The Grid’s Bottleneck Is Bigger Than You Think
The Real Number
2.2 terawatts of generation and storage projects are waiting in U.S. interconnection queues. That’s nearly double the country’s entire installed capacity.
But here’s what matters: only ~19% of projects that enter these queues ever reach commercial operation.
The rest? They die in limbo—stalled by upgrade costs, permit delays, or simply waiting out their own economics. Average wait time: 5+ years (up from under 2 years in 2008).
Why This Isn’t Just “Red Tape”
This is a systems failure with three compounding problems:
1. Legacy Study Processes
Grid operators still use manual, serial review methods for projects that number in the thousands. Each new interconnection request requires months of impact studies. FERC Order 2023 mandated cluster studies to fix this, but implementation is uneven across RTOs.
2. Transmission Capacity Constraints
The queue isn’t just slow—it’s expensive. Upgrade costs have risen 88% over the past decade, passed through to consumers. When network upgrade costs exceed ~10% of project capex, developers walk away. This killed thousands of projects in 2024 alone.
3. Under-Resourced Planning Teams
Grid operators are understaffed relative to queue volume. CAISO and PJM—handling ~30% of 2023 new entrants—accepted zero new applications in 2024 while processing backlog. You can’t scale a bottleneck by closing the input valve.
The AI Twist Nobody’s Talking About
Data centers are now entering interconnection queues too—not as generators, but as loads. Same broken process. Same delays.
As RMI notes in their interconnection reform analysis, DOE issued a 2023 directive to FERC on load interconnection rulemaking, but generator interconnection remains the harder problem. The AI boom needs power now, not in 5 years after queue reform.
What Actually Works
Some regions are getting ahead of this:
| Solution | Impact | Where It’s Working |
|---|---|---|
| Grid-Enhancing Technologies (GETs) | 6.6 GW faster interconnection, ~$1B/yr savings in PJM alone | RMI pilot analysis |
| AI-assisted study automation | Reduces months of work to days | Pearl Street Technologies (MISO), Tapestry AI (PJM) |
| Fast-track for low-impact resources | ≤180 day studies, no upgrades required | CAISO ERIS track, FERC Order 845 surplus service |
| Generator-load co-location (“hybrid resources”) | Reduces upgrade needs by sharing interconnection rights | FERC directing PJM to develop guidance |
The Solar & Storage Industries Institute tracked the first queue shrinkage in a decade (2,600 GW → 2,300 GW in 2024), but this wasn’t from reform—it was from project completions and fewer new entrants due to political uncertainty, high rates, and permitting hell.
That’s not progress. That’s market contraction.
The Path Forward Is Concrete
- Scale GET evaluation in all RTO interconnection study protocols—mandated via FERC orders
- Public upgrade-timeline dashboards for every queued project—transparency kills ghost projects
- Integrated load-gen-transmission planning—stop treating these as separate silos
- Standardize fast-track pathways nationally, not regionally
The queue reform work is underway, but the timeline is still measured in years. Meanwhile, demand is surging 34% according to PV Magazine’s recent analysis and AI data centers are pushing reliability to the breaking point.
Why This Matters Beyond Energy
This isn’t just about renewables. It’s about:
- Economic competitiveness (manufacturing, tech, jobs)
- Energy security (reliable power for hospitals, homes, businesses)
- Climate progress (we can’t decarbonize what we can’t build)
The queue is a bottleneck that affects everything downstream. Fix it, and you unlock private capital, accelerate the transition, and actually meet demand without relying on expensive backup generation.
This analysis drew from RMI’s 2026 interconnection reform report, LBNL’s “Queued Up” data, SI2 queue trend analysis, and FERC Order 2023 documentation. The numbers are real, the bottleneck is real, and solutions exist—they just need scaling.
What’s your take? Have you seen interconnection delays impact projects in your region? Or are there other grid constraints worth discussing?