💬 Discussion
Two Greek thinkers examine an electricity pylon, pipeline segment, permit scrolls and an hourglass.

Federal Permitting Reform

Federal environmental reviews of large projects take years, and a lawsuit can halt a project that has already been approved. On 30 September 2026 the Republicans Shelley Moore Capito and Mike Lee and the Democrats Sheldon Whitehouse and Martin Heinrich introduced S. 5653 to change that [4][6]. Oil and gas groups and much of the clean energy industry back the bill, environmental groups oppose it, and a floor vote is planned for after the November elections [16].

  • 4.2 years pass on average between a federal environmental review and the end of the appeal against it [22]
  • 28 percent of large energy and transport projects that needed a full review were sued before construction began [21]
  • 4.3 billion dollars of lines to connect data centers were approved in seven states of the largest grid region in 2024, mostly charged to every customer [20]

The proposal

Federal reviews of energy and infrastructure projects get deadlines of one to two years and lawsuits a limit of 150 days, while regions must plan power lines across their borders.

The debate

The sponsors and the energy industry argue that building lines, pipelines and plants takes too long and costs too much, and that data centers should pay for their grid links. Clean energy groups and a center-left think tank add that the rules for transmission matter more for emissions than anything the bill does for gas. Environmental groups and legal scholars answer that courts could no longer stop a project built on a flawed review and that pipelines and mines gain most. The Senate has not voted on the bill.

  • “That agreement is the product of many, many months of negotiations and years of work to address a problem Congress has talked about for far too long: It takes too much time and too much money to build things in this country, and that problem is becoming more urgent by the day.”

    argues for: Mike Lee, Republican, Utah Senate floor, 30 September 2026

  • “They don't get that preserving the status quo in permitting isn’t acceptable if you actually care about deploying clean energy and reducing emissions.”

    argues for: Josh Freed, Third Way Press release, 5 October 2026

  • “This language, if enacted, effectively guts the core purpose and effect of NEPA.”

    argues against: Deborah Sivas, Stanford Law School Stanford Law School blog, 4 October 2026

The analysis

Federal Permitting Reform

68% of the weight is for, 32% against.

An argument's weight combines three questions: how big the effect is, how certain it is and how much the good at stake matters. The larger the slice, the more the argument weighs.

For

Against

The federal permitting bill is a good proposal on the current analysis, and the reason is the grid.

Very likely better

Very likely better: Under the assumptions used, the proposal brings more than it costs in 99 of 100 model runs.

 Overall effect: large

How strongly the proposal works overall, in five steps from very small to very large, comparable across all debates. The ring is always full, even when the overall effect is small; this step says how large it is.

Like its sponsors, the analysis comes out for the bill, for a reason heard mainly from clean energy groups: regions must plan lines to each other and connect plants faster. That grid reform cuts carbon dioxide and running costs, and those two items together outweigh everything against the bill. The largest sum on both sides is the same money, the connection costs that data centers would pay instead of other customers, and it almost cancels. The strongest argument against, damage from projects built on a flawed review, is the one environmental lawyers press hardest, yet it weighs far less than the grid gain. Protection for issued permits and the say that states and tribes lose carry no figure, and the grid gain assumes that regions actually build the lines.

In detail

All arguments

Arguments for7 arguments

Fifty-seven million tonnes a year

18points

When wind, solar and storage can connect faster and their power can travel to the next region, gas plants run less. The modelling finds 71 million tonnes less carbon dioxide in 2035 in the four regions it covers, nine percent of their power sector emissions. Across the country and across a decade in which the lines arrive gradually, that averages about 57 million tonnes a year.

Value 7 · ClimateImpact 5.7Plausibility 4.5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is carbon dioxide not emitted from American power plants [1][2]. Its harm falls on people everywhere and for generations, not on those who use the electricity, which places it with the broad environmental goods on this scale. A tonne is priced at the cost of avoiding it elsewhere, which keeps the climate argument on the same footing as every other climate debate on this site. The local air pollution from the same chimneys is a different harm, to different people, and is counted separately in the next argument. The emissions that the bill may add through faster gas pipelines are counted against it on the other side, not netted here. The value is in the upper middle of the scale: a broad environmental good, priced at what it costs to secure it another way.

Impact

The modelling finds 71 million tonnes less carbon dioxide in 2035 in the four regions, a nine percent cut in their power sector emissions, as solar rises 22 percent, wind 14 percent and storage 65 percent against the business-as-usual case [1][2]. Scaled to the country outside Texas by the same factor of 1.6 that the modellers' national cost total implies, that is 114 million tonnes in 2035, with 1.0 as the floor [1][2]. Because the lines and the faster queue take effect from 2029 and rise to the 2035 level, the average over 2027 to 2036 is half the 2035 figure, in a range from three tenths to seven tenths [4]. That gives 57 million tonnes a year, in a range from 21 to 80 million. An earlier estimate for the transmission provisions of the 2024 permitting bill found 6.5 billion tonnes avoided through 2050; read as a straight ramp, that is in the same order of magnitude, but it was made under the tax credits for clean power that Congress has since ended, and by an organisation that campaigned for that bill [7]. Priced at the cost of avoiding a tonne elsewhere, the result is a Normalised Impact of 5.68 on the scale of this evaluation, in a range from 2.1 to 8.0. For scale, American power plants emitted 1,485 million tonnes in 2025. The Impact is the largest real effect in this debate and depends on how fast regions actually build the lines.

▸ Show calculation ▾ Hide calculation
71 million tonnes
Lower power sector carbon dioxide in the four modelled regions in 2035
[1][2]
× × 1.6 Setting, range 1.0 to 1.6: the same factor as for the cost saving, implied by the modellers' national total
113.6 million tonnes
From four regions to the whole country outside Texas
[1][2]
× × 0.5 Setting, range 0.3 to 0.7: effects start in 2029 and rise to the 2035 level, so between 21 and 80 million tonnes a year with the scaling
56.8 million tonnes a year
Average over 2027 to 2036 against the 2035 level
[4]
× × 100 euro-equivalent the cost of avoiding a tonne elsewhere, the rate this site uses; range 2.1 to 8.0 billion euro-equivalent
5.68 billion euro-equivalent a year
Value of a tonne
÷ 1 billion euro-equivalent a point scale of this evaluation
5.68
Normalised Impact
Score 5.68 Impact × 7 Value × 4.5 Plausibility ÷ 10 = 18 of 100

Plausibility

The counterfactual is the same business-as-usual case as for the cost saving: today's queue completion rates and no new interregional lines [2]. The design is a capacity-expansion projection with GenX, so the tonnes follow from modelled dispatch, not from a measured change [2]. Nobody has modelled the bill as introduced: the projection covers two reforms the bill contains and names no bill, and the earlier estimate for the 2024 bill assumed tax credits that no longer exist, so it supports the direction and little more [1][7]. The queue reform and the federal power to approve stalled lines follow from the text; whether regions then build the interregional lines the model assumes is not covered by anything [4]. The confounder that matters is the same as for the cost saving: if wind permits stay slowed, part of the new transmission carries less clean power [3][16]. Reverse causation does not arise. A climate scientist who reviewed the deal expects it to lower emissions on balance, with the fossil fuel provisions acting at once and the clean power gains only after some years [17]. Read as a statement about the world: a little under half of the modelled tonnes can be counted on within the decade. The Plausibility matches the cost saving, because both are the same modelled build seen from two sides.

Evidence: Projection · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Coverage gap · plausibility 4.5

Open: An independent model of the bill as introduced, under today's tax law, would replace the transfer from two modelled reforms and could lift the plausibility to the projection ceiling of 6.0.

Other customers stop paying

16points

Today the lines and substations built to connect a data centre are mostly added to the bills of every customer of the utility. In seven states of the largest American grid region, 4.3 billion dollars of such costs were approved in 2024 alone. The bill forbids shifting these costs for loads of 20 megawatts or more that connect after it takes effect, so households and other businesses keep about 3.9 billion dollars a year.

Value 5 · Electricity billsImpact 3.4Plausibility 9.5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is money on electricity bills: the cost of connecting large computing loads, which utilities today spread across all their customers [20]. No resource is saved; the same lines are built, and only who pays for them changes, which is why the payment by the data centres stands as its own argument on the other side. Households take about two fifths of American electricity, and a dollar on a household bill weighs slightly more than a dollar in a company's accounts, because electricity takes a larger share of small incomes. The receiving side is therefore counted at 1.02, the paying side at 1.0. The sponsors rest the word affordability in the bill's title largely on this provision [3][6]. The value is the middle of the scale, because the stream is money that changes hands.

Impact

Section 2107 of the bill adds a rule to the Federal Power Act: for a load of 20 megawatts or more, no part of the incremental cost of serving it may be shifted to other customers, and the utility must hold financial assurances before it builds [4][19]. The rule covers loads that connect after enactment. The Union of Concerned Scientists counted 4.3 billion dollars of transmission connection costs for data centres approved in 2024 in seven states of the PJM region, and more than 95 percent of the projects passed all of their cost on to customers' bills [20]. For the whole country the figure is set at one and a half times that, in a range from one to two times, because PJM holds the largest cluster of data centres but not all of them. Not all of it would be shifted without the bill: several states have begun to write their own tariffs for large loads, so three fifths is taken as the share the bill changes, in a range from two fifths to four fifths. That gives 3.87 billion dollars a year, in a range from 1.7 to 6.9 when both settings stand at their bounds together. At the weight of 1.02 for those who are relieved, the result is a Normalised Impact of 3.40 on the scale of this evaluation. Both multipliers are settings; the one measured figure is the 4.3 billion dollars of a single year in a single region. The Impact is the same money as in the argument on the data centres' payment, seen from the side that keeps it.

▸ Show calculation ▾ Hide calculation
4.3 billion dollars a year
Connection costs for data centres approved in seven PJM states in 2024
[20]
× × 1.5 Setting, range 1.0 to 2.0: PJM holds the largest cluster of data centres, not all of them
6.45 billion dollars a year
From seven PJM states to the whole country
× × 0.6 Setting, range 0.4 to 0.8: several states are writing their own tariffs for large loads; the rule covers loads of 20 megawatts or more that connect after enactment
3.87 billion dollars a year
Share the bill changes
[4][19]
÷ ÷ 1.16 exchange rate used throughout this evaluation; both settings at their bounds give 1.5 to 5.9 billion euro-equivalent (1.7 to 6.9 billion dollars)
3.34 billion euro-equivalent a year
In euro-equivalent
× × 1.02 two fifths households at 1.05, three fifths businesses at 1.0
3.4 billion euro-equivalent a year
Weight of a dollar for those who are relieved
÷ 1 billion euro-equivalent a point scale of this evaluation
3.4
Normalised Impact
Score 3.4 Impact × 5 Value × 9.5 Plausibility ÷ 10 = 16 of 100

Plausibility

The counterfactual is today's practice, in which connection costs enter the utility's general transmission rates [20]. That the allocation changes follows from the text of the bill: it is a legal consequence, not a behavioural response, so nothing has to be identified [4]. What is uncertain is the amount, because the national figure and the share that states would have reallocated on their own are settings, and that doubt sits in the range. Data centres may answer by building their own generation or by moving behind the meter; then less is paid, but less is built for them as well, and the two sides of this pair shrink together. Reverse causation does not arise. Both sides carry the same amount and the same plausibility, because money that is paid is money that is received. Read as a statement about the world: it is nearly certain that connection costs of this order stop reaching other customers' bills once the rule applies. The Plausibility is that of a rule written into the bill, with only its enforcement by regulators left open.

Evidence: Mechanism · plausibility at most 9.5 Evidence of effect: Definitional

A cheaper grid to run

11points

Lines between regions let cheap wind and solar in one region replace expensive gas in the next, and a faster queue lets that generation connect. Modelling of the two reforms finds about 7 billion dollars a year of lower system costs in 2035 in four regions that use half of American electricity, after paying for the lines. Spread across the country and across a decade in which the lines arrive late, that is about 5.6 billion dollars a year.

Value 5 · Electricity costsImpact 4.8Plausibility 4.5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The sponsors present lower electricity costs as the bill's main purpose [6]. The stream is money: what the power system costs to run, which falls when it needs less fuel and less standby capacity [1][2]. Most of it reaches businesses and the utilities that serve them: of the 125 billion dollars the modellers find for the whole country from 2026 to 2040, 19 billion, about 15 percent, is lower household electricity bills [1]. A dollar saved counts here at the standard weight, the same for a company and for a household of middle income. The lower risk of shortages that the modelling also finds is part of the same system cost and is not counted again [2]. The money that leaves gas suppliers when less gas is burned is partly their margin, not a saved resource; it is not separated out here. The value is the middle of the scale, because the stream is money at its ordinary weight.

Impact

The modelling commissioned by the Center for Climate and Energy Solutions finds that interconnection queue reform and new interregional lines lower grid costs in PJM, MISO, SPP and the non-Californian West by about 7 billion dollars in 2035, after 1.9 billion dollars of transmission investment [1][2]. Those four regions serve 54 percent of American demand, and the modellers' own national total of about 125 billion dollars over 2026 to 2040 implies a factor of 1.6 for the whole country outside Texas, and 1.0 is the floor if the rest of the country gains nothing [1][2]. The bill gives regions two years to set up the new planning processes, and a line takes several more years to build, so savings are taken to begin in 2029 and to reach the 2035 level only at the end [4]. Across 2027 to 2036 that averages half of the 2035 figure, in a range from three tenths to seven tenths. That gives 5.6 billion dollars a year. As a check, the national total of 125 billion dollars, spread on a straight ramp from 2028, gives between 5.5 and 6 billion dollars a year inside the same ten years. With high gas prices the modelled saving nearly doubles to 13 billion dollars in 2035, which sets the top of the range [2]. The range runs from about 2.1 to 14.5 billion dollars a year, wider above the central figure than below it. The Impact is the second largest real effect in this debate and it is net of the cost of the lines themselves.

▸ Show calculation ▾ Hide calculation
7 billion dollars a year
Lower grid costs in PJM, MISO, SPP and the non-Californian West in 2035
[1][2]
× × 1.6 Setting, range 1.0 to 1.6: the four regions use 54 percent of demand; 1.6 is the factor implied by the modellers' own national total of 125 billion dollars for 2026 to 2040
11.2 billion dollars a year
From four regions to the whole country outside Texas
[1][2]
× × 0.5 Setting, range 0.3 to 0.7: planning processes take two years under the bill, so savings start in 2029 and rise to the 2035 level; check: the national total on a straight ramp from 2028 gives 5.5 to 6 billion dollars a year in the same ten years
5.6 billion dollars a year
Average over 2027 to 2036 against the 2035 level
[4]
÷ ÷ 1.16 exchange rate used throughout this evaluation, 1.16 dollars to the euro; with high gas prices the modelled saving nearly doubles, which sets the top of the range at 12.5 billion euro-equivalent (about 14.5 billion dollars); the bottom, four regions only and a slow ramp, is 1.8 (about 2.1 billion dollars)
4.83 billion euro-equivalent a year
In euro-equivalent
[2]
× × 1.0 about 15 percent of the saving is lower household bills, the rest reaches businesses and utilities; both at the standard weight
4.83 billion euro-equivalent a year
Weight of a dollar saved on electricity bills
[1]
÷ 1 billion euro-equivalent a point scale of this evaluation
4.83
Normalised Impact
Score 4.83 Impact × 5 Value × 4.5 Plausibility ÷ 10 = 11 of 100

Plausibility

The counterfactual is the modellers' business-as-usual case: the same demand, fuel prices and tax rules, queues completing at today's rates and no new interregional lines [2]. The design is a capacity-expansion projection with the GenX model, which simulates every hour of the year; it is a mechanism with its parameters set, not a measured effect [2]. Part of what it models comes with the bill: the consolidated queue process, modelled on the Southwest Power Pool, becomes a requirement everywhere within 18 months, and the federal regulator's power to approve stalled lines applies at once [4]. The part that carries most of the quantity is not covered: the interregional lines are capped in the model at what the Department of Energy judged buildable, and the bill requires regions to plan them, not to build them [2][4]. The confounder that matters is the administration's slowdown of wind permits, which the sponsors describe as unresolved and which would shrink the gain without reversing it [3][16]. Reverse causation does not arise, because the saving is computed from the reform, not observed alongside it. The study was commissioned by an organisation that supports the bill, and nobody has modelled the bill as a whole. The doubt is therefore about how much of the modelled saving arrives, not about whether any does, and it is booked as a discount on the size. Read as a statement about the world: a little under half of the modelled saving can be counted on within the decade. The Plausibility is that of a projection whose queue half follows from the bill and whose transmission half still depends on regions deciding to build.

Evidence: Projection · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Coverage gap · plausibility 4.5

Open: The first interregional plans due two years after enactment, or an independent model of the bill as introduced, would show how much transmission is actually selected and could lift the plausibility to the projection ceiling of 6.0.

Fewer deaths from plant smoke

1points

Gas and coal plants that run less also emit less of the fine particles and nitrogen oxides that shorten lives downwind. Studies of wind and solar in the United States count about 1,400 deaths avoided in 2022 for about 275 million tonnes of carbon dioxide displaced. Because the plants displaced here are mostly gas, which is cleaner than coal, about 170 deaths a year are counted.

Value 10 · LifeImpact 0.3Plausibility 4
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is people who die earlier because they breathe the exhaust of power plants [9]. These deaths fall on identifiable neighbourhoods downwind of the plants rather than on the atmosphere, which is why they are counted apart from the carbon dioxide. A life is the highest good on this scale, and an avoided death counts at the full value of a statistical life. The grief of families is added at a tenth of that, as for every death on this site. Illness short of death, such as asthma attacks and hospital stays, is left out, which keeps the figure on the low side. The value is the top of the scale, because the stream is lives.

Impact

The quantity of carbon dioxide avoided comes from the argument above: 56.8 million tonnes a year across the decade [1][2]. A study of American wind and solar finds about 1,400 premature deaths avoided in 2022, when those sources displaced about 275 million tonnes of carbon dioxide, which is about 5 deaths per million tonnes for the mix of coal and gas they replaced [9]. Most of those deaths come from the sulphur of coal plants. The reform mainly displaces gas, whose particle and nitrogen oxide emissions per unit of power are far lower, so 3 deaths per million tonnes is used here, in a range from 1 to 5. That gives 170 deaths a year. Each counts at the value of a statistical life on this site, plus a tenth for the suffering of relatives: a Normalised Impact of 0.26 on the scale of this evaluation, in a range from 0.03 to 0.61, which combines the range of the death rate with the range of the tonnes. The Impact is small beside the carbon dioxide, because the plants displaced are already the cleaner ones.

▸ Show calculation ▾ Hide calculation
56.8 million tonnes a year
Carbon dioxide not emitted from gas and coal plants
[1][2]
× × 3 Setting, range 1 to 5: wind and solar avoided about 1,400 deaths for about 275 million tonnes in 2022, about 5 per million tonnes for a coal and gas mix; the reform mainly displaces gas
170 deaths a year
Premature deaths per million tonnes displaced
[9]
× × 1.4 million euro-equivalent the value this site uses for every avoided death
0.24 billion euro-equivalent a year
Value of a statistical life
× × 1.1 a tenth of the deaths on the same value, as for every death on this site; range 0.03 to 0.61 billion euro-equivalent
0.26 billion euro-equivalent a year
Suffering of relatives
÷ 1 billion euro-equivalent a point scale of this evaluation
0.26
Normalised Impact
Score 0.26 Impact × 10 Value × 4 Plausibility ÷ 10 = 1 of 100

Plausibility

The counterfactual is the same business-as-usual grid as above [2]. The design is a chain: modelled tonnes, then a death rate per tonne taken from a study that combines measured emissions with air quality and health models, not from observed deaths [9]. Every link is named and none has a finding against it, but the last one is a rate transferred from the mix of 2022 to a mix that is mostly gas in the 2030s. The confounder that matters is which plants run less: if the reform displaces coal, the rate rises toward 5; if it displaces only the newest gas plants, it falls toward 1, and that doubt sits in the range. Reverse causation does not arise. The argument inherits the open question of the two above, how much of the modelled build arrives, and adds the transfer of the rate, which is why it stands half a point below them. Read as a statement about the world: about two fifths of the deaths computed this way can be counted on within the decade. The Plausibility is that of a closed chain whose last link is a modelled rate carried over to a cleaner mix.

Evidence: Mechanism · plausibility at most 5 Evidence of effect: Mechanistic · at most 6 Placement: Chain closed, unevidenced · plausibility 4–5

Permits that stay issued

0points

A project that holds all its federal permits could no longer be stopped by a later decision of the same government, except for fraud, a court order or an urgent new danger. The rule covers permits in force since 16 September 2026 and applies to offshore wind farms as much as to gas export terminals. Investors would price less political risk into every large project.

Value 6 · Investment securityImpact 0Plausibility 5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is the value of investment that happens because a permit, once issued, can be relied on [4][18]. It is a gain in production, not a transfer, and it accrues to projects of every kind, which places it with the goods of a working economy on this scale. The five offshore wind farms halted by stop-work orders and a possible future pause on gas export approvals are the cases both parties have in mind [3]. The payments the Treasury would owe when an agency breaks the rule are a different stream and stand in the argument on mistakes that stay permitted. Whether the rule ends the current slowdown of wind permits is, by the sponsors' own account, unresolved [3][16]. The value is just above the middle of the scale: output and investment across the economy.

Impact

Section 1401 bars agencies from revoking, suspending or altering a federal authorisation in force on or after 16 September 2026 for a project with a non-federal sponsor, unless a court orders it, the holder broke the law or committed fraud, or new evidence shows specific and urgent harm [4]. The agency must prove the exception by clear and convincing evidence, and section 1402 lists 46 kinds of energy project that may not be singled out [4][18]. What this is worth depends on how much investment political risk holds back today and how much of it the rule releases. No source gives either number, for wind or for gas. A figure would have to start from the risk premium investors add to projects that need federal permits, and none has been published since the stop-work orders. The Impact is not assessed: the argument belongs to this debate, but its size cannot yet be derived.

Plausibility

The counterfactual is today's law, under which an agency can reopen or suspend an approval it has issued. That the protection exists follows from the text [4]. Whether it changes investment depends on how courts read the exceptions and on whether agencies find other ways to delay, such as withholding the next permit a project needs. Supporters point to the wind farms that were stopped, opponents to projects that deserved a second look; neither side has measured the effect on investment. The rule does not order any pending permit to be issued, so its reach is limited to projects that are already fully approved. The Plausibility is not assessed, because there is no quantity to which it could attach.

Recorded, not yet scored: Recorded but not assessed: no source estimates how much investment the political risk of a revoked permit holds back today, or how much of it the protection of issued permits would release. As soon as a study prices that risk for offshore wind or for gas infrastructure, the gain is booked here as production.

Evidence: Plausibility · plausibility at most 5

Gas for the Northeast

0points

Pipelines into New York and New England have been stopped for years by state water permits, and the region pays some of the highest winter gas prices in the country. If the bill lets such pipelines and capacity additions through, more gas reaches the region and prices fall. This is the main argument of the gas industry for the bill; how large the price effect would be has not been estimated.

Value 5 · Energy billsImpact 0Plausibility 5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is money: what households and businesses in the Northeast pay for gas and for the electricity made from it [4][18]. It is the other side of the argument on pipelines that get built: the same capacity that adds emissions there lowers prices here, and counting one without the other would tilt the ledger. A dollar saved on an energy bill counts at the standard weight. Part of any price fall is a transfer from gas producers and pipeline owners elsewhere to customers in the region and would have to be booked as such. The value is the middle of the scale, because the stream is money.

Impact

The bill confines state water certification of interstate gas pipelines to direct discharges and takes capacity additions inside existing corridors out of environmental review [4][18]. The argument on pipelines that get built counts about 0.75 billion cubic feet a day of capacity that would be in place on average across the decade because of these two changes. What that capacity does to prices in New York and New England depends on how tight the market is on the coldest days, and no source estimates it for this bill. The pipeline that states blocked longest, into New York City, was approved by New York and New Jersey themselves in November 2025, which takes part of the claimed gain out of the bill's account [23]. The Impact is not assessed: the argument belongs to this debate, but its size cannot yet be derived.

Plausibility

The counterfactual is today's law together with the approvals states have recently granted on their own [23]. That more pipeline capacity lowers regional prices is a standard mechanism; how much capacity the bill adds is itself uncertain, and the price effect per unit of capacity has not been measured for these routes. A fall in prices would also raise gas use, which is the emissions argument on the other side. Industry statements give the direction and no figure. The Plausibility is not assessed, because there is no quantity to which it could attach.

Recorded, not yet scored: Recorded but not assessed: no source estimates the effect on gas and electricity prices in the Northeast of the pipeline capacity the bill would let through. As soon as a market study gives a price effect per unit of added capacity for these routes, the saving is booked here, split into a real gain and a transfer from producers.

Evidence: Plausibility · plausibility at most 5

Faster reviews everywhere

0points

The deadlines and the limits on lawsuits apply to every federal project, not only to power lines: highways, transit, water works, mines and housing on federal land. When a project is finished earlier, its benefits start earlier. Court challenges to environmental reviews take about four years from the review to the end of the appeal.

Value 6 · EconomyImpact 0Plausibility 5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is the benefit of public and private projects outside the electricity grid that arrive sooner [4][5]. It is a gain in production and in public services, which places it with the goods of a working economy. The grid's share of this gain is counted in the first three arguments and is not repeated here. The damage a faster or thinner review may let through stands on the other side, in the argument on harm that courts no longer stop. Home builders and road builders support the bill for this reason. The value is just above the middle of the scale: output and investment across the economy.

Impact

The bill sets one year for an environmental assessment and two for an impact statement, takes projects whose only federal link is a loan or a guarantee out of review altogether, and sends challenges straight to an appeals court with 150 days to file [4]. A survey of 387 appeals court cases from 2013 to 2022 finds that 4.2 years pass on average between the review and the end of the appeal, and that agencies win about four in five [22]. How many projects outside energy are delayed by reviews and lawsuits, by how long, and what a year of delay costs has not been added up by anyone for this bill. A sum would need the number of affected projects and their yearly benefit, and both would have to be set here without a source. The Impact is not assessed: the argument belongs to this debate, but its size cannot yet be derived.

Plausibility

The counterfactual is today's procedure. That reviews get deadlines follows from the text; whether projects are then finished sooner depends on agency staffing, which the bill funds only in part, and on what else holds projects up, such as financing and state permits [4]. A legal scholar who opposes the bill expects the shorter filing period to change little, because most challenges are already filed within months [24]. No study links review deadlines of this kind to completion dates. The Plausibility is not assessed, because there is no quantity to which it could attach.

Recorded, not yet scored: Recorded but not assessed: no source counts the federal projects outside the electricity grid that the deadlines and litigation limits would speed up, or values the time gained. As soon as a study gives the number of projects and the delay avoided, the gain is booked here as production.

Evidence: Plausibility · plausibility at most 5

Arguments against8 arguments

Data centres pay

16points

The lines and substations that connect a large data centre would no longer be spread across all customers of the utility. Operators of loads of 20 megawatts or more that connect after the bill takes effect carry the full added cost themselves and must post financial assurances first. That is about 3.9 billion dollars a year which they pay and others no longer do.

Value 5 · Business costsImpact 3.3Plausibility 9.5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is money paid by the operators of data centres and other large computing loads [4][19]. It is the paying side of the argument on other customers who stop paying: the same amount, moving from one group to another. No resource is used up by the payment itself. A dollar in a company's accounts counts at the standard weight, however large the company. Whether higher connection costs slow the building of data centres, and what that would cost the economy, is a different question and is not counted here. The value is the middle of the scale, because the stream is money that changes hands.

Impact

The amount is the one derived in the argument on other customers: 4.3 billion dollars of connection costs approved in seven PJM states in 2024, one and a half times that for the country, and three fifths of it changed by the bill [20]. That is 3.87 billion dollars a year, in a range from 1.7 to 6.9. The rule covers loads of 20 megawatts or more that connect after enactment; arrangements approved before stay in force [4][19]. The federal regulator must also revise transmission pricing so that computing loads pay for the existing grid they use, with a lower rate for service that can be interrupted; that part has no figure and is not counted. The paying side counts at a weight of 1.0, so the Impact equals the amount. The Impact is the same money as in the argument on other customers, seen from the side that pays it.

▸ Show calculation ▾ Hide calculation
4.3 billion dollars a year
Connection costs for data centres approved in seven PJM states in 2024
[20]
× × 1.5 Setting, range 1.0 to 2.0, as in the argument on other customers
6.45 billion dollars a year
From seven PJM states to the whole country
× × 0.6 Setting, range 0.4 to 0.8: states' own tariffs for large loads; loads of 20 megawatts or more connecting after enactment
3.87 billion dollars a year
Share the bill changes
[4][19]
÷ ÷ 1.16 exchange rate used throughout this evaluation; range 1.5 to 5.9 billion euro-equivalent (1.7 to 6.9 billion dollars)
3.34 billion euro-equivalent a year
In euro-equivalent
× × 1.0 a company's dollar at the standard weight
3.34 billion euro-equivalent a year
Weight of a dollar for those who pay
÷ 1 billion euro-equivalent a point scale of this evaluation
3.34
Normalised Impact
Score 3.34 Impact × 5 Value × 9.5 Plausibility ÷ 10 = 16 of 100

Plausibility

The counterfactual is today's practice, in which connection costs enter the utility's general transmission rates [20]. That the allocation changes follows from the text of the bill, so nothing has to be identified [4]. The amount is uncertain, and that doubt sits in the range, shared with the receiving side. Operators who generate their own power or accept interruptible service pay less, and then less is built for them; both sides of the pair shrink together. Reverse causation does not arise. Read as a statement about the world: it is nearly certain that operators of large new loads pay connection costs of this order once the rule applies. The Plausibility is that of a rule written into the bill, and it equals that of the receiving side.

Evidence: Mechanism · plausibility at most 9.5 Evidence of effect: Definitional

Harm that courts no longer stop

3.2points

Under today's law a court can halt a project whose environmental review was flawed until the review is redone. Under the bill it sends the review back while construction continues, endangered species consultations get shorter deadlines, states lose most of their say over pipelines that cross their rivers, and hardrock mines gain waste sites and lighter reviews. About eight projects a year would be built on a review a court found flawed; what each of them damages is a setting.

Value 7 · EnvironmentImpact 1.5Plausibility 3
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is damage to rivers, wetlands, habitats and landscapes that a full review or a court order would have prevented or reduced [14][15]. It falls on local ecosystems and on the people who use them, and much of it cannot be reversed once a pipeline or a mine is built. That places it with the broad environmental goods on this scale, the same as the climate arguments. The loss of procedural rights for neighbours is a separate harm and stands in its own argument, and so do the sites that matter to tribes. Many of the projects that are sued are solar farms, wind farms and power lines, whose faster completion is counted on the other side; the damage they do locally still belongs here. The value is in the upper middle of the scale: lasting local environmental damage.

Impact

The bill makes remand without vacatur or injunction the remedy for flawed reviews, shortens the base period for endangered species consultation from 90 to 60 days, limits the Environmental Protection Agency's veto over wetland fill permits to the time before a permit is issued, confines state water certification of pipelines and power lines to direct discharges, and lets hardrock mines place waste rock on new five-acre sites [3][4][5]. Federal agencies finish about 142 impact statements a year [13]. Of the energy and transport projects that needed one between 2010 and 2018, 28 percent were sued before construction: nearly two thirds of solar projects, half of pipelines, 38 percent of wind farms and 31 percent of power lines [21]. Agencies won about four in five of 387 appeals from 2013 to 2022, so in about one case in five a court found the review wanting [22]. That gives about eight projects a year for which the legal consequence changes from a possible halt to a correction while building goes on. The first such rulings fall in 2029, so across the decade an average of 25 such projects stand built. The damage that a halt and a redone review would have avoided is set at a Normalised Impact of 0.06 a year for each of them, in a range from 0.02 to 0.2; no source gives this figure. The central value sits low in the range, because most of the projects that are sued are solar farms, wind farms, power lines and roads; a large pipeline or mine lies near the top, and a single fossil megaproject above it. The result is a Normalised Impact of 1.5 on the scale of this evaluation, in a range from 0.5 to 5.1. Not counted is the effect on agencies that review more carefully because a halt is possible, and the veto over wetland permits, used 14 times in over fifty years [14]. The Impact is the largest real effect on this side, and its last step is a setting.

▸ Show calculation ▾ Hide calculation
142 decisions a year
Final environmental impact statements a year
[13]
× × 0.28 energy and transport projects with an impact statement, 2010 to 2018; solar nearly two thirds, pipelines 50 percent, wind 38, power lines 31
39.76 projects a year
Share sued before construction
[21]
× × 0.2 agencies won about 80 percent of 387 appeals from 2013 to 2022
7.95 projects a year
Share in which the court finds the review flawed
[22]
× × 3.2 years first rulings under the new remedy in 2029; additions from then accumulate to an average of 3.2 years' worth
25.45 projects
Such projects standing built, averaged across the decade
× × 0.06 billion euro-equivalent Setting without a source, range 0.02 to 0.2: most sued projects are solar farms, wind farms, power lines and roads near the bottom, a large pipeline or mine near the top; gives 0.51 to 5.09 billion euro-equivalent
1.53 billion euro-equivalent a year
Damage a halt and a redone review would have avoided, per project and year
÷ 1 billion euro-equivalent a point scale of this evaluation
1.53
Normalised Impact
Score 1.53 Impact × 7 Value × 3 Plausibility ÷ 10 = 3.2 of 100

Plausibility

The counterfactual is today's law, under which a court can vacate an approval and a state can deny certification on broad water quality grounds [4][5]. That the remedies shrink follows from the text. The chain from there to damage is named, but two links that carry the quantity are unexamined: how often a redone review changed a route or a design, and what that change was worth. No study links the remedy to environmental outcomes. Supporters of the bill argue that most halted projects were later approved unchanged, opponents that the threat of a halt is what makes agencies review carefully; neither claim has been measured [15][24]. The counts of lawsuits come from a sample of large energy and transport projects and from appeals courts, and both have been criticised as unrepresentative of all reviews [21][22]. The confounder that matters is other law: a project that breaks the Clean Water Act or the Endangered Species Act itself can still be enjoined, which removes part of the damage from the bill's account [4]. Reverse causation does not arise. Read as a statement about the world: there is about a three in ten chance that the narrower remedies let through damage on this scale. The Plausibility is low because the chain is open where the quantity is decided, not because anything shows the damage to be absent.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Chain open · plausibility 3–3.5

Open: A study that follows projects after a court vacated their approval, and records how often route or design changed, would close the open links and could lift the plausibility to 4.0 or 5.0.

New money the bill authorises

1.4points

The bill authorises 200 million dollars a year for historic preservation offices, 150 million a year for states that take over endangered species consultations, and smaller sums for a coastal fund, a permitting fund and tribal records. Together that is about 330 million dollars a year across the decade. Congress still has to appropriate it, and usually funds less than it authorises.

Value 5 · Public financesImpact 0.3Plausibility 9.5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is money from the federal budget [4]. What it buys, faster historic preservation and species reviews, is part of what makes the deadlines in the bill achievable, and it is counted in the arguments on the grid rather than here. A public dollar counts at the standard weight. This argument is the payment and nothing else. The value is the middle of the scale, because the stream is public money.

Impact

The text authorises 200 million dollars a year for the Historic Preservation Fund from 2027 to 2036, 150 million a year for endangered species consultation and the states that take it over from 2027 to 2031, 50 million a year for a coastal fund from 2028 to 2032, 20 million a year for the federal permitting fund from 2027 to 2033, 20 million a year for local permitting help until 2031 and 20 million a year for tribal preservation records from 2027 to 2030 [4]. Within the ten years counted here that sums to 3.32 billion dollars, or 332 million a year. An authorisation is a ceiling, not an appropriation. If Congress funds only two fifths of it, the cost falls to about 130 million dollars a year, which sets the bottom of the range; the authorised amount is its top. Damages that developers could claim when an agency withdraws a permit or singles out a kind of project are paid from the Treasury's judgment fund without an appropriation; a single offshore wind farm could run into billions, but the amount depends entirely on how agencies behave and is not counted here [4][19]. No budget office score was available when this evaluation was written. The Impact is small: a few hundred million dollars a year against a federal budget of several trillion.

▸ Show calculation ▾ Hide calculation
3.32 billion dollars
Authorisations within 2027 to 2036
[4]
÷ ÷ 10 years
0.33 billion dollars a year
Per year
÷ ÷ 1.16 exchange rate used throughout this evaluation, 1.16 dollars to the euro; if only two fifths is appropriated, 0.11 billion euro-equivalent (about 130 million dollars)
0.29 billion euro-equivalent a year
In euro-equivalent
× × 1.0 the standard weight for public money
0.29 billion euro-equivalent a year
Weight of a public dollar
÷ 1 billion euro-equivalent a point scale of this evaluation
0.29
Normalised Impact
Score 0.29 Impact × 5 Value × 9.5 Plausibility ÷ 10 = 1.4 of 100

Plausibility

The counterfactual is the same budget without the new authorisations [4]. That the authorisations exist follows from the text of the bill, so nothing has to be identified. How much is appropriated is a decision of future Congresses, and that doubt sits in the range and not in the plausibility. The figures are taken from the bill as introduced; amendments on the floor could change them. Reverse causation does not arise. Read as a statement about the world: it is nearly certain that money is spent under these authorisations, in an amount between two fifths and all of what is authorised. The Plausibility is that of amounts written into the bill, with their funding carried in the range.

Evidence: Mechanism · plausibility at most 9.5 Evidence of effect: Definitional

Pipelines that get built

0.8points

The bill limits what states may weigh when they certify gas pipelines under the Clean Water Act, and it takes looping, compression and capacity additions on existing pipelines out of environmental review. Three pipelines into New York and New Jersey, together 2.15 billion cubic feet a day, were held up by state water permits in the past decade. Some projects like them would go ahead, existing lines would grow faster, and part of the gas would add to emissions.

Value 7 · ClimateImpact 0.4Plausibility 3
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is carbon dioxide and methane from gas that is burned because a pipeline was built or enlarged [12]. It is the same good as the carbon dioxide avoided on the other side and carries the same value, so the two sides of the climate ledger can be compared directly. A tonne is priced at the cost of avoiding it elsewhere. Damage to rivers and wetlands along the routes is a different harm and stands in its own argument. The text contains no order to hold more oil and gas lease sales and no deadline for gas export approvals, the two channels that dominated the debate over the 2024 bill; what it does contain on drilling is an exemption from federal drilling permits for wells on private land where the federal government owns a minority of the minerals [4][8]. Gas export terminals and refineries appear only in the list of projects whose issued permits are protected. The value is in the upper middle of the scale, the same as for every tonne in this debate.

Impact

Three interstate pipelines were held up by state water quality certification in the past decade: Constitution, 0.65 billion cubic feet a day, denied by New York in 2016; the Northeast Supply Enhancement, 0.4, refused several times from 2018; and PennEast, 1.1, abandoned in 2021 without New Jersey's certificate [11][23]. Together that is 2.15 billion cubic feet a day in ten years. The Atlantic Coast pipeline is left out, because it failed on a nationwide federal permit, forest crossings and cost, not on a state certificate [10]. Half of such capacity is taken to go ahead under the bill, in a range from a quarter to three quarters, because water permits were one obstacle among several. A second channel is new: capacity added by looping and compression inside existing corridors no longer needs an environmental review, and for that 0.1 billion cubic feet a day of extra capacity each year is set, in a range from 0.05 to 0.2 if such additions add gas at all, without a source [4][18]. Pipelines take about three years to build, so additions from 2029 accumulate to an average of 3.6 years' worth across the decade: 0.75 billion cubic feet a day. At 70 percent use and with 30 percent of the gas adding to emissions and the rest replacing heating oil, imported gas or other pipelines, in a range from 10 to 50 percent, burning it emits 3.1 million tonnes a year [12]; leaks in production and transport add a fifth. That is 3.8 million tonnes a year, a Normalised Impact of 0.38 at the cost of avoiding a tonne elsewhere, in a range from 0.12 to 0.76 with the ranges of the settings combined; whether the bill adds this gas at all is a question of plausibility. The Impact is a fifteenth of the carbon dioxide avoided on the other side, because pipelines are a small share of what the bill moves.

▸ Show calculation ▾ Hide calculation
2.15 billion cubic feet a day
Pipeline capacity held up by state water certification in ten years
[10][11][23]
÷ ÷ 10 years
0.22 billion cubic feet a day, each year
Per year
× × 0.5 Setting, range 0.25 to 0.75: water permits were one obstacle among several, and states have begun to approve such projects themselves
0.11 billion cubic feet a day, each year
Share that goes ahead under the bill
[23]
+ + 0.1 Setting without a source, range 0.05 to 0.2 if such additions add gas at all: looping, compression and capacity additions on existing interstate pipelines
0.21 billion cubic feet a day, each year
Capacity added inside existing corridors without environmental review
[4][18]
× × 3.6 years about three years to build, so additions from 2029 accumulate to an average of 3.6 years' worth
0.75 billion cubic feet a day
Capacity in place, averaged across the decade
× × 0.7 × 0.3 Setting, net share range 0.1 to 0.5: much of the gas replaces heating oil, imported gas or other pipelines
0.16 billion cubic feet a day
Use of the capacity and share that adds to emissions
× × 20 million tonnes per billion cubic feet a day 54.81 kg per thousand cubic feet × 365 days
3.14 million tonnes a year
Carbon dioxide from burning it
[12]
× × 1.2 Setting, range 1.1 to 1.5
3.77 million tonnes a year
Methane leaks in production and transport
× × 100 euro-equivalent the cost of avoiding a tonne elsewhere; with the ranges of the settings combined, the result lies between 0.12 and 0.76 billion euro-equivalent
0.38 billion euro-equivalent a year
Value of a tonne
÷ 1 billion euro-equivalent a point scale of this evaluation
0.38
Normalised Impact
Score 0.38 Impact × 7 Value × 3 Plausibility ÷ 10 = 0.8 of 100

Plausibility

The counterfactual is weaker than today's statute suggests. New York and New Jersey approved the Northeast Supply Enhancement themselves in November 2025, and the Environmental Protection Agency has proposed a similar narrowing of state certification by regulation; the bill mainly fixes that narrowing in law beyond the next change of government [15][23]. The design is a chain with every link named, but two links that carry the quantity are not established: that water permits, and not demand or cost, were what stopped these pipelines, and that review-free capacity additions add gas that would not have flowed anyway. A review of the 2024 bill's fossil fuel provisions judged their climate effect small and hard to quantify, which points the same way [8]. The confounder that matters is demand: where buyers for the gas exist, capacity is often added on another route, and then nothing is added by the bill. Reverse causation does not arise. A climate scientist reviewing the deal expects it to lower emissions on balance, with the fossil fuel provisions acting at once and the clean power gains later [17]. Read as a statement about the world: there is about a three in ten chance that the bill adds emissions through pipelines on this scale. The Plausibility is low because the chain is open at the links that decide the quantity, not because anything shows it to be false.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Chain open · plausibility 3–3.5

Less say for neighbours

0.042points

People who live near a pipeline, a mine or a highway lose part of their say. Comment periods are capped, only those who commented in detail and are directly harmed may sue, they have 150 days and one court instead of six years and two, and a court that finds the review flawed lets the project continue. Federal agencies finish about 140 full impact statements a year.

Value 7 · ParticipationImpact 0.01Plausibility 6
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is the procedural right of people near a project to be heard and to challenge the decision [4][5]. It is participation in a decision about one's own surroundings, not money or health; the material consequences of a weaker review are counted in the argument on environmental harm. Such rights weigh more than comfort but less than the vote itself, which is why they are counted as a fraction of a vote in a national election. The fraction depends on how much of the process falls away, and here a court stage is narrowed rather than abolished. The value is in the upper middle of the scale: participation in decisions about one's own surroundings, at a small measured size.

Impact

Federal agencies completed 1,276 final environmental impact statements with a record of decision between 2010 and 2018, about 142 a year [13]. The bill caps public comment at 45 to 60 days, limits standing to those who commented in sufficient detail and suffer direct harm, cuts the time to sue from six years to 150 days, sends the case straight to an appeals court, and makes remand without halting the project the only remedy for a flawed review [4][5]. A legal scholar who opposes the bill regards the shorter deadline as largely symbolic, since most challenges are filed within months already, and the loss of the remedy and the standing rule as the substance [24]. For the households around a project this takes away much of what the court stage could achieve, which is counted as one stage of participation falling away: 2 percent of the value this site gives a vote in a national election, per household and decision. The number of households affected by an average impact statement is set at 5,000, in a range from 1,000 to 20,000, because linear projects such as pipelines and highways cross many communities. That gives a Normalised Impact of about 0.014 on the scale of this evaluation, in a range from 0.003 to 0.057. Smaller projects under environmental assessments are left out, which keeps the figure low. The Impact is the smallest in this debate, because each household loses only part of one stage, once per project.

▸ Show calculation ▾ Hide calculation
142 decisions a year
Final environmental impact statements a year
[13]
× × 5,000 Setting, range 1,000 to 20,000: pipelines, transmission lines and highways cross many communities
710,000 households a year
Households affected per decision
× × 20 euro-equivalent one stage of participation falls away: 2 percent of a vote in a national election (1,000 euro-equivalent); comment capped, standing narrowed, 150 days to sue, remand without a halt; range 0.003 to 0.057 billion euro-equivalent
0.01 billion euro-equivalent a year
Value of the lost participation per household and decision
[4][5]
÷ 1 billion euro-equivalent a point scale of this evaluation
0.01
Normalised Impact
Score 0.01 Impact × 7 Value × 6 Plausibility ÷ 10 = 0.0 of 100

Plausibility

The counterfactual is today's procedure: comment periods set by each agency, six years to sue, and vacatur or injunction as possible remedies [4][5]. That the procedural rights shrink follows from the text; what is set is the number of households and the fraction of a vote, and both sit in the range. The design is a mechanism without behaviour in it, so the confounder is not who reacts but how many people actually take part: most neighbours never comment, which a per-household value already absorbs. Reverse causation does not arise. Agencies could still lengthen comment periods to 75 days with the sponsor's consent, and courts may still enjoin a project that breaks a law other than the review statutes, which leaves part of the right intact [4]. Read as a statement about the world: it is likely, about six in ten, that neighbours of federal projects lose part of their say on this scale. The Plausibility is that of a direct legal change whose size depends on a setting.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6

Mistakes that stay permitted

0points

The protection of issued permits also binds a government that finds an approval was a mistake. It may act only on new evidence of specific and urgent harm, must prove its case by clear and convincing evidence, and otherwise owes the developer a quarter to half of the project's cost, or half to all of it if a kind of project was singled out. The money comes from the Treasury without a vote of Congress.

Value 5 · Public financesImpact 0Plausibility 5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

Two streams meet here. One is public money: damages paid from the Treasury's judgment fund when an agency withdraws a permit or delays a kind of project [4][19]. The other is the harm from projects that a later government would have stopped or changed on better knowledge and no longer can. The first is a transfer to developers and would be booked with a receiving side; the second is real damage and would count at the higher value of the environmental arguments. The list of protected projects runs from wind farms to coal mines, refineries and gas export terminals, so the rule cuts in both political directions [4]. The value is set at the middle of the scale for the money; the damage would stand higher once it can be sized.

Impact

Section 1401 allows a court to award delay costs and between 25 and 50 percent of the project's cost when an agency interferes with a fully permitted project, and section 1402 between 50 and 100 percent of the expected construction cost for a proven pattern of unequal treatment [4][19]. Both are paid from the permanent judgment fund and need no appropriation. A single offshore wind farm or export terminal costs several billion dollars, so one case could exceed everything the bill authorises in a year. How many cases there would be depends on how future governments behave, and the purpose of the rule is that there are none. Neither the expected payments nor the damage from approvals that can no longer be corrected has been estimated. The Impact is not assessed: the argument belongs to this debate, but its size cannot yet be derived.

Plausibility

The counterfactual is today's law, under which an agency can reopen its own approval and owes no damages of this kind. That the liability exists follows from the text [4]. Whether it is ever paid depends on agency conduct and on how courts read a standard of proof that is new in this field. A pattern of unequal treatment requires proof of intent and a statistical comparison with the previous five years, which is a high bar. No precedent exists for damages measured as a share of project cost. The Plausibility is not assessed, because there is no quantity to which it could attach.

Recorded, not yet scored: Recorded but not assessed: neither the damages the Treasury could owe under the protection of issued permits nor the harm from approvals that can no longer be corrected has been estimated, and both depend on how future governments act. As soon as a budget office score or a first court award gives a figure, the payment is booked here as a transfer and the harm on the environmental value.

Evidence: Plausibility · plausibility at most 5

States lose the say

0points

A high-voltage line that a state has not approved within a year can be approved by the federal regulator instead, no longer only inside corridors the Energy Department has designated, and the federal permit carries the power to take land. States and tribes also lose most of their grounds for refusing water certification to pipelines and power lines. The federal siting power has existed since 2005 and has hardly been used.

Value 6 · State authorityImpact 0Plausibility 5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is the authority of states to decide what is built on their territory and on whose land [4][5]. It is a good of the federal order, not money and not the environment: a state that would have said no for reasons of its own loses that choice even where the line is useful. Landowners along a route are affected through the same door, because a federal permit opens the federal courts for condemnation. The benefit of the lines that get built this way is counted in the arguments on the grid. The objection is raised from the side of the states, whatever the kind of project. The value is just above the middle of the scale: the room states have to decide for themselves.

Impact

Section 2101 lets the federal regulator permit an interstate line of 230 kilovolts or more once a state has withheld approval for a year, without the corridor designation today's law requires, and for lines of 345 kilovolts or more the federal and state proceedings may run side by side [4][5]. Condemnation of rights of way then follows the federal rules of civil procedure. For water certification, states and tribes get nine months and must show by clear and convincing evidence that no change to the project would meet water quality standards [4]. How many lines and pipelines would be approved against a state's will under the new rules, and how a lost decision of that kind should be valued, has no source and no settled rule on this site. The Impact is not assessed: the argument belongs to this debate, but its size cannot yet be derived.

Plausibility

The counterfactual is today's law, with federal siting confined to designated corridors and broad state discretion over water certification. That the powers shift follows from the text [4]. How often they are used is open: most lines are approved by states in the end, and a regulator in Washington may be slow to overrule a governor. States keep their role in the federal proceeding and their say over retail rates. No count of lines blocked by a single state exists that would give the frequency. The Plausibility is not assessed, because there is no quantity to which it could attach.

Recorded, not yet scored: Recorded but not assessed: no source counts the transmission lines and pipelines that would be approved against a state's will under the new siting and water certification rules, and this site has no settled value for a decision taken from a state. As soon as both exist, the loss is booked here on state authority.

Evidence: Plausibility · plausibility at most 5

Sacred sites less protected

0points

Federal projects must take account of places that matter to tribes, and tribes are consulted before a decision. The bill counts only direct alterations of such a place as harm, not what is seen or heard from it unless that significantly diminishes it, narrows which places qualify, ends consultation when the environmental review ends, and limits tribal water certification to direct discharges. What is lost in a given case is a place that cannot be rebuilt.

Value 7 · Tribal heritageImpact 0Plausibility 5
▸ Show reasoning & sources ▾ Hide reasoning & sources

Value

The stream is the protection of places of religious and cultural importance to tribes, and the say tribes have before a federal project touches them [3][4][18]. It is closer to a right than to a comfort: the places are tied to the continued practice of a community and cannot be replaced once altered. This site has no value of its own for cultural heritage, so the argument is filed under participation, the nearest good on the scale. The general loss of say for neighbours of projects stands in its own argument; this one concerns a duty the federal government owes to tribes specifically. The value is in the upper middle of the scale: a community's say over places it cannot replace.

Impact

Section 2301 limits an adverse effect on a historic property to a reasonably foreseeable, direct alteration and leaves out visual, atmospheric and audible effects unless they significantly diminish the property [4]. A property of traditional religious or cultural importance must be an identifiable location or feature, which makes whole landscapes harder to protect, and tribal preservation offices are reviewed every two years and can have their functions taken over [4][18]. Consultation follows fixed deadlines, an agency may proceed if a party has not answered within 30 days, and challenges must be filed within 150 days [5][18]. How many sites would lose protection, and how the loss of one should be valued, has no source. The bill also gives tribes more freedom to grant rights of way across their own land, which points the other way. The Impact is not assessed: the argument belongs to this debate, but its size cannot yet be derived.

Plausibility

The counterfactual is today's consultation under the historic preservation law, without statutory deadlines and with indirect effects in scope. That the definitions narrow follows from the text [4]. How often a project was changed because of visual or audible effects on a tribal site is not recorded anywhere. The bill adds money for preservation offices, which could speed consultation without weakening it. Tribal governments had not taken a common position when this evaluation was written. The Plausibility is not assessed, because there is no quantity to which it could attach.

Recorded, not yet scored: Recorded but not assessed: no source counts the tribal sites that would lose protection under the narrower definitions of harm and of eligible places, and this site has no value for cultural heritage. As soon as a count of affected consultations exists and the value is set, the loss is booked here.

Evidence: Plausibility · plausibility at most 5

How robust the result is

Under the assumptions used, the proposal brings more than it costs in 99 of 100 model runs.

gains more than it costscosts more than it gains

Each dot stands for 200 of the 20,000 model runs.

The result rests on how much of the modelled grid gain is counted. The faster queue and the federal power to approve stalled lines come with the bill, but it orders the planning and not the building, so a little under half of the modelled gain is counted. On that basis the bill comes out ahead in 99 of 100 runs. Carbon dioxide is priced at the cost of avoiding a tonne elsewhere; at its estimated damage, three and a half times higher, climate gain and pipeline emissions would both grow and the gap would widen. The payment for connecting data centres is counted in full on both sides, as money their operators pay and as money everyone else no longer pays; it cancels except for a small difference in weight. The damage that narrower court remedies let through rests on a set value per affected project; it would have to average three times the top of the range used before the expected balance turned.

Show the distribution of the model runs

20,000 simulated futures

Very likely better · 99 %

In 99 of 100 simulated futures the proposal brings more than it costs.

No effect is known exactly; every argument has a range. The calculation draws a possible size for each argument from its range 20,000 times and counts how often the benefits exceed the costs.

Net effect in points per future. Left of zero the drawbacks outweigh the benefits, right of zero the benefits win. Eight in ten futures fall between +10 and +43 points.
Scale and assessment

Scale of this evaluation: Normalised Impact, unitless and calibrated to this topic. For comparison, one point here corresponds to about 1 billion euro-equivalent a year.

Size class:medium

Effect large (+25) · 85 % of the effect is positive

Areas show individual argument weights. The positive share of the effect nets out money transfers between the two sides; the probability of a positive net effect is determined separately.

How we score →

Review and sources

AI evaluation · not yet reviewed by a human

This evaluation was produced and sourced by an AI model; a human review is still pending. Figures and conclusions may still change. The review log is at the foot of the page.How review works →

Last reviewed by Claude Opus 5.5 · October 29, 2026 · 6× AI, not yet reviewed by a human
  1. October 29, 2026AI reviewClaude Opus 5.5record updated

    Page texts for the new page layout (draft G) with quotes from primary sources; numbers unchanged.

  2. October 21, 2026AI reviewClaude Fable 5.1record updated

    Money amounts in argument texts stated in dollars per style-en.md; scores, chains and balance unchanged

  3. October 20, 2026AI reviewClaude Fable 5.1record updated

    Summary and scale note rewritten to the style of 20 Oct 2026; numbers unchanged

  4. October 16, 2026AI reviewClaude Opus 5.5record updated

    Pipeline emissions range rederived in quadrature instead of multiplying the bounds; the review-free additions range now starts above zero, since whether they add gas is carried by plausibility.

  5. October 16, 2026AI reviewClaude Opus 5.5record updated

    Text aligned with the card values (Bilanz 2.2)

  6. October 12, 2026AI reviewClaude Opus 5.5First evaluation

    KI-Entwurf der Themen-Automatisierung, am 12.10.2026 unabhängig nachgeprüft und korrigiert übernommen

Evaluations are produced with AI support and reviewed on a schedule for new developments; human passes are marked separately.How we review →

Show all 25 sourcesHide sources
  1. Center for Climate and Energy Solutions (30 September 2026): New analysis finds federal transmission reforms would slash power sector emissions, boost grid reliability and save consumers billions. c2es.org
  2. Greenline Insights for C2ES (29 September 2026): Grid and economic impacts of interconnection and interregional transmission reform — GenX modelling of PJM, MISO, SPP and the West for 2035. static1.squarespace.com
  3. Heatmap News (30 September 2026): The Senate's big bipartisan permitting deal, explained. heatmap.news
  4. US Senate, 119th Congress (30 September 2026): S. 5653, Bipartisan American Affordability and Jobs Act of 2026, text as introduced. govinfo.gov
  5. Pillsbury Winthrop Shaw Pittman (October 2026): Senate permitting bill would reshape environmental and energy reviews. pillsburylaw.com
  6. Senate Committee on Energy and Natural Resources (30 September 2026): Lee, Capito, Heinrich and Whitehouse introduce the Bipartisan American Affordability and Jobs Act of 2026. energy.senate.gov
  7. RMI (August 2024): Estimating the climate impacts of the transmission provisions of the Energy Permitting Reform Act of 2024. rmi.org
  8. Center on Global Energy Policy, Columbia University (20 December 2024): Fossil fuel provisions of the Energy Permitting Reform Act of 2024 and their effects on climate goals. energypolicy.columbia.edu
  9. Millstein, O'Shaughnessy and Wiser, Cell Reports Sustainability (2024): Climate and air quality benefits of wind and solar generation in the United States from 2019 to 2022. tethys.pnnl.gov
  10. Argus Media (5 July 2020): Dominion and Duke cancel the Atlantic Coast pipeline. argusmedia.com
  11. World Oil (27 September 2021): PennEast gas pipeline halted in latest blow to US energy infrastructure. worldoil.com
  12. US Energy Information Administration (2024): Carbon dioxide emissions coefficients by fuel. eia.gov
  13. Council on Environmental Quality (June 2020): Environmental impact statement timelines, 2010 to 2018. nepa.gov
  14. Natural Resources Defense Council (2 October 2026): Hard questions about the permitting bill. nrdc.org
  15. Mother Jones (9 October 2026): Why this long-sought, bipartisan Senate deal has environmentalists feuding. motherjones.com
  16. Mining Weekly / Reuters (1 October 2026): US senators hit deal on energy project permitting bill, vote after midterm elections. miningweekly.com
  17. Energy Connects / Bloomberg (2 October 2026): What the Senate permitting deal means for US climate emissions. energyconnects.com
  18. National Law Review (October 2026): Senate negotiators reach agreement on permitting reform text. natlawreview.com
  19. McGuireWoods (October 2026): Bipartisan American Affordability and Jobs Act of 2026 proposes sweeping reform for energy and infrastructure projects. mcguirewoods.com
  20. Union of Concerned Scientists (2025): Billions of dollars in unreported data center costs in PJM: 4.3 billion dollars of transmission connection costs approved in 2024 in seven states. ucs.org
  21. Bennon and Wilson, Environmental Law Reporter (October 2023): NEPA litigation over large energy and transport infrastructure projects. elr.info
  22. Breakthrough Institute (2024): Understanding NEPA litigation: 387 appeals court cases from 2013 to 2022. thebreakthrough.org
  23. Argus Media (7 November 2025): New York approves Williams' NESE gas pipeline. argusmedia.com
  24. Stanford Law School, Deborah Sivas (4 October 2026): Permitting reform legislation would take a meat axe to environmental law when what we need is a surgical scalpel. law.stanford.edu
  25. Vinson & Elkins (October 2026): To the finish line: ten ways the new Senate permitting bill is a boon for energy infrastructure. velaw.com

More debates in this area

Very likely better · 100 %
Size: large
Restore the Electricity Credits

Put the clean electricity production and investment credits back on their original open-ended schedule, reversing the 2025 phase-out.
Likely worse · 30 %
Size: large
Repeal the Vehicle Standards

Withdraw the 2009 finding that greenhouse gases endanger health, and with it every federal greenhouse gas standard for cars and trucks.
All debates in Energy and Climate →