NuScale Power Is Down 84% From Its 52-Week High. Is It Finally Time to Buy the Dip in the Nuclear Start-Up?

NuScale (SMR) remains the only U.S.-NRC standard-design-approved SMR provider, but its competitive edge is eroding as DOE’s Reactor Pilot Program and Executive Order 14300 aim to cut licensing timelines from several years to ~18 months. The company has one project in Romania targeting deployment of six 77 MWe modules by 2033 (with NuScale building one first), and it is targeting a potential up-to-6 GW TVA deal by end-2026. Despite the AI/data-center tailwind for reliable baseload nuclear power, shares are down ~84–85% from their peak, and the lack of firm commitments beyond Romania keeps the setup high-risk.

Analysis

The key market mechanism is not “nuclear is back,” but that scarcity premium around SMR’s regulatory lead is eroding faster than the addressable market is expanding. If licensing can be compressed toward ~18 months, the moat shifts from approvals to execution, financing, and customer credit quality — areas where pre-revenue developers typically get punished. That argues for multiple compression in SMR unless it converts the paper advantage into contracted backlog quickly.

For hyperscalers, the implication is more favorable than the headline suggests: they gain negotiating leverage by having more credible power options, which should improve procurement terms versus taking a single-vendor bet. The near-term beneficiaries are existing nuclear owners, grid-adjacent utilities, and equipment/service names with de-risked sites, because they can deliver firm power sooner than greenfield SMRs. In other words, demand is real, but the capture rate likely migrates away from pure-play developers.

Catalyst timing matters. Over the next few weeks, SMR can still squeeze on any TVA headline or Romania milestone, but absent a firm commitment by year-end, the market will refocus on dilution risk, long project timelines, and the gap between approvals and cash flow. Over 6-18 months, the structural winner is likely whoever owns the asset base and interconnection, not who owns the most design approvals. The contrarian miss is that faster regulation is bearish for the best bull case in SMR: it removes the “years ahead” narrative without solving commercialization.

Key Decisions for Investors

• Short SMR on strength or via 3-6 month put spreads; thesis is valuation reset as regulatory scarcity fades. Falsify if SMR announces a binding TVA contract with meaningful customer funding/prepayment, or if backlog inflects materially in the next earnings cycle.
• Use SMR as a tactical event-driven long only after a signed offtake/financing package, not on headline optimism. Entry should be post-announcement with a tight stop if the deal is non-binding or heavily conditional.
• Prefer long exposure to the power-secured hyperscaler basket (GOOGL, MSFT, META) on any AI-capex drawdown rather than chasing SMR; these names have the balance sheets to lock in firm power and arbitrage vendor competition.
• Relative-value idea: long hyperscalers / short SMR as a way to express “AI power demand is real, but pure-play SMR commercialization is not.” Best implemented on any sector-wide nuclear rally, with SMR the more fragile leg.
• Watch item: if TVA or another Tier-1 utility signs a firm SMR commitment before year-end, cover shorts immediately; that would be the first evidence that customer validation is catching up to regulatory approval.