733 GW Withdrew from U.S. Interconnection Queues in 2025

U.S. interconnection queues saw record withdrawals in 2024, but even those record-breaking numbers were eclipsed by the broad purge that occurred in 2025. 733 gigawatts of proposed generation capacity withdrew across 4,061 projects.¹

The shotgun approach to queue applications has become prohibitively costly since the widespread implementation of FERC Order 2023. Higher deposits and stricter site control requirements at earlier deadlines were implemented to disincentivize exactly this behavior. And from the 2024 and 2025 data, this approach appears to be working. MISO completed massive study cycles. PJM continued its transition. And across every region, the result was the same: projects that couldn't back up their applications were removed earlier than in previous years.

We're left with a leaner, more serious queue. But queue reform is far from the only factor that drove unexpected behavior in 2025. The One Big Beautiful Bill Act (OBBBA), supply chain constraints, and the era-defining data center buildout all influenced generation queues in unforeseen ways.

¹ Our 2024 retrospective counted 296 GW in withdrawals, but that figure only captured projects with explicitly recorded withdrawal dates, an estimated half to two-thirds of actual withdrawals. This year's methodology is more comprehensive (see Methodology section below). Even adjusting for the difference, 2025's total represents a massive escalation.

The Year in Withdrawals

Withdrawals arrived in waves, largely tied to tightly-bound cluster deadlines. December peaked at 131 GW across 775 projects, but every single month of 2025 saw meaningful exits from the queue. Q3 and Q4 saw massive withdrawn capacities. Q3 withdrawals were driven largely by MISO's DPP 2022 cluster (272 withdrawn projects totaling 48.4 GW) and DPP 2023 cluster (195 withdrawn projects totaling 36.4 GW).

The first half of the year saw 283 GW exit. The second half was significantly larger at 450 GW. Only December exceeded 100 GW, but March (93 GW), April (87 GW), and July (87 GW) all came close. Unlike a one-time purge, this was a sustained, rolling cleanup, largely the product of staggered decision point deadlines hitting across different power markets at different times.

The Net Impact: A Shrinking Queue

The queue absorbed 733 GW in withdrawals, a record-shattering number. 481 GW of new capacity entered the queue in 2025, and another 50 GW of existing projects reached commercial operation. But additions couldn't keep pace with withdrawals, and the net result is a meaningfully leaner queue.

The U.S. interconnection queue shrank from 2,273 GW at the end of 2024 to roughly 1,971 GW at the end of 2025, a 13% reduction and the second consecutive year of contraction. That reversal is historically significant. From 2019 to 2023, queues grew 20-40% annually as cheap application fees and low barriers to entry created a classic tragedy of the commons: developers flooded the shared queue with speculative projects, degrading study timelines and grid access for everyone. FERC Order 2023 imposed the guardrails that the commons lacked, and the spray-and-pray era appears to be ending.

What's Left: Technology Shifts

Not all technologies were impacted equally. Natural gas was the only main technology to see meaningful growth, while all other technologies contracted (excluding some growth of the "Other" category, which covers a long-tail of less-common technologies like nuclear and pumped storage).

Natural gas was the standout technology to increase its presence in the queue. Nationally, gas capacity rose from 183 GW to 255 GW, a 39% increase. The growth was concentrated in the interior: MISO surged from 30.6 GW to 71.6 GW (+134%), SPP from 14.9 GW to 35.3 GW (+137%), and ERCOT from 29.4 GW to 43.1 GW (+47%).

Every other major technology contracted. Solar fell 21% from 921 GW to 731 GW, a massive absolute decline but also a sign that the remaining solar pipeline is more commercially serious. Standalone storage dropped 18%, offshore wind collapsed 46% as project economics continued to deteriorate, and onshore wind fell 13%. The hybrid category (solar-plus-storage combinations) declined 17%.

The Long Wait: Queue Timelines

How long does it take to actually get through the queue? The answer in 2025 was: longer than ever. Projects that reached commercial operation last year waited a median of 5.9 years from queue entry, up from 5.1 years in 2024 and continuing a steady upward climb.

Queue reform was designed to fix this. But the projects completing in 2025 entered the queue around 2019 to 2020, before any of the current reforms took effect. Streamlined processes, cluster studies, and milestone-based screening may eventually reduce these timelines, but it will take years for those changes to show up in completion statistics. For now, the queue remains a long road.

What's Actually Getting Built

With so much attention on what's leaving the queue, it's worth asking: what's actually making it to the finish line? In 2025, at least 50 GW reached commercial operation across 400 projects, a tangible addition to the grid that gets lost in the headline withdrawal figures.

The most concrete indicator of near-term buildout is the pool of projects with signed Generator Interconnection Agreements (GIAs). These projects have cleared interconnection studies, negotiated contracts, and committed real capital. As of the end of 2025, roughly 488 GW of capacity holds signed GIAs, including about 19 GW in suspended status. ERCOT leads all markets with 133 GW, followed by the West at 86 GW, CAISO at 65 GW, MISO at 60 GW, and PJM at 53 GW.

This is the pipeline that matters most. While early-stage queue volumes are noisy and volatile, inflated by speculation and deflated by reform, GIA-backed projects represent the best signal we have of what's actually coming online. The fact that this pool grew even as the broader queue shrank suggests that the projects surviving the 2025 cleanup are further along in development, better capitalized, and more likely to reach operation. The queue is smaller, but the buildable pipeline is healthier than the top-line numbers imply.

Looking Ahead: The OBBBA Deadline

The next inflection point is already on the calendar. The One Big Beautiful Bill Act (OBBBA), signed into law in August 2025, set July 4, 2026 as the construction commencement deadline for solar and wind projects seeking federal tax credits — the investment tax credit (ITC) and production tax credit (PTC) that underpin the economics of most renewable energy development. The IRS followed with Notice 2025-42, which eliminated the 5% safe harbor that had previously allowed developers to qualify by ordering equipment alone. Projects must now demonstrate physical work of a significant nature at the project site by the deadline.

The queue that emerges from 2025 is fundamentally different from the one that entered it. It's smaller: 1,971 GW versus 2,273 GW a year earlier, a 13% reduction. Speculative projects have been culled. The remaining pipeline is weighted toward developers with signed agreements, secured sites, and real capital at risk. The path from queue to operation still takes nearly six years on average, and it's too early to tell whether reforms will meaningfully shorten that timeline. But the projects in line today are more likely to complete the journey. The central question for 2026 is whether the projects that remain can build fast enough to meet the deadlines now bearing down on them.

Methodology

Withdrawal dates: We use the reported withdrawal date when available. When not reported, we attribute the withdrawal to the month when we first detected the status change in our monthly snapshots.

Scope: The 733 GW headline figure counts projects that transitioned out of the Active or Suspended pool during 2025. Projects already withdrawn at the start of 2025 are excluded.

Comparison to 2024: Our 2024 retrospective reported 296 GW withdrawn, but that figure only included projects with explicitly recorded withdrawal dates. We estimate that captured between 50% and 67% of actual 2024 withdrawals, implying a true 2024 total of roughly 440 to 590 GW. This year's 733 GW figure uses a more comprehensive methodology based on our daily historical dataset, falling back to observed status transition dates when explicit withdrawal dates are not available. The methodological improvement means that a year-over-year comparison against our 2024 report is not strictly apples-to-apples, but even adjusting for the difference, 2025 withdrawals significantly exceeded 2024.

Data sourced from interconnection.fyi's database of queue filings across ISOs and utilities nationwide.