Energy & Utilities
For a Decade, Power Was Assumed. It Has Become the Largest Constraint on Industrial Growth.
Only 13% of interconnection requests filed 2000-2019 reached commercial operation by 2024. Withdrawal, not connection, is the modal outcome — and utility relationship management is now a core commercial capability.
By Vaxa Market Intelligence Team
4 min Read
Published on Januay 4, 2026
Sector Snapshot — Energy Transition
01
~2 TW of generation and storage capacity currently stuck in U.S. interconnection queues — almost double the country's entire installed grid capacity
02
150+ GW in additional generation capacity needed within five years, by 2030, to meet current demand growth
03
13% of interconnection requests filed between 2000 and 2019 had actually reached commercial operation by the end of 2024
04
60% demand growth PJM, the largest U.S. grid operator, now projects over the next 20 years — after a decade of essentially flat load growth
Market Consensus & Vaxa's Position
Grid interconnection is a queue to get through — submit the request, wait your turn, and the project connects when its number comes up.
Getting into the queue and getting through it are two different processes with almost no relationship to each other. Only 13% of capacity that entered a queue between 2000 and 2019 had reached commercial operation by 2024. Withdrawal, not connection, is the modal outcome — "waiting your turn" is not actually how most of these projects resolve.
The Shift
For a decade, power was assumed. It has become the single largest constraint on industrial growth.
From roughly 2010 to 2020, U.S. electricity demand grew by less than half a percent a year — essentially flat. Utilities planned, priced, and staffed for a world of incremental growth. That world ended abruptly. Data center load, reshored manufacturing, EV production, and broader electrification are now arriving simultaneously, and PJM — the largest grid operator in the country, covering thirteen states and Washington, D.C. — now projects summer peak demand growing 2.4% a year over the next 20 years, a cumulative rise of roughly 60% and the steepest sustained curve the operator has forecast in years.
The consequence is structural, not cyclical. Any company whose growth plan depends on new or expanded power-intensive capacity — a factory, a fabrication plant, a distribution center, a data center — is now negotiating with a utility and a grid operator before it is negotiating with a contractor or a landlord. Power access has become the first constraint in the sequence, not a checklist item confirmed after the site is chosen.
Two industrial manufacturers each file for grid interconnection the same month — same regional utility, same 40-megawatt load, same expansion timeline tied to a customer contract eighteen months out.
The first treats the filing as the finish line. It submits the request, assumes the queue will process in the order received, and waits. Eighteen months later, the interconnection study alone hasn't cleared — the same fate that 77% of requests filed in this era ultimately meet. The customer contract's delivery date passes with no power to run the new line, and the expansion quietly disappears from the next budget cycle.
The second treats the filing as the opening move in a negotiation, not a request to be granted. It engages the utility directly on cost-sharing for the grid upgrades its own load will trigger, offers demand-response flexibility and real-time telemetry in exchange for priority sequencing, and structures a behind-the-meter backup as a hedge against queue delay. The utility, now negotiating terms rather than simply processing an application, moves the project onto its fast-track path. The expansion goes live on schedule.
Same utility. Same month. Same load. One company waited in a queue that fails seven times out of eight. The other negotiated — because it understood the queue had already stopped being a queue.
The Queue Problem
Getting into the interconnection queue is easy. Getting through it is not.
Nearly 8,200 projects are currently seeking grid interconnection in the U.S., representing roughly 1,312 gigawatts of generation and 749 gigawatts of storage — collectively, capacity that dwarfs what's actually installed today. The process to connect a new generation source to the grid, or a new large load to draw from it, was not built for volume at this scale. Interconnection wait times have more than doubled over the past fifteen years, and projects now spend an average of five years in queue before reaching commercial operation, if they reach it at all.
Of all capacity that submitted interconnection requests between 2000 and 2019, only 13% had actually reached commercial operation by the end of 2024 — per Lawrence Berkeley National Laboratory's "Queued Up" tracking, the standard reference dataset for U.S. interconnection queues. The rest — 77% — was withdrawn outright, with the remainder still stuck in various stages of review. Nearly 80% of withdrawn projects cite unpredictable, multi-year delays and prohibitive grid upgrade costs, which for withdrawn projects can consume 30-37% of the total project budget on their own. PJM's backlog alone exceeded 300 gigawatts of projects before the operator reopened its queue in 2026 after a multi-year pause — a freeze significant enough that industry groups said it had effectively stalled large volumes of new generation development nationally.
The cost of this dysfunction is not abstract. Analysis commissioned by GridLab found that if just 10% of the renewable capacity sitting in PJM's pre-2024 queue had been built in time for the 2026-2027 capacity auction, it would have saved PJM consumers an estimated $3.5 billion. Delays in transmission investment more broadly translate to $150-370 million in lost net benefits annually for every $1 billion of delayed investment.
.png)
The New Negotiation
Utilities are starting to negotiate terms with large loads, not just connect them.
Faced with this volume, utilities and regulators are beginning to change the terms of engagement rather than simply working the backlog faster. In Indiana, rate-case settlements now require hyperscalers to directly pay for the grid upgrades their projects trigger. Oregon has passed legislation creating an entirely separate utility rate class specifically for large data center loads. ERCOT now requires large-load projects to align their approvals with ramp-up plans and real-time telemetry, and is building dynamic tariffs that could expose flexible large loads to granular, real-time price signals rather than fixed rates.
PJM's fast-track interconnection process now explicitly favors "shovel-ready" projects, and 2026 is expected to bring performance-based interconnection more broadly — tying queue priority to a project's demonstrated flexibility and telemetry, not just when its application was filed. The practical effect is that the utility relationship has shifted from a service being requested to a negotiation being conducted, with cost-sharing, operational flexibility, and demand-response commitments now standard terms rather than exceptions.
This is compounding an already tight labor market. The utility workforce is aging into a wave of retirements at the exact moment demand for engineering, construction, and systems operations expertise is spiking — a talent bottleneck sitting directly beneath the capital bottleneck everyone is focused on.
What This Means for Strategy
Utility relationship management is now a core commercial capability, not a peripheral one.
For any company planning power-intensive growth — not just hyperscalers, but manufacturers, EV producers, and any reshoring effort dependent on new domestic capacity — the interconnection queue is no longer background infrastructure risk. It is a primary planning variable that has to be assessed before site selection, capital allocation, or timeline commitments are finalized. Treating a projected commercial operation date as a bankable milestone, when the data shows only 13% of historical requests actually reached that milestone, is a planning error with direct financial consequences.
The organizations adapting fastest are building utility and grid relationships as a dedicated capability — tracking interconnection queue positions, transmission upgrade costs, and state-level rate case activity the way they would track any other market intelligence, and structuring offtake agreements and cost-sharing arrangements to reflect delay risk rather than assuming it away. Behind-the-meter generation and flexible interconnection agreements, once niche options, are increasingly the default path for any project that cannot tolerate a five-year queue wait.
Two industrial manufacturers each file for grid interconnection the same month — same regional utility, same 40-megawatt load, same expansion timeline tied to a customer contract eighteen months out.
The first treats the filing as the finish line. It submits the request, assumes the queue will process in the order received, and waits. Eighteen months later, the interconnection study alone hasn't cleared — the same fate that 77% of requests filed in this era ultimately meet. The customer contract's delivery date passes with no power to run the new line, and the expansion quietly disappears from the next budget cycle.
The second treats the filing as the opening move in a negotiation, not a request to be granted. It engages the utility directly on cost-sharing for the grid upgrades its own load will trigger, offers demand-response flexibility and real-time telemetry in exchange for priority sequencing, and structures a behind-the-meter backup as a hedge against queue delay. The utility, now negotiating terms rather than simply processing an application, moves the project onto its fast-track path. The expansion goes live on schedule.
Same utility. Same month. Same load. One company waited in a queue that fails seven times out of eight. The other negotiated — because it understood the queue had already stopped being a queue.
DIAGNOSTICS CHECKLIST
Bringint it together
01
Only 13% of interconnection requests filed 2000-2019 had reached commercial operation by 2024 — most growth plans that depend on new power capacity are underwriting against a milestone that fails to hold roughly seven times out of eight.
02
PJM's demand curve has inverted from essentially flat (sub-1% annual growth, 2010-2020) to a forecast 60% cumulative rise over the next 20 years — the planning assumptions built during the flat decade no longer apply.
03
Withdrawal, not connection, is the modal outcome for a queued project — treating "submitted a request" as equivalent to "capacity is coming" misreads what the data actually shows.
04
The dysfunction has a real, measured cost: GridLab's analysis puts $3.5 billion in avoidable PJM consumer costs on just a 10% improvement in queue-to-build conversion for one capacity auction cycle alone.
05