Summary
Latitude Media reporting on Lawrence Berkeley National Laboratory (LBNL) annual “Queued Up” interconnection-queue analysis. Headline finding: 2,600 GW of total capacity sits in US interconnection queues — more than double the 1,279 GW of installed US capacity. ~11,600 active projects, 1,570 GW generation + 1,030 GW storage. Wait times have grown from under 2 years (2008) to 5 years (2023); only 14% of solar and 11% of battery projects requesting interconnection 2000-2018 were operational by end of 2023; ~70% of projects ultimately withdraw. Article published April 11, 2024; updated March 6, 2025. The LBNL “Queued Up: 2025 Edition” (Dec 2025) is the primary citable document for the most current data.
Key Points
- Headline ratio: 2,600 GW in queues vs. 1,279 GW US installed capacity — queue is roughly 2x installed.
- Active projects: ~11,600 seeking interconnection.
- Capacity composition: 1,570 GW generation + 1,030 GW storage.
- Resource mix: 95% solar/storage/wind; 3% gas; >80% solar+storage.
- Wait time evolution: 2008 <2 years → 2015 3 years → 2023 5 years.
- Completion rates (2000-2018 requests): 20% overall operational by end-2023; solar 14%; battery 11%.
- Project size growth (2015 → 2023): Solar +250%; standalone batteries +330%; wind +66%.
- Withdrawal rate: ~70% of projects ultimately withdraw; 1,250+ requests / 200+ GW withdrew in 2023 alone.
- Cost increase: Interconnection costs in 2019-2023 were 44% greater than the preceding 5 years.
Newsletter Angles
- The queue is twice the grid. This is the single most arresting framing for the Interconnection Queue story — easily quotable, immediately legible, structurally accurate. The article in development should lead Item 3 with this ratio.
- The 70% withdrawal rate is the structural failure mode. Projects don’t fail to clear the queue because of bureaucracy; they fail because the wait is longer than the financing horizon. Each withdrawal is a project that was viable when it entered and isn’t when it exits.
- The wait-time progression is a doubling/tripling story. From <2 years (2008) to 5 years (2023) — that’s a 2.5x increase over 15 years. The trajectory is widening, not narrowing.
- The 44% cost increase compounds the timeline problem. Even projects that survive the queue are paying substantially more to clear it. This is what “the queue is the document the market doesn’t price in” means concretely: interconnection costs are not in the AI capex projections.
Entities Mentioned
- Lawrence Berkeley National Laboratory (LBNL) — primary data source (deferred stub)
- Phoebe Skok — Latitude Media co-author
- Bianca Giacobone — Latitude Media co-author
Concepts Mentioned
- Interconnection Queue — directly thematic; this source is foundational
- AI Buildout Grid Constraint — queue as one of four sub-constraints
- Chokepoint Control — RTOs as legitimate-path chokepoints
Quotes
“Nearly 11,600 projects representing 1,570 gigawatts of generator capacity.” — LBNL “Queued Up” report (per Latitude Media)
“Completion rates are even lower for solar (14%) and battery (11%).” — LBNL findings
“Interconnection costs … were 44% greater than during the preceding 5 years.” — LBNL study
Notes
Latitude Media is energy trade press; treat as a high-quality summary of the LBNL “Queued Up” annual report (the primary citable document). The article was published in April 2024 with the 2024 LBNL data; the March 2025 update incorporates the 2025-edition data. The LBNL “Queued Up: 2025 Edition” (released Dec 15, 2025) extends this dataset further; that PDF should be ingested separately if the article needs the most current quarterly snapshot. The 2,600 GW vs. 1,279 GW ratio is the headline figure most cited; the 70% withdrawal rate is the structural-failure framing most cited. The article does not break out figures by RTO (PJM/ERCOT/MISO individually) — for that, the LBNL report itself or RTO-specific dashboards are needed.