By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel consensus is bearish on FRNW, citing extreme valuations, limited liquidity, and significant risks such as grid interconnection queues, policy changes, and interest rate fluctuations.
Risk: Grid interconnection queues and the physical inability of utilities to upgrade transmission capacity at hyperscaler speed, leading to potential earnings misses and multiple compression.
Opportunity: None identified.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
- FRNW gained 33% YTD, tripling SPY's 11% return, a performance driven by AI data center power demand rather than traditional clean energy tailwinds.
- GEV booked $18.3B in Q1 orders, up 71% organically, including $2.4B in data center electrification that already exceeds the total for all of 2025.
- GEV now trades at 98x EV/EBITDA and ORA at 70x earnings, meaning buyers today are paying for the data center thesis rather than clean energy.
- It sounds nuts, but SoFi is giving new active invest users up to $1,000 in stock for a limited time, and all it takes is a $50 deposit to get started. See for yourself (Sponsor)
A $10,000 position in Fidelity Clean Energy ETF (NYSEARCA:FRNW) on the last trading day of 2025 was worth about $13,330 by the close on June 4, 2026, a 33% run in a little over five months. The same money in SPY would have grown to about $11,100, an 11% return over the identical window. FRNW has roughly tripled the S&P 500 year to date, which is the kind of headline that draws a crowd, and almost none of it is happening for the reason the fund's name suggests.
FRNW is a small, plain-vanilla index ETF from Fidelity's Covington Trust lineup, with a 0.39% net expense ratio and just $63.75 million in net assets as of the March 31, 2026 NPORT filing. The fund opened the year at $20.30 and closed June 4 at $27.06. Over the trailing twelve months the move is even more striking, about 84% versus the S&P 500's about 27%. Five-year returns tell a more honest story about the category, with FRNW up just about 11% since October 2021 while the S&P 500 returned about 79% over the same horizon. Clean energy has been the place capital went to die for most of this decade. Something changed.
The change driving the run is electricity demand from data centers, and FRNW happens to own the picks-and-shovels names that sell into that demand.
FRNW's biggest U.S. equity position is GE Vernova (NYSE:GEV) at 4.42% of net assets, and GEV is up 47.59% year to date. GE Vernova's Q1 2026 report is the document that explains the ETF. The company booked $18.3 billion in orders, up 71% organically, and called out $2.4 billion of Electrification equipment orders for data centers in a single quarter, which exceeded all of 2025. CEO Scott Strazik told investors that "demand is accelerating for our Power and Electrification solutions from a diverse set of customers, with our backlog growing by more than $13 billion quarter-over-quarter." Management raised 2026 revenue guidance to $44.5 billion to $45.5 billion and free cash flow to $6.5 billion to $7.5 billion. The stock trades at 28x trailing earnings and 34x forward, with analysts carrying a consensus target of $1,216.
Ormat Technologies (NYSE:ORA), a 2.67% FRNW position, is up 29.21% YTD on the same thesis dressed in different clothes. Ormat sells geothermal baseload, which is exactly what a hyperscaler needs when its data center wants 24/7 carbon-free electrons and a 15-year price lock. In Q1 the company posted revenue of $403.9 million, up 75.8% year over year, with adjusted EPS of $1.30 against a $0.90 consensus. CEO Doron Blachar tied the run directly to the AI buildout, citing "a 15-year portfolio PPA of up to 150MW to supply Google's data center electricity needs through NV Energy, and a 20-year agreement with Switch for approximately 13MW from the Salt Wells power plant." The stock now trades at a steep 70x earnings, which is the part of the story Wall Street is still arguing about.
First Solar (NASDAQ:FSLR) sits at 4.14% of the fund and is up 20.56% YTD, with most of that move concentrated in the past month. The thesis here is policy more than physics. First Solar's Q1 2026 guidance assumes $2.10 billion to $2.19 billion of Section 45X manufacturing tax credits this year, all of which flow to the bottom line. CEO Mark Widmar framed the competitive position around "a domestic manufacturing footprint, and independence from Chinese crystalline silicon supply chains." Q1 revenue of $1.04 billion came with a 50% adjusted EBITDA margin, which is what happens when Washington pays you to build solar panels in Louisiana.
SoFi Active Invest is offering a limited-time promotion. Open an account, fund it with $50 or more, and you could receive up to $1,000 in complimentary stock for Active Invest accounts. **See for yourself by clicking here now. **
And then there is the lottery ticket. Plug Power (NASDAQ:PLUG) is a tiny 1.52% of the fund but is up 82.74% YTD off a depressed base, and up 287% over the trailing year. Q1 GAAP gross margin came in at negative 13%, which sounds terrible until you remember it was negative 55% a year earlier. New CEO Jose Luis Crespo is targeting positive EBITDAS in Q4 2026. Reddit's wallstreetbets crowd noticed in May, with sentiment scores running 82 to 88 on a short-squeeze thesis. PLUG still burns ~$150 million of operating cash a quarter, which is the part the squeeze chatter tends to leave out.
What ties these four names together is one chart from the EIA. The Annual Energy Outlook 2026 notes that after 15 years of nearly flat U.S. electricity consumption, demand has increased by 2.1% per year, on average, over the last five years, with data center server energy projected to grow more than 16 times its 2020 level by 2050 in the high-demand case. Hyperscalers are signing 15- and 20-year contracts for electrons today because they cannot wait three years for a grid interconnect tomorrow. FRNW happens to own grid equipment makers, geothermal developers, domestic solar manufacturers, and a hydrogen call option. The fund's index construction got lucky on what would matter in 2026.
The international portion of the book has not been the story. Vestas Wind Systems is FRNW's largest position at 5.15%, with Ørsted, EDP, Acciona, and a handful of Chinese solar names rounding out the geography. Several of those have been dead money. The U.S. names did the lifting.
The conditions that produced this run are intact but no longer cheap. GE Vernova has tripled off the March 2024 spinoff price of $130.77 and now trades at 98x EV/EBITDA, a multiple that requires the data center order book to keep growing at the pace it did in Q1. Ormat at 70x earnings needs each new hyperscaler PPA to ratify the multiple. First Solar's earnings depend on Section 45X surviving in its current form, and the One Big Beautiful Bill Act of 2025 leaves the credits intact through 2030 with a phase-out through 2033. That cliff is the single most important policy date in this fund.
Two macro variables matter most for what comes next. The 10-year Treasury sits at 4.49%, in the 95th percentile of its trailing 12-month range, and renewable developers are the most rate-sensitive cohort in the market because their projects are bond-like. A sustained move back toward the May 19 peak of 4.67% would chip away at the relief rally that started in late winter. The other variable is the hyperscaler capex line, which is the actual demand signal. Track GE Vernova's quarterly Electrification book-to-bill, currently ~2.5x, and Ormat's new PPA announcements. Those two data points are the leading indicators that drove the move and will drive the next one.
One last thing worth saying out loud. FRNW has $63.75 million in net assets and 13.9% sitting in cash as of the March filing. This is a small fund that benefited from owning the right names in a regime where AI capex turned into power capex. The mechanism is structural and durable. The valuations the mechanism produced are stretched. If you are walking in now, you are paying for the data center thesis, not the clean energy one, and the price tag reflects it.
Want Up To $1,000? SoFi Is Giving New Active Invest Users Free Stock
Looking to grow your money but unsure where to begin? SoFi Active Invest is offering a limited-time promotion—open an account, fund it with $50 or more, and you could receive up to $1,000 in complimentary stock for Active Invest accounts.
From $0 commission trading to fractional shares and automated investing, this app is designed to simplify investing for everyone, whether you’re just starting or already experienced. Its easy to sign up and secure your bonus.
Four leading AI models discuss this article
"Current valuations for grid-infrastructure providers like GEV have decoupled from fundamental earnings growth and are now purely speculative bets on sustained, infinite hyperscaler capex."
The thesis that FRNW is a 'clean energy' play is a misnomer; it is essentially a levered bet on the physical infrastructure of the AI grid. While GEV and ORA are seeing massive order book expansion, the valuation compression risk is extreme. Trading at 98x EV/EBITDA, GEV is priced for perfection, assuming zero execution friction in a supply chain already stretched by labor shortages and long lead times for high-voltage transformers. Investors are conflating 'essential demand' with 'guaranteed margin expansion.' If hyperscaler capital expenditure cycles decelerate due to AI ROI skepticism, these multiples will contract violently, regardless of the long-term energy deficit.
The structural energy deficit is so acute that these firms possess significant pricing power, potentially allowing them to maintain elevated margins even if broader AI capex growth slows.
"FRNW's outperformance reflects multiple expansion on an already-priced thesis, not discovery of a cheap opportunity—and valuations now require flawless execution and accelerating capex to justify entry."
The article conflates two separate theses and mistakes one for the other. Yes, data center power demand is real and structural—the EIA chart backs that. But FRNW's 84% run wasn't driven by owning the right thesis cheaply; it was driven by multiple expansion on names that already priced in the thesis. GEV at 98x EV/EBITDA and ORA at 70x earnings aren't cheap entry points into a secular trend—they're expensive bets that capex growth accelerates from here. The fund's $63.75M AUM and 13.9% cash position mean it's been a beneficiary of narrative momentum, not fundamental outperformance. First Solar's 50% EBITDA margin is entirely synthetic—it's a tax credit arbitrage, not operational excellence. If Section 45X phases out or Congress reprices it, that margin evaporates.
Data center power demand genuinely is accelerating at 16x by 2050 per EIA, and 15-20 year PPAs lock in decades of cash flows—these aren't sentiment trades, they're structural. The article may be right that valuations are stretched, but it's also possible the market is correctly pricing a multi-decade tailwind that justifies current multiples.
"FRNW investors today pay data-center multiples for names already up 47-82% YTD, leaving little margin if 10-year yields rise or 45X credits face revision."
FRNW's 33% YTD outperformance rests on data-center electrification orders at GEV and ORA, yet the ETF's $63.75M AUM and 13.9% cash drag already signal limited liquidity. GEV at 98x EV/EBITDA and ORA at 70x earnings embed 71% order growth continuing indefinitely while the 10-year Treasury at 4.49% raises financing costs for rate-sensitive projects. First Solar's 50% EBITDA margin hinges on Section 45X credits surviving intact through 2030. International holdings like Vestas remain dead money. The structural demand shift is real, but entry multiples now price perfection rather than the clean-energy thesis the fund name implies.
Hyperscaler capex could exceed even the EIA's high-demand case, sustaining GEV's 2.5x book-to-bill and Ormat's 15-20 year PPAs at current multiples without rate or policy disruption.
"The strongest risk to the FRNW thesis is that the rally has priced in an extended, high-cost data-center electrification cycle, leaving little room for error if rates rise or hyperscalers slow capex."
FRNW has jumped about 84% in the 12 months ending June 4, 2026, driven by a data-center electrification thesis rather than pure clean-energy demand. The fund's top holdings are concentrated in GE Vernova and Ormat, names trading at sky-high multiples (GEV ~98x EV/EBITDA, ORA ~70x earnings), so continued upside depends on a persistent, ramping hyperscaler capex cycle. The piece acknowledges policy and rate risk, but underplays two red flags: (1) FRNW is a tiny fund (~$63.8m net assets; ~14% cash) with outsized name concentration, amplifying idiosyncratic risk; (2) a shift in interest rates or a stumble in data-center demand could compress multiples faster than earnings grow.
The strongest counter: the rally is pricing in an unusually long, high-cost data-center cycle; even a small policy shift or uptick in rates could snap the multiple compression and wipe out perceived gains, especially with high concentration and a tiny fund size.
"The market ignores that grid interconnection backlogs will force a reality check on the revenue recognition timelines currently baked into GEV and ORA valuations."
Claude, your focus on Section 45X tax credit arbitrage misses the primary risk: grid interconnection queues. Even if subsidies hold, the bottleneck isn't just policy—it's the physical inability of utilities to upgrade transmission capacity at hyperscaler speed. GEV and ORA are being priced as if they operate in a frictionless regulatory environment. When projects stall in the multi-year permitting backlog, the revenue realization for these firms will lag their current aggressive valuation multiples, leading to inevitable earnings misses.
"Grid queues are real, but the bull case hinges on GEV/ORA solving them—not just suffering from them—which the current valuation may or may not reflect."
Gemini's grid interconnection queue argument is the most underexplored risk here. But it cuts both ways: if GEV and ORA are solving that bottleneck—not just benefiting from demand—their moats widen and justify premium multiples. The question isn't whether queues exist; it's whether these firms' execution actually accelerates project timelines versus merely capturing upside from a solved constraint. That distinction determines if we're pricing genius or just riding tailwinds.
"Current multiples embed execution perfection that regulatory and financing frictions make improbable."
Claude frames interconnection queues as a potential moat-widener, yet this ignores that GEV's 98x EV/EBITDA already capitalizes on frictionless delivery. Historical permitting lags average 4-7 years; even modest acceleration still collides with 4.49% Treasury yields raising project hurdle rates. Cash flows from 15-20 year PPAs arrive too late to justify today's multiples if utilities cannot clear the backlog faster than hyperscalers demand.
"FRNW's small size and heavy concentration in two names make it vulnerable to idiosyncratic project delays, potentially causing NAV losses and quicker multiples compression than earnings growth."
FRNW's concentration risk is underappreciated. Even if interconnection queues are real frictions, a tiny fund with ~63.8m AUM and ~14% cash can’t absorb a single large GEV/ORA project delay without nav and price impact. The argument hinges on macro grid bottlenecks; the market may punish idiosyncratic execution risk first, triggering faster multiple compression than earnings growth if a project stalls or a contractor hiccups.
The panel consensus is bearish on FRNW, citing extreme valuations, limited liquidity, and significant risks such as grid interconnection queues, policy changes, and interest rate fluctuations.
None identified.
Grid interconnection queues and the physical inability of utilities to upgrade transmission capacity at hyperscaler speed, leading to potential earnings misses and multiple compression.