Queue Analysis: America's Interconnection Crisis
Nearly 2× the power that currently exists on the US grid is waiting to connect. An interactive analysis.
Queue by Technology
Active capacity in US interconnection queues as of end of 2024: LBNL Queued Up 2025 Edition
Technology Mix (Donut)
Active Queue by Technology (GW)
Key insight: Solar and storage together make up ~81% of the queue, driven by the Inflation Reduction Act. Gas capacity surged 72% year-over-year to 136 GW, reflecting new data center and AI demand — and that was before Texas's 80.6 GW gas development pipeline for DCs and the SoftBank/AEP 9.2 GW Ohio announcement in March 2026. Meanwhile, solar dropped 12%, wind dropped 26%, and storage fell 13%, reflecting historic withdrawal rates and a 12% overall decrease in active queue volume per LBNL.
Queue by RTO/ISO
Queue size and historical completion rates by grid operator
Queue Capacity vs. Completion Rate by RTO
The West dominates. Non-ISO Western regions hold 706 GW, the largest queue in the country. CAISO follows at 523 GW. PJM's queue is smaller but among the most congested, with some of the lowest completion rates. ERCOT processes requests fastest but has less stringent study requirements.
The Wait
Average years from interconnection request to commercial operation, it's getting worse
Wait Time Trend (Years)
Wait Time by Technology (Years)
Estimated Wait Time by RTO (Years)
A decade of deterioration. The median project built in 2018–2024 waited over 4 years from interconnection request to commercial operations, up from 3 years in 2015 and less than 2 years in 2008 (LBNL). Wind projects wait the longest (~5.5 years) due to siting and transmission constraints. Solar has shortened slightly thanks to distributed siting, but storage waits are climbing fast.
The Dropout
Of all projects that entered the queue from 2000–2019, only 13% reached commercial operation
The Interconnection Funnel
Completion Rate by Technology
Project Outcomes (2000–2019)
87% of capacity never gets built. Gas projects have the best completion rate at 31%, followed by wind (20%), solar (13%), and batteries (11%). Most projects withdraw during the study phase when they discover upgrade costs they can't afford, often hundreds of millions of dollars allocated to a single project.
The Cost of Waiting
Queue delays don't just waste time, they waste billions
Study Deposits
Per project for interconnection studies. Deposits have increased 3–5× since FERC Order 2023 to deter speculative applications. Large projects can face $10M+ in study costs alone.
Network Upgrades
Average upgrade allocation for later-queued projects. "Cost cascade" effects mean one project's withdrawal can trigger $50–200M in cost reallocation to remaining projects.
Lost Generation
Estimated clean energy generation delayed or lost annually due to queue backlogs. That's roughly 15% of total US electricity generation, enough to power 55 million homes.
Consumer Impact
Higher electricity costs from delayed competition, continued reliance on expensive peaker plants, and missed clean energy price reductions. Costs ultimately passed to ratepayers.
The hidden tax on the energy transition. When a 500 MW solar project waits 5 years instead of 2, that's ~2.4 TWh of clean energy never generated. Multiply across thousands of projects and the climate cost is staggering. Meanwhile, developers burn through millions in holding costs, studies, and legal fees, costs ultimately borne by consumers.
Reform Tracker
What's being done to fix the queue crisis, and what's actually working
FERC Order 2023 Active
- First-ready, first-served: Replaces serial first-come, first-served with cluster study approach
- Financial commitments: Higher deposits ($5M+ for large projects) and "readiness" milestones to deter speculative queue entries
- Study deadlines: Penalties for transmission providers that miss study timelines
- Compliance deadline: May 16, 2024 for all transmission providers
- Status: All RTOs have filed compliance plans. First cluster studies beginning 2025–2026. Full impact expected by 2027–2028.
RTO-Specific Reforms Mixed
- ERCOT: Already used cluster approach. Best processing times nationally. Added fast-track for smaller projects.
- CAISO: Cluster study process since 2010. Strong track record but overwhelmed by volume.
- PJM: Transition window clearing legacy backlog. New rules effective mid-2025. Slowest current processing.
- MISO: Queue cap proposal rejected by FERC. Working on tranche-based processing.
- SPP: Implemented zonal approach. Backlog clearing underway.
- ISO-NE / NYISO: Compliance filings accepted. Transitional studies in progress.
State Actions Emerging
- Texas (SB 1287): New transmission planning process. $4.6B in new lines approved for ERCOT.
- California: SB 887 allows "connect and manage" for some projects. CPUC proactive transmission planning.
- New York: Power Grid Study identifying transmission needs. $11B in grid investments planned.
- Virginia: Dominion has 70,000 MW of large-load requests in queue — nearly tripled since January 2025. Only 25,000 MW have set connection dates. GS-5 rate class now requires DCs to fund their own infrastructure. PJM's Expedited Interconnection Track (EIT) filed Feb 2026 — 10 fast-track slots per year for 250 MW+ projects with state siting approval.
- Multi-state: MISO's Long Range Transmission Plan ($10B+) creating new capacity.
What's Working? Too Early
- Higher deposits are reducing speculative applications, queue growth may be slowing
- Cluster studies process more projects simultaneously vs. serial approach
- Transition period is creating a "study gap" where few new projects are being studied
- Legacy backlog of ~2,600 GW still needs to be processed, will take years
- Transmission buildout remains the fundamental bottleneck. Queue reform without grid expansion only goes so far.
- Workforce shortages in grid engineering are limiting study capacity across all RTOs.
Interactive Queue Explorer
Filter and sort queue data by RTO, technology, and state
| RTO ⇅ | State ⇅ | Technology ⇅ | Capacity (GW) ⇅ | Est. Projects ⇅ | Completion % ⇅ | Avg Wait (yr) ⇅ | Queue Size |
|---|
Last updated: March 2026 · Based on LBNL 2025 Edition (end of 2024 data). ← Back to Queue Tracker · Download raw data →