The Permitting Clock

The problem

Nine proposals in this catalog require somebody to build something. Transmission lines (№ 024), housing (№ 032), water infrastructure (№ 034), mines (№ 037), reactors (№ 044), ports, fabs, and the rural buildout of № 036. Every one of them terminates in the same bottleneck, and it is worth treating that bottleneck as its own proposal rather than as an aside in nine others.

The National Environmental Policy Act was signed in 1970. It is a procedural statute: it does not forbid anything, it requires the government to study the environmental consequences of its actions and publish them before acting. That is a good idea and it remains a good idea.

The Council on Environmental Quality's own review found that the average environmental impact statement took about four and a half years to complete, with a median near three and a half, and a substantial tail running past a decade. The average document ran to roughly 600 pages, not counting appendices that routinely run to thousands.

Then the document is finished, and the litigation begins. A NEPA case does not usually argue that a project is harmful. It argues that the study of the project was inadequate, in which case the remedy is to do the study again. A challenge filed years after a decision can restart the clock on work already done.

The costs of this are not evenly distributed, and they are not what people assume.

Cape Wind spent 16 years in review and litigation and was never built. Vineyard Wind took roughly a decade. Interconnection queues for new generation now hold something on the order of 2,600 gigawatts of proposed capacity, more than the entire installed capacity of the United States, with typical waits around five years and only a minority of projects ever emerging.

For contrast: in 2022, facing an energy emergency, Germany permitted and built a floating LNG import terminal in roughly 200 days.

Here is the part that should trouble everyone regardless of their priors. NEPA now applies most punishingly to the things that reduce emissions, because those are the things that need new construction. A solar farm, a transmission line and a battery plant all need permits. An existing coal plant already has its permits and needs nothing. The statute has become asymmetrically hostile to building the replacement.

The proposal

One lead agency, one document, a hard deadline, and a short window to sue. Not less review. The same review, finished on a date.

How it would work

1. A real clock. Two years for an environmental impact statement, one year for an environmental assessment, measured from initiation to record of decision. The 2023 amendments established these deadlines; the missing half is a consequence when they are missed. If the deadline passes without a decision, the applicant may seek an order from a court compelling a decision within 90 days.

1. A decision, not an approval. This is the essential distinction and this proposal will be misread without it. The clock forces the agency to answer, and no is a complete answer. Nothing here creates automatic approval by default, which would convert a procedural statute into a rubber stamp and would deserve to be struck down.

1. One lead agency, one document. A single agency designated for each project, with authority to set the schedule for all cooperating agencies and to produce one document that satisfies all of them. Projects have been known to require concurrent reviews from a dozen agencies, each with its own timeline and its own view.

1. Page limits that mean something. 150 pages, 300 for genuinely complex projects. A 600-page document is not more rigorous than a 150-page one; it is a document written to survive litigation rather than to inform a decision, which is a description of a defensive artefact rather than an analysis.

1. A short statute of limitations. 150 days from the final decision to file, hard. A project that has cleared review and begun construction should not face a challenge in year four. This shortens the period of legal uncertainty without narrowing the grounds on which a challenge may be brought.

1. Remand without vacatur as the default remedy. Where a court finds a review inadequate, the ordinary remedy is to order the defect fixed, not to void the permit and halt the work. Vacatur remains available where the defect is fundamental. Today the threat of vacatur is what makes the 600-page document rational.

1. Staff it. Deadlines without reviewers produce either missed deadlines or bad reviews. Multi-year funding for permitting staff at the agencies that carry the load, and the ability to pay competitive salaries for the specific scientific and engineering skills that are scarce.

1. Categorical exclusions, expanded and honest. Reconductoring an existing transmission line, adding capacity within an existing footprint, and siting in a previously disturbed industrial area should carry categorical exclusions. The great majority of federal actions already proceed this way, and extending it to work that reuses existing corridors is the lowest-risk, highest-yield item on this list.

The numbers

Time is the number. A project financed at 8 percent that sits in review for an additional four years carries roughly 36 percent more in accumulated financing cost before it earns a dollar. That is often the difference between a project happening and not happening, and it is a cost paid entirely to nobody.

2,600 gigawatts sit in interconnection queues. Even a modest improvement in throughput dominates any generation subsidy in effect, because the binding constraint is not whether projects are wanted but whether they can be connected.

The comparison that should end the argument: the Interstate Highway System, 41,000 miles, was substantially built between 1956 and the early 1970s. It would not be permittable today on anything like that schedule, and the country that built it is now unable to run a wire from a wind farm in Iowa to a city in Illinois inside a decade.

The cost of item 7 is on the order of a few hundred million dollars a year in agency staffing, against projects worth hundreds of billions waiting on those staff. This is the cheapest item in this catalog and probably the highest-return.

The honest objections

"NEPA exists because government used to bulldoze neighbourhoods without asking. Speeding it up means going back to that." The most serious objection and it rests on real history: highways were routed through Black and poor neighbourhoods in city after city precisely because nobody was required to study or disclose the consequences, and NEPA and its state analogues are why that is harder now. Nothing above removes a substantive protection, narrows a ground for challenge, or reduces what must be studied and disclosed. It compresses the schedule and shortens the window for suit. But the concern deserves a specific safeguard rather than a reassurance: the consultation requirements for affected communities should be strengthened, not merely preserved, and the time saved from litigation defensiveness should be partly reinvested there.

"Shot clocks with fixed resources produce sloppy work or automatic denials." Correct, which is why item 7 is in the list rather than in a footnote, and why item 2 makes denial an available answer. An agency that cannot complete a review in two years with adequate staff will deny, and the applicant's remedy is to fix the project. That is a functioning system. The current system, where the applicant simply waits, is not.

"Most permitting delay is state and local, not federal." Frequently true, particularly for housing and transmission, and it is why № 032 and № 024 carry their own state-level mechanisms. Federal reform is necessary rather than sufficient. It is also the part Congress can actually do.

"Shortening the limitations period cuts off legitimate challenges from communities that lack the resources to organise quickly." A real and specific harm, and 150 days is genuinely tight for an under-resourced community that learns of a project late. The mitigation is that the clock runs from the final decision, which is the end of a public process running years, not from a surprise. If evidence shows the window is systematically excluding legitimate challenges, it should be lengthened. We would rather set it, measure it, and adjust than leave it indefinite.

"This is a giveaway to industry dressed up as climate policy." It benefits whoever wants to build, which includes pipelines and mines as well as transmission and solar. That is a genuine consequence and we are not going to claim the reform is selective, because a permitting regime that runs fast only for projects the administration favours is worse than a slow one that runs the same for everybody. The bet is that on current margins, a country that can build again builds more of what it now lacks, which is transmission, housing, clean firm generation and the physical basis for everything in this catalog.

Sources

• National Environmental Policy Act of 1969, 42 U.S.C. § 4321 et seq.
• Council on Environmental Quality, Environmental Impact Statement Timelines (average completion approximately 4.5 years; average document length approximately 600 pages) (whitehouse.gov/ceq)
• Fiscal Responsibility Act of 2023, NEPA amendments: two-year and one-year deadlines, page limits, lead agency designation
• Seven County Infrastructure Coalition v. Eagle County, 605 U.S. ___ (2025), on the scope of NEPA review and judicial deference
• Interconnection queue volumes and typical wait times; Lawrence Berkeley National Laboratory, Queued Up series (emp.lbl.gov)
• Cape Wind project history (2001–2017); Vineyard Wind permitting timeline
• Germany's Wilhelmshaven LNG terminal, permitted and commissioned in approximately 200 days under the LNG Acceleration Act (2022)
• Proposals № 024 (The Grid Dividend); № 032 (The Housing Dividend); № 034 (The Water Compact); № 036 (The Rural Compact); № 037 (The Hardrock Royalty); № 044 (The Atomic Compact)