The command to impose and collect a US methane charge still stands and its first charged year is calendar year 2034 as read 2026-08-08 — the duty was timed, not removed

What is happening?

The Waste Emissions Charge is the only charge the United States federal government has ever imposed directly on a greenhouse gas. Congress wrote it into the Inflation Reduction Act of 2022 and set it to begin with methane emissions reported for calendar year 2024. As of 2026-08-08 it begins with emissions reported for calendar year 2034 — a ten-year deferral enacted on 2025-07-04.

The statute is section 136 of the Clean Air Act, codified at 42 U.S.C. 7436. Subsection (c) directs the Administrator to impose and collect a charge on methane emissions above a threshold from any facility that reports more than 25,000 metric tons of carbon dioxide equivalent a year under subpart W of the federal greenhouse gas reporting rule. Subsection (d) names nine industry segments — onshore and offshore production, gathering and boosting, processing, transmission compression, transmission pipelines, underground storage, LNG storage, and LNG import and export. Subsection (e) sets a rate ladder of USD 900 for emissions reported for calendar year 2024, USD 1,200 for 2025, and USD 1,500 for 2026 and each year after. Subsection (f) sets the thresholds at 0.20 percent of gas sent to sale for production, 0.05 percent for nonproduction and 0.11 percent for transmission.

The scale, as EPA estimated it in 2024 using emissions reported for 2022, was about 364 facilities screened down from 2,112 subpart W reporters, about 20 million metric tons of carbon dioxide equivalent subject to the charge, and revenue averaging about USD 525 million a year between 2024 and 2027. Those 20 million metric tons are 0.3 percent of total United States greenhouse gas emissions in 2022. All four figures come from the same agency analysis and are keyed to 2022, not to today.

The charge was switched off three times, by three instruments of different kinds. Congress disapproved the implementing rule under the Congressional Review Act and the President signed that resolution as P.L. 119-2 on 2025-03-14; EPA then removed the rule text from the Code of Federal Regulations on 2025-05-19. P.L. 119-21, signed 2025-07-04, amended subsection (g) at title VI, section 60012(b), replacing calendar year 2024 with calendar year 2034. The same law rescinded the unobligated money on the grant side of section 136. Because subsection (e) was never renumbered, the first rate that can now apply is the USD 1,500 tier, and the USD 900 and USD 1,200 rates sit in the code naming years the charge can no longer reach.

Whose problem is this?

Every body that could restore the charge is a body that already voted to defer it, and the deferral was written as two words inside a reconciliation bill. Nothing outside those bodies has a lever, because nothing is being violated.

Where does this problem end?

The boundary here is not whether methane warms the planet but whether a charge that Congress already wrote is presently collectible. That question has a clean answer and the answer is no.

What is the state now, and what should it be?

Now

Needs a new measurementthe target state: section 136 states thresholds and prices, not an outcome. The one outcome standard inside it is subsection (f)(6)(ii), which asks whether compliance with the section 111 performance standards would achieve emissions reductions equivalent to or greater than the proposed rule — and that proposed rule was nullified in 2025, so the benchmark now points at a document with no legal force. No source opened here states what absolute methane reduction the charge was expected to produce, and none scores what the ten-year deferral is worth. The GAO major rule report of 2024-12-03 gives about USD 48 million in social costs through 2035 and about USD 180 million in net benefits, but those score the compliance rule rather than the charge.

How big is it?

Between 17.6 million and 46.6 million people, and that range is a proximity proxy rather than a count of the people the charge would have protected.

The low end is 17.6 million Americans living within one mile of an active oil or gas well, from the first peer-reviewed nationwide proximity measurement, published 2017-08-23. It covers the production segment only, and the source does not state the year of the well inventory behind it. The high end is 46.6 million people, 14.1 percent of the contiguous United States population, living within about a mile of at least one piece of fossil fuel infrastructure across five stages, from a study reported on 2025-11-17. That count includes refining and end use, which the charge never reached. Inside it the study gives more than 20 million near extraction sites and more than 6 million near storage — the two categories that map most closely onto the segments in subsection (d).

The subpart W shaped middle, extraction plus storage, would land near 26 million before removing double counting, but no source publishes it as a figure in its own right, so it is not used as a bound here.

What the range cannot do is narrow to the regulated universe. The charge would have applied to about 364 facilities above the thresholds, not to every well and tank, and the roster of those facilities has never been published. It also counts neighbours rather than the climate pathway; methane is a global forcer, so on that axis the affected population is everyone and the number stops being useful. Both framings are wrong in opposite directions and no source opened here offers a third. Finally it is a stock and not a flow. The same person is counted once here and is exposed across ten deferred reporting years, and nothing converts proximity into person-years.

Under what conditions does it arise?

1. A duty with a start date behaves differently from a duty that was repealed. Subsection (c) still says impose and collect, and EPA is in perfect compliance with it, because subsection (g) now says the collecting begins with emissions reported for 2034. There is no final agency action to challenge, no unreasonable delay to compel, and no party who can say a duty was skipped. A repeal would have created a defendant and a record. A moved date creates neither.

2. The three instruments are of different kinds, so they have different undo routes. A Congressional Review Act joint resolution is not undone by ordinary legislation restoring a date. A reconciliation amendment to subsection (g) is not undone by a rulemaking. A rulemaking that suspends subpart W is not undone by either. Restoring the charge requires the date moved back, a rule that would otherwise be substantially the same specifically authorized, and the reporting substrate still standing. Any single restoration leaves the charge inoperative.

3. The reciprocal is cheap and the restoration is not. Section 60012(b) is a two-word amendment inside a reconciliation bill. Doing the same thing again in 2033 costs the same two words and needs no hearing, no record and no finding. Undoing it costs three separate legislative and regulatory acts that have to line up. That asymmetry is the mechanism, and it does not depend on anyone acting in bad faith.

4. The charge is computed from a measurement program that the same agency can switch off. Subsection (c) reads the charge off what a facility reports under subpart W, and subpart W is a regulation rather than a statute. On 2025-09-16 EPA proposed to remove reporting obligations for most source categories and to suspend the remaining subpart W segments until reporting year 2034. Pricing built on top of a discretionary measurement program inherits the fragility of that program.

What has been tried?

Two directions ran at once and both point away from collection. The instruments that switched the charge off were legislative, and the instruments that would have let it lawfully retire under subsection (f)(6) — the section 111 performance standards — are themselves being narrowed and delayed. No document opened here connects the two.

What was found?

Why is it still unsolved?

Enforcement absent — the obligation is on the books in the present tense, and for ten reporting years there is nothing to enforce it against.

Start with what did not happen. Nobody repealed the charge. Subsection (c) reads today exactly as it read in 2022, and an agency that collects nothing under it is not in violation of it, because subsection (g) now says the collecting starts with emissions reported for 2034. That is why this is not an exemption. An exemption carves a class out of a regime that keeps operating on everyone else; here all nine segments and all of the roughly 364 facilities remain inside the definition, and it is the regime itself that has been switched off. The duty was timed, not removed. Section 136 does contain a genuine exemption, at subsection (f)(6), for facilities already covered by equivalent section 111 standards — but that provision is not what stopped the charge, and its trigger requires an agency determination that has never been made.

The second movement is the layering. It is tempting to describe this as one postponement, and the assignment that reached this desk did. It is three. The rule was annulled under the Congressional Review Act, which also bars EPA from reissuing anything substantially the same unless Congress specifically authorizes it. The start year was moved by a reconciliation amendment. The grant side was defunded by the same law. Each of the three has its own undo route and none of the routes overlaps. A Congress that moved the date back to 2027 would find no rule and no lawful way for the agency to write one quickly; an agency granted authorization to write the rule would find the date still at 2034. This is what makes a deferral more durable than a repeal, and it is what a single-lock reading misses.

The third movement is the quietest. The charge is arithmetic performed on subpart W reports, and on 2025-09-16 EPA proposed to suspend those reports for the same segments until reporting year 2034. The two 2034 dates match exactly and no document opened here mentions both. If that proposal is finalized, then a Congress that restored the 2024 start date in, say, 2030 would find no reported emissions for the intervening years to charge against. Meanwhile the record for the years the charge was originally supposed to price is already not public. The latest published subpart W figures are for reporting year 2023, and the deadline for reporting year 2025 has been moved to 2026-10-30. The charge is off, the measurement of what it would have priced is going dark, and the standards that were supposed to be able to replace it are being narrowed. Nothing here required a single violation.

What observation would mean it is solved?

Candidates — (a) Congress moves the start year in subsection (g) back and specifically authorizes a rule to implement it (b) EPA publishes a final rule under section 136 with a stated effective date and a first collection year (c) reported methane from subpart W facilities, or vented and flared volumes, fall for several consecutive years.

(a) alone restores a date and not a charge. A start year with no implementing rule is what the statute already has. The Congressional Review Act bar sits between the date and any rule that resembles the one annulled in 2025, and no agency, court or committee document opened here says how far a rule written for 2034 would have to differ to clear it. Moving the date without settling that question buys a line in the code and nothing else.

(b) alone can be written to collect nothing. Thresholds, netting procedures and the exemption in subsection (f)(6) all live below the level of the effective date. A rule can be final, in force, and still reach far fewer than 364 facilities. It is also reversible by the same three routes that produced the present state, which means a rule observed today is not a rule observed in 2034.

(c) alone measures the wrong thing twice over. Reported emissions move with commodity prices, production volumes and reporting rules, not only with abatement, and the reporting rule itself is proposed for suspension — a fall could mean fewer emissions or fewer reports. The vented and flared series is a gas volume rather than the methane tonnage the charge is computed on, so it cannot substitute. The three have to be read together, and (c) has to be read against whether subpart W was still collecting data in the years it covers.

What is it connected to?

Fills with researchthe Congressional Review Act as an instrument for durable deregulation, the greenhouse gas reporting program as shared infrastructure for other climate rules, the section 111 performance standards for the oil and gas sector, methane pricing and border adjustment regimes outside the United States, and the wider question of what happens to statutory duties whose effective dates are moved rather than repealed. Relation type and evidence grade were not confirmed in this round.

What these sources do not say

• The agency page for this program never states when the charge applies. It was last updated 2026-04-29, ten months after the start year moved, and it describes the rule nullification and the CFR removal without a single reference to 2034 or to the law that put it there. A reader who goes to the canonical federal page for the Waste Emissions Charge cannot learn from it when the charge is due.
• The wider methane program page still describes the charge in the present tense. Updated 2026-02-25, it says the program establishes a Waste Emissions Charge and cites USD 1.36 billion in assistance, mentioning neither the 2034 date nor the rescission of unobligated funds that two other sources report.
• Nobody scores the deferral. The only revenue figure in circulation is about USD 525 million a year across 2024 to 2027, keyed to 2022 emissions and to the original rate ladder. Multiplying it by ten is invalid — the ladder is now flat at USD 1,500 from 2034, the netting procedures that produced the 364-facility screen lived in the annulled rule, and emissions for 2024 through 2033 are unknown and partly unreported. No budget estimate for section 60012 appears in any source opened here.
• Nobody says what evaluating options and obligations obligates the agency to do, or by when. There is no implementing rule, no deadline, no milestone, no reporting requirement to Congress and no published work plan. The agency says it will provide more information at an appropriate time and no source defines that time.
• No document connects the two 2034 dates. The proposal that would suspend subpart W reporting until reporting year 2034 does not mention the Waste Emissions Charge or section 136 anywhere in its abstract, even though the charge is computed from subpart W reports. Neither does the deadline extension that followed it.
• The roster of covered facilities has never been published. About 364 exists only as an aggregate in a 2024 agency analysis. Without a list there is no way to say which states bear the deferral, who lives near those facilities, or whether 364 is still the right number for 2034, and no source projects the count forward.
• Nobody says whether the 2024 and 2025 liabilities were extinguished or merely never assessed. Under the original text those calendar years closed with rates attached. One secondary source treats them as gone. No source opened here states the mechanism by which that happened.
• Whether the reissue bar would block a rule written for 2034 is unresolved in print. The strongest statement available is that it is uncertain whether the agency could take a similar course of action, and no agency, court or committee document opened here addresses it further.

See the evidence

No primary congressional document was read. The statutory text was read from a code service rather than from the Office of the Law Revision Counsel, because that office refused the connection twice, and congress.gov returned HTTP 403 to the enrolled text of the law that moved the date. So the single most load-bearing citation in this document — the attribution of the 2034 date to title VI section 60012(b) — rests on one editorial amendment note in one code service and was not corroborated. Four Federal Register documents and two Federal Register queries were read through the JSON API, because the human readable pages began redirecting to a rate limit interstitial partway through the round; the API returns the same abstracts and metadata. The agency pages, the emissions data page, the energy statistics series and the two proximity studies were opened directly. Everything about scale — the 364 facilities, the 20 million metric tons, the USD 525 million — comes from a single secondary report of a single agency analysis, and no second independent source for any of those four numbers was found. Where sources disagree the disagreement is left visible rather than resolved. One secondary source states that the funding provisions of section 136 were repealed, while the code service still shows subsections (a) and (b) with no repeal note and carries only the amendment to subsection (g); either the rescission operated on unobligated balances without striking the authorizing text, or the code service has not caught up, and the enrolled text that would settle it could not be opened. The two sources also give different totals for the money — USD 1.36 billion on the agency page as of 2026-02-25 and over USD 1.5 billion in the client alert as of 2025-07-25 — and neither states the appropriated, obligated and rescinded amounts as three separate figures. One further item is reported but not verified and is therefore not asserted anywhere in this document: a committee stage appropriations bill approved 2025-07-22 that would bar the agency from spending funds to impose the charge, whose enactment no source opened here confirms. The analysis from the law school program predates the 2025-07-04 amendment and does not mention it, so its statement that the obligation to pay survives is true of the command in subsection (c) and not of any emissions year before 2034. This is a Path A output (research-based definition), so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range 17,600,000–46,600,000 As of 2017 to 2025

Sensitivity The width is the gap between one segment on a nine year old inventory and every stage of fossil fuel infrastructure including two stages the charge never covered. The same high end study reports more than 20 million people near extraction sites and more than 6 million near storage, the two categories that map most closely onto the segments in subsection (d) of 42 U.S.C. 7436, which would put a subpart W shaped middle near 26 million before removing double counting. That middle is not published as a figure in its own right and is therefore not used as a bound. What this number fails to count is the actual regulated universe. The charge would have applied to about 364 facilities above the statutory thresholds, on an agency estimate made in 2024 from emissions reported for 2022, and the roster of those facilities has never been published, so proximity cannot be narrowed to them. It also fails to count exposure over time. This is a stock rather than a flow, and the deferral runs across emissions reported for calendar years 2024 through 2033, with no source converting proximity into person years. In the opposite direction the number is far too small. Methane is a global climate forcer and a precursor of ground level ozone, so its burden is not proximity limited, and on that axis the affected population is everyone and the figure stops being useful. Both framings are wrong in opposite directions and no source opened in this round offers a third. Separately, no source states how many people the rescinded monitoring and abatement assistance under subsections (a) and (b) of the same section would have reached.

Regional breakdown No source opened in this round breaks either proximity count down by state or region, and the roster of the roughly 364 facilities that would have been subject to the charge has never been published, so there is no basis for saying which states bear the deferral. Splitting a national figure by population share is not permitted and would be meaningless here, because oil and gas infrastructure is concentrated in a small number of states rather than distributed with population.