A $10,000 position in the iShares Global Clean Energy ETF (NASDAQ:ICLN) on the last trading day of 2025 is worth about $14,455 as of Tuesday’s close, with shares moving from $16 to $24 in five months. The S&P 500 did fine over the same window. The SPDR S&P 500 ETF Trust (NYSEARCA:SPY) returned 11% year to date. ICLN’s 45% gain is just shy of four times that, which is the quadrupling the headline is pointing at.
If you owned this fund any time between 2021 and late 2024, that sentence reads like a typo. ICLN was the textbook example of a thematic ETF that got bought at the wrong end of a regime, held through a brutal multi-year unwind, and quietly sold for tax losses by people who had given up. Over the trailing five years, the fund has returned 15% in total, against 81% for SPY. The one-year number tells a different story. ICLN is up 92% over the trailing twelve months, against 28% for SPY. Something changed.
What Actually Did the Work
The clean energy trade of 2020 and 2021 was a rates and subsidies story. Cheap money discounted long-duration cash flows generously, and the Inflation Reduction Act promised a wall of federal support. When rates went up and the IRA’s edges started getting renegotiated, the trade collapsed. The 2025 and 2026 revival is built on something different, which is why it has held. The new buyer needs electrons.
U.S. data center electricity demand has gone from a rounding error to the dominant variable in utility resource planning. Data centers grew from 1.9% of total annual U.S. electricity consumption in 2018 to 4.4% in 2023, and the Department of Energy now projects data centers will account for up to 12% of U.S. electrical demand by 2028. An individual hyperscale facility can pull over a gigawatt of power, equivalent to powering approximately 750,000 homes. You cannot meet that load on natural gas turbines alone fast enough, and the people writing the checks know it. The EIA’s own base case has the combined generation share of natural gas, wind, and solar rising from about 60% in 2025 to around 80% in most cases by 2050.
That reframing, from policy bet to infrastructure bet, is what news flow has been chewing on for six months. Industry coverage in December described the shift as the "investment narrative shifted from government subsidies to critical infrastructure" and called out "massive power demands from technology hyperscalers" as the catalyst. A January piece framed ICLN as having moved from a "speculative growth-at-any-cost phase to infrastructure-heavy focus". Inside the fund, the result has been concentrated. Bloom Energy alone was reported to have contributed a 435% surge to ICLN’s gains, and the top holding, China Yangtze Power, comprises around 6% of the portfolio, which gets you exposure to the operator of the Three Gorges Dam inside a fund most U.S. buyers think of as solar and wind.
The Tax-Credit Cliff Sitting in July
There is a second mechanism layered on top, and it has a date attached. Industry trade press has been writing for months about a "construction cliff" requiring projects to begin by July 4, 2026, to qualify for full tax incentives. Developers do not let free money expire. The order books for solar modules, inverters, electrolyzers, and balance-of-system equipment have been pulled forward into the first half of 2026 because they had to be. The same article noted ICLN delivered a +47% return in 2025 and was already up 8% YTD in early January 2026 on this dynamic alone.
Institutional positioning has been consistent with a real bid rather than a meme. Florin Court Capital made ICLN its largest holding at 13% of AUM last November, and even after trimming 199,800 shares in May 2026, ICLN remained its largest holding at 18.5% of assets. That is the profile of a manager taking some chips off the table after an 80% rally. Other funds have sold entire positions on the same logic. Perbak Capital exited a $6.33 million position in late December, framing it as "lock in gains and rebalance".
What a Repeat Would Require
This is where the article earns its keep. The setup that produced 44% in five months is not a forever setup, and pretending otherwise would be cheerleading. Three conditions have to keep holding for ICLN to keep working, and each is now visibly under pressure.
First, the AI power-demand story has to stay in the price. It probably does. The Department of Energy’s up to 12% of U.S. electrical demand by 2028 projection has not weakened, and EIA’s own May 2026 outlook revised utility-scale solar generation 1.4% higher than in the previous STEO. The watch item is hyperscaler capex guidance from the next two earnings cycles. If that flattens, the bid flattens.
Second, the tax-credit pull-forward is a one-shot. After July 4, 2026, developers who started construction get the credit, and developers who did not, do not. The bulge in equipment orders that has driven first-half earnings for ICLN’s manufacturers will not repeat in the second half. Expect a digestion phase. The honest read is that some portion of the 44% YTD gain was demand pulled out of late 2026 and 2027. The April 2026 index rebalancing flagged in trade coverage adds a second mechanical wildcard.
Third, the political backdrop has to not get worse. Trump’s Davos energy pitch in January already pressured clean energy ETFs and rotated capital into nuclear names, and $1.5 billion in outflows from clean energy funds showed up in early April even as ICLN’s price kept rising. Price up on net outflows is what happens when a few large buyers are absorbing supply from many small sellers, which works until it doesn’t.
The cleanest way to think about ICLN here is that the fund stopped being a rates trade and became a power-grid trade, and the power-grid trade is real. The fund still carries a low 0.39% expense ratio and a global mandate, which is part of why a single Chinese hydropower operator can sit at the top of the book. Prediction sentiment composite reads 62, bullish with medium confidence, which is consistent with a market that believes the story and is aware of the July date. What to watch from here is straightforward. Hyperscaler capex guidance, the post-July 4 order book at the fund’s solar and inverter manufacturers, and whether the next round of clean energy fund flows turns positive or keeps bleeding while price drifts. The mechanism that produced the run is the right mechanism to watch for the run to keep going, or to end.