Third quarter 2026 earnings call, November 2, 2026

BWX Technologies reported third quarter revenue of $953 million, up 10% (7% organic), and Adjusted EBITDA of $180 million, up 19%. Adjusted EPS rose 26% to $1.26. Management left its 2026 operating guidance unchanged at roughly $667 million of Adjusted EBITDA and $4.72 to $4.77 of EPS, both at the midpoint of prior ranges. The call came just over a month after the company's Investor Day, so the most useful new information concerned capital commitments, the cadence of commercial orders and the first quantified view of 2027.

A U.S. Heavy Components Plant Is Going Ahead Without Firm Orders

The clearest decision signal was on U.S. manufacturing capacity. BWXT is finalizing design for a new commercial facility to make large steam generators and reactor pressure vessels, with a final investment decision targeted by year-end. Asked by Seaport's Jeffrey Campbell whether the plant would require firm orders, as a potential Poland investment would, CEO Rex Geveden answered directly: "That U.S. facility does not depend on firm orders. That said, there's so many things in the pipeline that we're certainly compelled to go ahead with that U.S. plant." He added, "We're going ahead with the U.S. plant regardless."

This is a speculative capacity bet ahead of contracts, and it raises the stakes on the timing of reactor orders. Poland is held to a stricter standard: any manufacturing investment there would be "backed by firm orders and appropriate government partnerships," with Geveden noting that discussions involved gigawatt-scale demand for microreactors and small modular reactors.

Capacity is otherwise being stacked in sequence. The Cambridge, Ontario expansion, which lifts heavy component capacity by about 50%, was finished slightly under budget and roughly on the original schedule. The recently closed PCG acquisition brought close to $500 million of backlog, and management believes the sites are about 50% underutilized, with modest capital able to unlock another 50%. Roughly 75% to 80% of PCG is naval work, which is expected to grow at a low single-digit rate, while commercial work will be built into the idle capacity.

Commercial Order Timing Is Drifting, With No Change to 2030 Targets

Large-reactor order conversion remains the main swing factor, and the timeline is looser than at the start of the year. Geveden said he had expected new U.S. reactor plant orders by the end of 2026 and "I still think that's a possibility for us, but certainly no later than about the first half of the next year." He sees SMRs ahead of large reactors, with sequencing running first through Darlington units 2, 3 and 4, then the 10 TVA reactors (5 at Clinch River and 5 at Bellefonte), followed by AP1000s.

Baird, BTIG and others probed whether a slip would endanger the back-end-weighted 2030 targets of $5.5 billion to $6 billion of revenue and $1.1 billion to $1.2 billion of Adjusted EBITDA. Geveden said it would not, stating the plan has "multiple ways to win" and incorporates only "modest success in commercial orders for new builds." That is reassuring on its face, but investors should note the company has not quantified how much commercial order flow the targets assume.

On what is holding up final investment decisions, Geveden, who said he is not in the negotiations, pointed to risk allocation rather than financing: "everyone bleeds, but no one bleeds to death." He noted the 10 X-300s and 10 AP1000s are expected to be funded by sovereign investment from South Korea and Japan. On a potential Korean funding role, he expects BWXT to compete with Doosan and IHI, with Westinghouse making content decisions, though global capacity for these components is limited. Geveden said component manufacturing risk, such as for reactor pressure vessels and steam generators, is distinct from project delivery risk, which BWXT does not intend to carry, though price pressure should be expected and is embedded in the 2030 margin profile. He also dismissed election risk, noting these deals sit outside congressional appropriations.

2027 Outlook Nudged Higher on Government Margins

CFO Mike Fitzgerald guided to high single-digit to low double-digit growth in Adjusted EBITDA and EPS for 2027. Asked by TD Cowen's Marc Bianchi about an apparent increase from Investor Day, he called it "a refinement of our expectations, nothing material." Government Operations is guided to high single-digit to low double-digit revenue and EBITDA growth, led by Special Materials. Commercial Operations is guided to high single-digit organic revenue growth but only low single-digit total growth because of the planned medical divestiture, which is on track to close later this year. Commercial margins are expected to rise 50 to 100 basis points.

Free cash flow is the soft spot. The 2026 range was narrowed up at the low end to $350 million to $360 million, but 2027 is guided flat to slightly higher. Fitzgerald cited a roughly $10 million pension headwind tied to Kinectrics, working capital for growth and milestone timing on HPDU and new enrichment programs. Capital spending stays near 6% of revenue in both years. For a company growing EBITDA near 10%, flat cash flow shows how capital-intensive the buildout is.

Government Operations Margin Strength Is Core-Driven, Not Mix-Driven

Government Operations posted a 22.6% Adjusted EBITDA margin, with EBITDA up 22% to $144 million. Fitzgerald attributed this to "a substantial increase in the efficiency of the volume of the work that we performed specifically in our naval nuclear business," plus higher technical services equity income. Newer programs contributed little to the margin beat, since the company does not take large margin adjustments on them until they are further along. Segment revenue rose only 4%, held back by lower naval materials procurement against a strong prior-year quarter. The slower year-to-date growth reflects the HPDU program performing well: under percent-complete accounting, strong margin performance lowers recognized revenue. HPDU revenue should step up in the fourth quarter and 2027 as the new Jonesborough, Tennessee facility, part of the $1.6 billion contract for 300 metric tons annually, is built.

A new naval fuel contract of $189 million was booked, and the company is negotiating the next multiyear pricing agreement. Fitzgerald noted the business is coming off older pricing arrangements, which is now helping the financials. Mix dilution from scaling newer programs is expected to be offset by productivity gains in 2027.

Advanced Nuclear: Janus, Prodigy and TRISO Capacity

BANR picked up two wins. The U.S. Army's Janus program targets more than 20 microreactors across 5 bases, with Fort Campbell first; the company is still working with partners "to fully define the program's costs, execution approach and operating plan," so economics are not yet settled. Prodigy Clean Energy selected BANR for a project in Belledune, New Brunswick, the first international microreactor win. Geveden declined to disclose pricing but acknowledged a three-reactor deal would be significant. He was also candid: "super early days," with financing still to be lined up and only an MOU for a power purchase agreement in place. Wyoming, Poland and Canada are running in parallel behind Fort Campbell.

Geveden argued BANR's scale is a differentiator: 22 MWe, 75 MWth, high temperature, and rail-shippable. He contrasted it with SMRs: "When we talk about small modular reactors, those are not small. And frankly, they're not modular. They're built in the field." He said the Navy's Crane microreactor effort is smaller and being done by a competitor.

On TRISO, BWXT signed a multiyear deal to supply an initial 1 metric ton to Antares, its first large-scale commercial fuel customer. The Lynchburg line produces a couple of hundred kilograms a year, and management can at least quadruple that output through technology changes and operating improvements. A larger Wyoming plant would require two gates: concrete demand, with Kairos and its Hermes reactor as anchor tenant plus multiple BANR orders, and commercial availability of HALEU. Neither is in hand today.

Enrichment and Capital Allocation

The defense fuels centrifuge prototype remains on track for testing by year-end. Geveden would not define success criteria, saying the aim is to reduce complexity, exercise the supply chain and build an unobligated system: "just spinning that thing will be the proof that we're looking for." The company is pursuing adding LEU and HALEU to the scope alongside the HEU facility in Erwin, Tennessee, but offered no details.

On capital allocation, Fitzgerald said the 2030 targets are anchored to 5% to 6% core capex and require no specific M&A. The $3 billion to $4 billion of potential deployment capacity cited by CJS's Bob Labick is therefore upside, and uses are prioritized as organic capacity, Special Materials and microreactors, with capital returned if opportunities fall short. Only modest 2027 buybacks to offset dilution are planned. Fitch assigned an inaugural investment-grade rating and Moody's revised its outlook to positive.

The Commercial Operations segment grew revenue 25% (15% organic) with Adjusted EBITDA up 17% to $41 million and a 13.2% margin. Management said the implied fourth-quarter step-up in the segment reflects normal seasonality and PCG contribution, with nothing unusual, and described PCG integration as going "very, very well" without specific cost synergy targets.