Nuclear M&A Hits Seven-Year High but Reactor Delivery Lags by a Decade
Western governments and hyperscalers are pouring capital into nuclear at record rates, but the IEA's 70-gigawatt build pipeline won't ease near-term data-center load pressure.
The URA uranium exchange-traded fund was down 1.33% at $48.05 on Thursday (2026-08-27), a move that reflects growing market scrutiny of the gap between nuclear's investment surge and actual reactor delivery timelines.4
The M&A data tell the bullish story. White & Case and Mergermarket figures show 25 transactions closed in 2025, up from 17 in 2024, with total deal value reaching $1.5 billion — the highest annual total in seven years. Corporate power buyers, sovereign funds and infrastructure investors are competing for a finite pool of assets. Signed deals, though, do not produce electrons.2
The demand side is moving faster than any reactor build can match. IEA data project U.S. data-center power consumption rising from 34.7 gigawatts in 2024 to 106 GW by 2035, more than tripling in eleven years. Microsoft, Amazon and Google have all signed offtake agreements tied to nuclear supply, according to Forbes, racing to lock in carbon-free baseload before competition for it tightens further. The IEA put global data-center investment at roughly USD 580 billion in its most recent review.1
Washington's policy response is ambitious but long-dated. The U.S. government is targeting a quadrupling of domestic nuclear capacity by 2050, with accelerated permitting for advanced reactor designs. The IEA expects global nuclear spending to exceed USD 100 billion annually under stated policies, and projects more than 70 GW of new nuclear capacity coming online by the mid-2030s — the strongest build rate in 30 years, the agency said.1
That mid-2030s horizon offers limited relief to grid operators and corporate buyers managing load growth right now. Nuclear investment has expanded by more than 70% over five years, per IEA figures, but the spending funds projects whose output arrives well after the near-term capacity crunch peaks.1
Europe faces the same arithmetic from a weaker starting position. European Commission President Ursula von der Leyen declared in March 2026 that the continent's retreat from nuclear had been a "strategic mistake" and pledged €200 million for a new generation of small modular reactors. SMR programs across Europe remain at demonstration stage, with no commercially operating SMR project in Western Europe providing a proven delivery template.1
Germany's exposure is the most direct. Berlin closed its last reactors in 2023, and German power traded at €136.55/MWh on Thursday (2026-08-27), a price that reflects the cost of a grid that depends on gas imports when domestic baseload generation is thin. Reversing the phase-out would take years of regulatory and legislative work that Berlin has not begun.2
The Council on Foreign Relations, writing on August 20 (2026-08-20), flagged that expanding civilian nuclear programs carries proliferation risk because enrichment and reprocessing capabilities developed for power generation can be directed toward weapons-grade material. That concern does not stop capital from moving, but it adds complexity to licensing processes — particularly in countries lacking established nuclear regulatory infrastructure.3
The commercial logic for small modular reactors, which have attracted the largest share of advanced nuclear investment and political attention, depends on achieving series-production economics no developer has yet demonstrated at scale. Nuclear first-of-kind builds carry a consistent record of cost overruns and schedule delays. The IEA's 70-GW projection is a stated-policy estimate, not a committed delivery schedule.1
NYMEX Henry Hub front-month gas was at $2.87/MMBtu on Thursday (2026-08-27), cheap enough to keep gas-fired generation competitive across most of the United States and blunting near-term financial pressure on nuclear developers to move faster. In Europe, ICE Endex TTF front-month gas held at €65.63/MWh on Thursday (2026-08-27), sustaining long-term demand for zero-carbon baseload alternatives without providing any bridge supply while new reactors remain under construction. [live prices]
The pace at which the U.S. Nuclear Regulatory Commission processes advanced reactor license applications is the next concrete test of whether mid-2030s delivery dates are achievable. First-wave permits in nuclear programs have historically slipped; if they do again here, the 106-GW data-center load the IEA projects by 2035 collides with a supply shortfall that no volume of deal activity can quickly close.1