Large generation announcements can make electricity supply look like a simple construction problem: propose enough power plants, then build them. A project can carry a gigawatt figure while it is still asking for permission to connect. Between a proposal and commercial operation sits an interconnection process that can require studies, new equipment, network upgrades, cost allocation, permitting, and years of work.
Electricity supply is limited by the rate at which projects can connect to a planned, permitted, equipped, and operated grid.
Here, the grid means the networks and equipment used to deliver electricity, together with the system operators that decide how proposed plants may connect. Interconnection is the process through which a proposed project obtains an electrically workable path into that system.
From proposal to connection
Lawrence Berkeley National Laboratory’s Queued Up describes the formal entry point in the United States. Transmission operators require proposed power plants seeking a grid connection to undergo impact studies before the plants can be built. Those studies establish which transmission equipment or upgrades may be required and assign their costs. The lists of projects undergoing this process are called interconnection queues.
A useful schematic sequence is: request, impact studies, identification and costing of required network work, connection, and commercial operation. Operators need not use identical stages or timelines.
The distinction between stages is essential because a proposal’s requested capacity becomes visible before its power becomes available. Requested capacity is evidence of demand for grid access; later milestones provide stronger evidence that the capacity may reach operation.
What queue volume measures
At the end of 2025, about 8,200 projects were actively seeking grid interconnection in the United States, representing 1,312 GW of generation and approximately 749 GW of storage. The scale is large, but Berkeley Lab reports that most projects applying for interconnection are eventually withdrawn. Projects that are built have also been taking longer on average to complete their required studies and begin operating.
The historical outcomes give the queue a useful denominator. Of the capacity that submitted interconnection requests from 2000 through 2020, 13% had reached commercial operation by the end of 2025, while 75% had been withdrawn and 10% remained active. Berkeley Lab describes queue data as a general indicator of future capacity additions. The withdrawal record shows why the full queue cannot be carried directly into a forecast of operating supply.
Queue totals also combine technologies at different stages. The 2025 active queue included generation and storage, so a headline total covering both categories does not describe generation capacity alone.
The gap after an agreement
Progress within the queue provides more information than entry alone, but it still leaves a gap to operation. Berkeley Lab reports that 549 GW already had a draft or executed interconnection agreement at the end of 2025 without having reached commercial operation. That total included 256 GW of solar, 161 GW of storage, 76 GW of wind, and 45 GW of natural gas.
An interconnection agreement and commercial operation are separate milestones. An agreement can mark substantial progress while construction, upgrades, or other remaining work still stand between the project and available supply. Berkeley Lab does not divide the 549 GW by the specific reason each project had yet to begin operating.
For projects built in 2025 in regions with available data, the median duration from interconnection request to commercial operation exceeded five years. That measure covers the full interval rather than attributing the delay to one step.
Network buildout runs on a longer clock
The same timing problem appeared globally in the International Energy Agency's 2023 grid report. The IEA reported that at least 3,000 GW of renewable power projects were waiting in grid connection queues, including 1,500 GW in advanced stages. The full amount was equivalent to five times the solar and wind capacity added during 2022.
That estimate is incomplete by construction. Queue data were accessible for countries accounting for half of global wind and solar capacity, and the agency says the worldwide total is therefore likely higher.
The physical network develops more slowly than many projects seeking to use it. According to the agency, new grid infrastructure often requires five to 15 years to plan, permit, and complete. New renewable projects typically require one to five years, while new electric-vehicle charging infrastructure can require less than two years. The timing mismatch allows connection requests and electricity demand to accumulate faster than the supporting network can be expanded.
The required buildout is also large relative to the existing system. The 2023 report estimated that meeting national goals would require adding or refurbishing more than 80 million kilometres of grids by 2040, an amount equivalent to the entire existing global grid. It reported that global grid investment had remained around USD 300 billion per year while renewable investment increased rapidly, and estimated that grid investment would need to nearly double to more than USD 600 billion per year by 2030 to meet national climate targets.
Four parts of the constraint
The two reports describe four parts of the bottleneck:
- Planning and permission. Grid infrastructure must be planned and permitted, a process included in the agency’s five-to-15-year development range.
- Physical equipment. Interconnection studies identify the transmission equipment and upgrades required before a project can connect.
- Cost assignment and investment. Operators assign upgrade costs through the interconnection process, while the broader grid requires substantially greater investment.
- System operation. Connecting capacity does not remove the need to operate the system through changing patterns of output. In a scenario consistent with national climate goals, the agency estimates that required system flexibility doubles between 2022 and 2030 as variable renewable generation expands.
The four-part decomposition organizes the reported mechanisms; neither source measures it as one composite variable. It helps locate the relevant constraint when a large capacity announcement appears.
Reading a capacity proposal
A proposed-capacity figure becomes more informative when paired with its position in the connection process. Useful questions include:
- Is the capacity represented by an initial request, completed studies, an interconnection agreement, or commercial operation?
- Which transmission equipment or upgrades have the impact studies identified?
- Have the costs of that work been assigned?
- Does the total combine generation with storage?
- How much capacity from comparable requests has historically reached operation?
- Does the surrounding grid have the planning, permitting, investment, equipment, and operating flexibility required to accept the project?
Grid constraints can then propagate into product systems. Fertilizer, Nitrogen, and Food Security follows one chain from energy inputs through industrial production, trade, farm use, and crop risk.