Google takes the nuclear option for €13B Finland expansion

Google plans to invest at least €13 billion ($15 billion) in Finnish digital infrastructure and low-carbon energy projects over the next two years, expanding its Hamina campus and building new datacenters near Kajaani, Muhos and Vaala. The company signed a 22-year PPA with Fortum that will support extending the Loviisa nuclear plant's operations through 2050; Google is expected to receive up to 50% of the plant's output from 2030-2049. Google also contracted 629 MW of new Finnish onshore wind capacity and a 94 MW battery system near Kajaani, due to connect in late 2027, supporting data-center power supply and grid stability.

Analysis

For GOOG, the strategic value is less Finnish search revenue than securing European AI/cloud capacity with lower carbon intensity and stronger data-residency credentials. The spend raises the probability of enterprise-workload wins against AWS and Azure in regulated verticals, but it also brings forward depreciation and fixed-power commitments before incremental Cloud revenue is observable; the market should demand disclosed MW capacity, utilization, and Cloud margin commentary rather than credit the full investment as growth immediately.

FORTUM gains a long-duration contracted cash-flow stream that improves the financeability of life-extension capex and materially reduces the binary risk of losing a major generating asset after 2030. The trade-off is foregone merchant-power upside: once the contracted volume ramps, Fortum will have less exposure to Finnish wholesale-price spikes, so valuation should increasingly be anchored to contract economics and extension-capex discipline rather than a simple Nordic power-price beta.

Second-order beneficiaries are European data-center electrical and cooling suppliers—VRT, ETN, ABB and Schneider Electric (SU)—but procurement timing is likely staggered across permitting, grid connection and construction milestones. The underappreciated risk is local transmission scarcity during cold, low-wind periods: if connection queues or balancing costs rise, the economics of additional capacity can deteriorate even with nominally low-carbon power supply.

Consensus may overstate the near-term GOOG earnings impact and understate the strategic hedge against European power and regulatory constraints. A more investable 6-18 month catalyst is evidence that the capacity converts into accelerating Google Cloud backlog or regional AI-service adoption; absent that, this is principally a capex and depreciation story.

Key Decisions for Investors

• Maintain a neutral-to-selectively long GOOG stance rather than chase the announcement; reassess after the next two earnings reports for Cloud revenue acceleration, capex guidance, and depreciation/margin commentary. Thesis is supported by improving Cloud growth without material operating-margin erosion; falsified if capex rises while Cloud growth decelerates.
• Initiate or add to FORTUM on weakness over a 12-24 month horizon, contingent on disclosure that contracted pricing covers extension capex and preserves acceptable returns. The upside is a lower risk premium from asset-life certainty; key risk is a below-market PPA price or extension-cost escalation, which would turn the agreement into a margin cap rather than a rerating catalyst.
• Build a watchlist long basket of VRT, ETN, ABB and SU for contract-award confirmation rather than immediate exposure. Favor VRT/ETN where power-density and cooling requirements are highest; exit the thesis if project permits, grid-connection schedules, or disclosed facility capacity are delayed.
• Monitor Finnish/Nordic wholesale power spreads and nuclear-extension approvals through 2027-28. A sustained rise in balancing costs or a regulatory delay would weaken both the data-center buildout economics and FORTUM's expected cash-flow de-risking.