On May 13, Fervo Energy went public at $27 a share and closed its first day of trading up 35%, pushing the company’s market capitalization past $10 billion. In the lead-up, demand was strong enough that Fervo increased the number of shares on offer and raised its asking price twice before the offering closed.
The customer book was anchored by the largest geothermal power purchase agreement ever signed: for a 320-megawatt deal with Southern California Edison, for power from Fervo’s under-construction Cape Station project in Utah.
It’s certainly the most significant commercial milestone in U.S. geothermal history. The $1.89-billion offering gave an industry that had operated for two decades as a series of pilot projects and government cost-share grants a publicly traded flagship.
Hyperscaler demand has made clean baseload capacity a scarce resource in the U.S. data center buildout. Fervo had spent the prior eighteen months signing contracts to deliver exactly that, including a 115-MW agreement with Google for its Corsac Station project in Nevada that also yielded the hyperscaler’s “clean transition tariff,” a novel incentive model to get emerging clean energy projects built.
However, what the IPO does not price is the dependency stack underneath Fervo’s success so far. These dependencies have a history of hitting capital-intensive energy technologies hardest at the moment of greatest enthusiasm. The most recent example is U.S. offshore wind, which at the height of its momentum in 2022 attracted $4.4 billion at a single federal lease auction.
Only four years later, the offshore wind industry finds itself on life support — though almost none of the damage is the result of problems with the technology itself.
The offshore wind experience
Over three days in late February 2022, the Bureau of Ocean Energy Management awarded six lease areas in the New York Bight to a mix of European utilities, oil majors, and infrastructure investors. The largest single payment came from Bight Wind Holdings, a joint venture of RWE Renewables and National Grid, which paid $1.1 billion for a single tract, more than eight times the prior record for a U.S. offshore wind lease.
This was the culmination of months of market enthusiasm. State procurement programs across the Atlantic coast were stacking up, the Biden administration had set a 30-gigawatt deployment target for 2030, and major developers were competing against each other for development rights. The technology worked. The capital was there. The only question that seemed left was how fast the industry could build.
Ten months later, though, warning signs emerged. In December 2022, Avangrid filed to dismiss its power purchase agreements for the 1,232-MW Commonwealth Wind project in Massachusetts, saying the contracted $72-per-megawatt-hour price could no longer be financed under post-pandemic inflation and rising interest rates. The company eventually paid $48 million in termination penalties to exit.
By November 2023, Ørsted had taken approximately $4 billion in impairment charges and cancelled its Ocean Wind 1 and Ocean Wind 2 projects in New Jersey. In January 2024, Equinor and BP terminated their offtake agreement for the 1,260-MW Empire Wind 2 project in New York after the state rejected their request to raise the contracted strike price. Within two years of the NY Bight auction, more than five GW of contracted U.S. offshore wind capacity had collapsed.
Again, very little of this had anything to do with the technology; offshore wind turbines work. The problem was that the market embraced the industry because of a set of assumptions that stopped being true at the same time.
The Federal Reserve raised its benchmark interest rate from near-zero in March 2022 to above 5% by mid-2023, well after developers had locked in fixed-price contracts sized for a lower-rate environment. Supply chain costs surged simultaneously, with equipment caught up in Russia-Ukraine war disruptions and general post-COVID delays.
Then, in early 2025, the second Trump administration froze offshore wind permitting on day one, removing the political foundation the offshore wind industry had built its pipeline around.
Geothermal’s multi-conditional bet
Fervo’s IPO is priced on a set of assumptions that are each individually plausible — but collectively fragile. The chart below shows what happens to Fervo’s target LCOE when any single one of them breaks. What it can’t show is what happens when they break together.
Cape Station’s first phase, scheduled to come online by the end of this year, is delivering power at $7,000 per kilowatt, competitive with nuclear power but roughly three times the cost of a new gas plant. Competing with gas would mean getting down to $3,000 per kW, which would require Fervo to replicate and improve its drilling performance across many more sites than it has built to date. Between 2022 and 2025, the companyreduced drilling times by roughly 75%, and per-foot drilling costs by about 70%, a genuinely impressive rate of improvement. However, those gains came from iterating on a single project in a single geology. The learning curve that gets geothermal to $3,000 per kilowatt is a fleet learning curve, not a pilot improvement curve.
Of Fervo’s disclosed $7.2 billion in potential revenue backlog, 658 MW is backed by binding power purchase agreements with Southern California Edison, Shell, and Google. It also has a non-binding 3-GW framework agreement with Google that does not obligate the hyperscaler to purchase power or provide financing.
While the company’s S-1 implies that Fervo has agreed to restrict its ability to accept capital from a broad category of entities Google defines as competitors, Sarah Jewett, Fervo’s senior vice president of strategy, told Latitude Media that it isn’t an exclusivity agreement and that the company is already in talks with other hyperscalers. Time will tell how this unusual framework agreement plays out — and whether the restrictions it suggests inform Fervo’s growth.
Geothermal is more sensitive to interest rates than most clean energy technologies because, like offshore wind, the capital stack is heavily front-loaded, with drilling costs paid before a single MWh is generated. Fervo posted revenue of $138,000 in 2025 against a net loss of $70.5 million, and Phase 2 of Cape Station alone will require approximately $2.2 billion in capital expenditure through 2028.
Those numbers work at current rates. At rates two percentage points higher, however, the financing cost alone adds roughly $10 per MWh to the project’s levelized cost of energy. And unlike a solar project, where you can slow deployment while you wait for rates to drop, a partially drilled geothermal well is not a modular asset.
Political and technology risks
The GOP’s One Big Beautiful Bill preserved geothermal’s investment and production tax credits at full value through 2033, suggesting that the industry has far more bipartisan support than than wind and solar received. However, the offshore wind industry also had what looked like durable federal support in 2022, and the IRA’s wind and solar credits initially appeared structurally protected — before being effectively terminated within three years.
Geothermal’s current political insulation comes partly from its overlap with oil and gas, sharing drilling equipment, workforce, and investors, and from the personal history of Energy Secretary Chris Wright, whose company Liberty Energy invested $10 million in Fervo in 2022. That protection is real, but it is also contingent on geothermal remaining both small and politically convenient. As the industry potentially scales to tens of gigawatts of baseload capacity — and therefore starts to compete with gas — its friends may change.
Fervo’s technology is the integration of horizontal drilling, hydraulic fracturing, and fiber optic monitoring techniques adapted from the shale industry, and none of those techniques are proprietary in isolation. Eavor has raised more than $560 million for a closed-loop approach that avoids induced seismicity entirely, and Quaise is developing millimeter-wave drilling to reach depths and temperatures current enhanced geothermal cannot access. If the technology proves out at scale, the tools Fervo pioneered become the industry standard, available to every well-capitalized entrant that follows — and the valuation premium that comes with being first disappears.
There’s also the hurdle of scaling. Block Island Wind Farm came online in December 2016 as the first commercial offshore wind farm in the United States, a 30-MW project off the coast of Rhode Island that proved the technology could work in U.S. market conditions. It was a genuine milestone — and it was also almost entirely uninformative about whether 30 GW would work, because the success conditions at those two scales had almost nothing in common.
Block Island didn’t need a Jones Act-compliant installation fleet, a domestic supply chain for monopiles and turbine components, or a permitting environment that could survive a hostile administration. It just needed five turbines installed in relatively shallow water off the coast of a wind-friendly state.
Cape Station has the same problem; one 500-MW commercial-scale project in Utah, backed by hyperscaler PPAs and favorable geology doesn’t tell you whether enhanced geothermal can scale across less-ideal geographies — much less under a shifting policy environment.
Now, none of these risks is guaranteed to materialize; maybe none of them will. The chart above is not a forecast but rather a map of the assumptions the IPO is priced on, and a reminder that on the day of the NY Bight auction, a chart like it for offshore wind would have included a list of variables that looked equally manageable.
But these risks are not independent from each other, meaning that cascade risk is also a possibility.
A rate environment that squeezes geothermal’s financing cost would also make the cost curve harder to achieve, because it would make the capital required to drill the wells more expensive. A policy reversal that curtails the investment tax credit sooner than 2033 makes the financing cost worse. A hyperscaler that declines to convert its framework agreement into binding offtake could limit the revenue visibility that future project financing depends on.
Each assumption is load-bearing for the others. And that is exactly what made offshore wind so vulnerable at the moment it looked most secure
Two things can be true, and the tension between them is what makes this moment worth paying attention to. Geothermal has a real path to succeed on a long enough timeframe, and Fervo will likely play a meaningful role in that growth. At the same time, the commercial launch phase is also exactly where good technologies get derailed, not by the technology failing but by the broader market conditions moving in the wrong direction at the wrong time.
The right question isn’t whether geothermal works but whether the industry navigates this moment without repeating offshore wind’s mistake of treating good technology and visible momentum as sufficient on their own.
