Constellation Energy (CEG)
The largest listed US nuclear operator: 31.7 GW owned capacity pre-Calpine, FY2025 Adjusted Operating Earnings of $2,944M, clean-baseload PPA optionality, and ~$89B market cap.
Educational analysis for professional use. This profile is not investment advice or a recommendation to buy or sell any security, and it is not personalised.
Snapshot
The Nuclear Baseload Benchmark
Constellation is the cleanest way to learn how baseload nuclear trades in a generation market obsessed with contract quality. At FY2025 year-end the company owned 31,676 MW of generation capacity: 22.1 GW nuclear, 7.0 GW gas and oil, and 2.6 GW renewables, plus 4.8 GW of third-party capacity under long-term unit-specific purchase contracts. Nuclear supplied 68% of total electric supply in GWh terms. FY2025 Adjusted Operating Earnings were $2,944 million ($9.39 per share), and market capitalisation was roughly $89.0 billion as of 9 June 2026. Constellation publishes neither adjusted EBITDA nor CAFD, the cash-available-for-distribution metric yieldcos report; the Adjusted Operating Earnings line above is the filed cash-proxy anchor instead.
Baseload quality plus data-centre offtake set the thesis. Crane restart (~835 MW) carries a 20-year power purchase agreement (PPA) with Microsoft and a DOE loan guarantee of up to $1.0 billion. Nuclear runs at house planning capacity factors around 91% (EIA CY2024 actual 90.8%), which is why one gigawatt of nuclear delivers roughly 3.8 times the MWh of one gigawatt of utility-scale solar at 24% planning norms. That maths matters when hyperscalers want around-the-clock clean power rather than shaped renewables.
Post-Calpine pro forma capacity of ~55 GW (closed 7 January 2026) is a separate base from the FY2025 standalone fleet above. Do not blend the two when screening leverage, earnings, or contract proxies.
Business Overview
Constellation operates the largest US nuclear fleet among listed independent power producers. Gas and oil fill regional gaps; renewables are a minority of owned MW but part of the forward clean-energy sales path. Management's March 2026 deck points to a path to add or extend ~9,350 MW from 2026 through 2030 and beyond, separate from the Calpine acquisition footprint.
Contract disclosure here is forward-looking and split by product rather than reported as one filed PPA percentage. For 2026, roughly 25% of expected baseload clean MWh (nuclear plus geothermal) sits under long-term agreements, while 72% of natural gas and other base gross margin is attributable to contracts; the remainder is merchant exposure on gas margin and most nuclear MWh. Forward margin splits instead of a consolidated generation share are typical for merchant-leaning baseload names, which is why the PPA vs merchant guide puts contract profile ahead of headline multiples.
How the Economics Work
Adjusted Operating Earnings is Constellation's primary non-GAAP earnings metric, aggregating nuclear uptime, capacity-market revenues, retail and wholesale margins, and corporate costs into one line analysts can track quarter to quarter. Without adjusted EBITDA, Constellation cannot sit on the same EV/EBITDA grid as Vistra or Brookfield Renewable; the comp set uses operating earnings and capacity quality instead.
Net debt was approximately $5.35 billion at 31 December 2025 (total debt $8,992 million less cash $3,641 million), a figure Constellation does not label directly. Dividing that by Adjusted Operating Earnings gives roughly 1.8× leverage on this derived proxy, conservative relative to merchant peers. The comparison only works with a label swap, though: the metric is earnings-based, not EBITDA-based, so cross-company leverage screens need to convert first.
Capacity revenue and energy margin pull in different directions by region. Cap-bound PJM capacity auction years can clear well above the $250/MW-day level this site uses as a planning assumption. Nuclear units with high capacity factors capture both energy and capacity payments; the capacity factor guide shows why baseload uptime dominates long-term average (LTA) generation maths.
Valuation Framework
Constellation screens on baseload scarcity, contract additions (Crane, Microsoft), and Adjusted Operating Earnings growth rather than a filed EV/EBITDA. Peer merchant and yieldco names in this vertical trade on adjusted EBITDA or CAFD; Constellation discloses neither, so any EBITDA multiple applied to it would be invented rather than filed.
Nuclear uptime plus incremental LT clean MWh sales tied to data-centre load growth is the scenario that plays in Constellation's favour. Merchant re-exposure on the majority of baseload clean MWh still sold into hub markets is the offsetting risk, alongside integration risk as Calpine's ~55 GW pro forma fleet layers onto a balance sheet that was lightly levered on FY2025 standalone numbers.
Key Risks
Merchant nuclear margin. Only ~25% of forward baseload clean MWh is contracted. Hub price moves and capacity-market reform still flow through most of the nuclear book.
Calpine integration boundary. Pro forma ~55 GW is not in FY2025 standalone stats. Mixing bases will mis-state leverage and contract profile until consolidated filings catch up.
Metric mismatch in comps. Screening Constellation against Vistra's $5,912 million Ongoing Operations Adjusted EBITDA or Brookfield's CAFD requires converting frameworks, not copying multiples.
Execution on Crane and the ~9.4 GW path. Restart timing, PPA counterparty credit, and permitting on the forward pipeline determine whether baseload scarcity feeds through to reported earnings or stays a narrative.
What the Screening Shows
Against the Renewables and IPPs Sector Primer thresholds:
- Capacity: 31.7 GW owned pre-Calpine; 22.1 GW nuclear. Baseload-heavy; post-Calpine ~55 GW flagged separately.
- Contract profile: Forward proxy ~25% clean MWh LT, 72% gas margin contracted (2026 deck). Not a filed consolidated PPA %.
- Cash metric: Adjusted Operating Earnings $2,944M. No adjusted EBITDA or CAFD.
- Leverage: ~1.8× net debt ÷ Adjusted Operating Earnings (derived). Conservative on that proxy; not comparable to EBITDA-levered peers without conversion.
- Pipeline: Crane ~835 MW with Microsoft PPA; ~9,350 MW forward path per Mar 2026 deck.
Filed numbers here are nuclear-heavy capacity and Adjusted Operating Earnings, not yieldco CAFD. The screening question is how fast LT clean MWh contracts close the gap with Brookfield-style ~89% contracted generation proxies on a book where only ~25% of forward baseload clean MWh is contracted today.
Renewables & IPPs Sector Primer
A multi-period DCF splits contracted PPA years from the merchant tail, pricing each off realised power price and term, then checks the value against spark spread and CAFD.
The Excel model is the primer's two archetype builds live across 10 sheets: a multi-period DCF with a contracted-or-merchant toggle, step-down discount rates for the PPA term and the merchant tail, a spark-spread build for the merchant case and a CAFD-and-payout bridge for the yieldco. Change the realised PPA price, the contracted discount rate or the merchant spread and the value per share moves; the premium and payout sheets update alongside it. It also splits the contracted premium into the part the pipeline explains and the part contract quality explains, splits the PV between contracted years and the merchant tail, and reads EV per kW on both archetypes.