Those are considerably better numbers.

They are also a different company. Pro forma restates the accounts for the planned reorganisation ahead of listing, and a pro forma operating cash flow of positive $158.2 million cannot be set alongside an actual negative $248.5 million as though the difference were performance. It is a difference in perimeter. Any analysis that quotes the pro forma revenue and the actual free cash flow in the same paragraph is producing a number that describes no entity that has ever existed. This piece uses the actual basis throughout and says so.

The two Holtecs the price has to choose between

Palisades has not restarted. This matters, because a good deal of the coverage reads as though it has. The plant stopped operating in 2022. The restart slipped from an October 2025 target to early 2026 and then further. Holtec described the closeout of the last major project in July 2026 as a watershed moment, and plant staff began loading fuel into the reactor vessel on 31 August 2026. If it completes, Palisades will be the first US nuclear plant to restart after entering decommissioning. That is a genuinely significant thing, and it is prospective.

The SMR-300 programme is earlier still. Holtec has submitted a licensing application to construct two units adjacent to Palisades. Per-unit capacity is reported inconsistently, with the design name implying 300 MWe and the American Nuclear Society reporting about 340 MW per unit, so readers should treat any precise figure with caution. Either way, no SMR-300 has been built.

The bull case, stated properly

The strongest argument against everything above is that free cash flow is simply the wrong lens. The $983.6 million of capex is a one-time asset build, not a run rate. You would have said the same things about any utility halfway through constructing a plant, and you would have been wrong about all of them. On this reading the operating income line is depressed precisely because the company is spending to create a generating asset that does not yet produce revenue, and the pro forma figures are the better approximation of the post-restart entity.

That argument is reasonable and may well be right. It rests on one assumption, which is that Palisades comes back and generates on something close to schedule. The counter is not that the bull case is illogical, but that the price already contains it.

Which is what the $15-to-$18 range is really choosing between. The $1.7 billion spread between the bottom and the top of the range is roughly three times the entire 2025 revenue of the company, on a business whose revenue fell about 25% last year. At $15 you are paying for a construction-services contractor with a large option attached. At $18 you are paying for the option.

What would settle it is straightforward and will take a few quarters: an operating income line that stands up without the trust gains, and the first reported revenue from a Palisades that is actually generating. Until then, the honest description of HNUC is that the headline profit number is not a profit number, and the most important asset is not yet running.