3 Things Smart Investors Know About the Nuclear Power Comeback

Nuclear power is regaining global policy and financing support, with more than 20 countries pledging at COP28 to triple nuclear capacity by 2050 and the World Bank ending its multi-decade nuclear-financing ban in June 2025. AI hyperscalers are securing long-term nuclear supply: Microsoft signed a 20-year Constellation Energy PPA tied to restarting Three Mile Island Unit 1, while Meta has 20-year agreements with Constellation and Vistra. Small modular reactors could reduce deployment timelines to roughly 24-36 months, with GE Hitachi BWRX-300 projects targeted for Ontario in 2029-30 and under consideration by TVA.

Analysis

The investable bottleneck is not nuclear policy but deliverable firm power at specific grid nodes. CEG and VST monetize this scarcity first because existing generation can be contracted years before greenfield capacity is available; the read-through is higher forward power-price capture and longer-duration EBITDA visibility, not merely incremental volume. Data-center load also raises transmission congestion, creating a second beneficiary set in GEV through grid equipment and services even if new-reactor schedules slip.

SMR equities remain structurally different from incumbent generators: OKLO and SMR are long-duration financing, licensing, fuel-supply and first-of-a-kind construction options, with little near-term cash-flow support. HALEU availability, reactor-component capacity and NRC execution could delay deployments enough to force additional equity issuance; a hyperscaler memorandum or non-binding development agreement should not be valued like a contracted, financeable PPA. GEV has superior risk-adjusted exposure because reactor awards are upside while electrification equipment demand supports nearer-term earnings.

Over the next 1-3 months, announced data-center PPAs or utility integrated-resource-plan revisions can sustain the nuclear premium, but the consensus may be extrapolating bilateral contract prices across whole portfolios. The key falsifier for CEG/VST is a decline in realized forward power curves at their relevant hubs or management guidance showing contracting fails to offset outage, refueling and regulatory costs. Over 6-18 months, rising power-demand forecasts should favor firm generation, but lower AI capex, faster gas generation additions, or materially weaker natural-gas prices would compress the scarcity multiple.

Key Decisions for Investors

• Prefer long GEV over an equal-dollar short basket of OKLO and SMR for 6-12 months. This expresses nuclear/electrification spending while hedging first-of-a-kind reactor execution; reassess if GEV orders or grid-equipment margins weaken at the next two earnings prints, or if either SMR developer secures a binding, fully funded construction contract.
• Maintain CEG/VST only as a selective power-scarcity exposure rather than chase broad nuclear beta. Add following confirmed long-dated contracted-load disclosures or post-earnings guidance upgrades; take risk down if forward hub power prices fall materially for two consecutive months or capacity-factor/outage guidance deteriorates.
• Monitor uranium enrichment and HALEU procurement milestones as a gating alert for OKLO, SMR and XE rather than initiating on policy headlines. A government-backed fuel award with disclosed volumes, delivery dates and customer-funded offtake would convert part of the current optionality into a tradable catalyst; absent that, dilution risk dominates.
• Use a 3-6 month pair of long CEG or VST versus short a broad renewables proxy only if regional power-price data show sustained congestion and firm-power premiums. The trade fails if gas prices decline enough to restore gas-fired dispatch economics or if data-center project cancellations reduce utility load forecasts.