이스트 팔레스타인을 리허설처럼 보이게 만드는 조용한 규칙 제정

Key Takeaways

• The EPA quietly proposed on February 13, 2026 to strip out 14 major provisions of the 2024 chemical accident prevention rules, with zero major news coverage.
• The central target is the RAGAGEP (Recognized and Generally Accepted Good Engineering Practices) gap analysis requirement, the rule that forces plants to know whether their safety practices are dangerously out of date.
• The lobbying trail is explicit: The American Chemistry Council praised the rollback within hours, calling it a return to a “data-driven approach,” the same language it used in 2019 and 2017 to describe every prior rollback.
• The historical parallel is exact: It took the Bhopal disaster in 1984 to create these protections in the first place. The arc is now being reversed.
• The comment period runs approximately 45 days after Federal Register publication, meaning public input closes before most Americans know this is happening.

The Rule Nobody Reported

On February 13, 2026, the U.S. Environmental Protection Agency (EPA) released the prepublication text of a proposed rule titled “Common Sense Approach to Chemical Accident Prevention.” It is one of the most consequential regulatory rollbacks of the year. Odds are you never read about it.

The proposal targets the Risk Management Program (RMP), a set of rules under the Clean Air Act (CAA) that governs roughly 12,000 to 15,000 chemical facilities across the United States that handle threshold quantities of toxic or flammable substances. These are places that store chlorine, ammonia, hydrogen fluoride, vinyl chloride, and dozens of other chemicals that, if released into the air, can kill people within minutes.

The RMP is not new. It has existed in some form since 1996. But the 2024 “Safer Communities by Chemical Accident Prevention” (SCCAP) rule, finalized under the Biden administration after years of rulemaking, had significantly strengthened it, adding provisions that required facilities to:

• Conduct gap analyses comparing their safety practices to current engineering standards
• Submit to third-party audits after accidents
• Analyze safer technology alternatives before assuming the most dangerous chemical process is the only option
• Improve employee participation in safety planning
• Strengthen community notification systems

The February 13, 2026 proposal rolls back all of it. It targets 14 specific areas on an accelerated timeline, because the EPA wants to finalize the rollback before the May 10, 2027 compliance deadlines for the 2024 provisions kick in. If successful, facilities will never have to comply with the stronger rules at all.

What RAGAGEP Actually Means and Why Gutting It Matters

To understand why the gap analysis requirement is the most consequential item on the chopping block, you need to understand what RAGAGEP stands for: Recognized and Generally Accepted Good Engineering Practices.

Think of RAGAGEP as the building code equivalent for chemical plants. Just as a 1960s building might be structurally legal but violate every modern fire and electrical code, a chemical facility built or designed in 1995 might be using equipment, valve specifications, or process controls that the American Petroleum Institute (API) or the American Society of Mechanical Engineers (ASME) has since classified as dangerously substandard.

The 2024 SCCAP rule required Program Level 3 facilities (the highest-risk category, handling the most dangerous chemicals) to conduct a gap analysis: a one-time review comparing what they are actually doing on the floor to what current RAGAGEP standards say they should be doing. The gap analysis does not require facilities to immediately fix everything. It requires them to know what they do not know.

The 2026 proposed rule eliminates that requirement entirely. Under the new proposal, facilities only need to maintain a general obligation to keep process safety information “up to date,” a standard so vague it is functionally unenforceable.

The American Chemistry Council (ACC), the primary trade group for chemical manufacturers whose members include Dow Chemical, ExxonMobil, and BASF, praised this specific change within hours of the proposal’s release. The ACC called it “a data-driven, performance-focused approach to RMP.” This is identical language to what the ACC used in 2019, when the first Trump administration rolled back the 2017 Obama-era strengthening, and to what it used in 2017 when it opposed those 2017 additions originally. The ACC has opposed every strengthening of the RMP since its inception.

The math of why is straightforward: a gap analysis is expensive. For a large facility handling multiple Program Level 3 processes, a comprehensive RAGAGEP audit can cost hundreds of thousands of dollars and require operational downtime. Avoiding that cost is worth lobbying for. The question is who pays the price when the gap goes unidentified.

The Safer Technology Requirement: The One That Could Have Changed Everything

The STAA, or Safer Technology and Alternatives Analysis, is the provision that deserved the most public attention and received the least.

Under the 2024 SCCAP rule, facilities undergoing process hazard analyses (PHAs) were required to actively ask: Is there a less dangerous way to do this? Can chlorine gas be replaced by a less toxic form? Can anhydrous ammonia (NH₃, a colorless gas with a TLV occupational ceiling of 25 ppm and an IDLH of 300 ppm) be substituted with aqueous ammonia in lower concentrations for the same industrial purpose?

The STAA did not require facilities to change anything. It required them to document that they considered whether they could. This is the chemical equivalent of requiring a driver to check mirrors before changing lanes: not to guarantee safety, but to establish that basic due diligence occurred.

The 2026 proposal eliminates the STAA practicability assessment requirement for existing non-new processes. The EPA’s stated reason: it duplicates existing requirements under OSHA’s Process Safety Management (PSM) standard. But OSHA’s PSM standard does not contain an equivalent STAA requirement. The duplication claim is, at best, imprecise.

The actual effect is that facilities will no longer have to document that they asked themselves the most basic safety question in chemical engineering.

The Historical Rhyme: From Bhopal to EPCRA to Now

The arc of chemical safety regulation in the United States has a clear origin point: December 3, 1984.

At approximately 12:30 a.m. local time in Bhopal, India, roughly 45 tons of methyl isocyanate (MIC) gas, a chemical used in pesticide manufacturing, escaped from a Union Carbide India Limited plant. Wind carried it into surrounding slums while residents slept. Estimates of the immediate death toll range from 3,800 (Union Carbide’s own figure) to 15,000 to 20,000 (advocacy organizations and Indian government estimates). More than 500,000 people were exposed. Survivors experienced permanent blindness, chronic respiratory illness, and neurological damage. Their children faced elevated rates of birth defects.

The Union Carbide facility had no functional gas scrubber (disabled three weeks prior), no working gas flare system (out of action for three months), and a refrigeration system whose coolant had been drained weeks earlier for use elsewhere in the plant. Nearby residents had no idea what chemicals the plant was storing, no emergency notification system, and no evacuation plan.

Two years later, the United States Congress passed the Emergency Planning and Community Right-to-Know Act (EPCRA) as part of the Superfund Amendments and Reauthorization Act. EPCRA required chemical facilities to disclose what hazardous substances they stored and in what quantities. This foundational “right to know” would eventually become the backbone of the RMP. In 1996, the EPA codified the RMP rule itself under the Clean Air Act, giving communities the first systematic means of knowing what chemical risks existed in their neighborhoods and requiring facilities to have plans for preventing and responding to accidents.

It took the deaths of tens of thousands of people to build these protections. The 2026 proposed rule is dismantling them on grounds of “administrative burden.”

The pattern is not new. The RMP has been a 30-year political pendulum:

Year	Action
1996	RMP rule finalized under the Clean Air Act
2017	Obama administration finalizes expanded RMP amendments
2019	Trump administration rescinds most 2017 additions
2024	Biden administration restores and expands protections (SCCAP rule)
2026	Trump administration proposes rollback to mid-2000s standards

What has not changed across this entire pendulum: the American Chemistry Council has been on the same side of every single vote.

Third-Party Audits: Who Gets to Grade Their Own Homework

One of the 2024 SCCAP rule’s most significant provisions was the requirement for third-party audits following chemical accidents. If a facility had an accidental chemical release (one of the events the entire RMP system is designed to prevent), an independent auditor would review what went wrong and whether the facility’s safety management systems were adequate.

The 2026 proposal either rescinds this requirement entirely or limits it dramatically: third-party audits would only be triggered after two accidents within a 10-year period, with a sunset clause on that restriction.

The logic of the rollback, as stated by EPA, is that third-party audits are “duplicative” of internal incident investigations and OSHA inspections. The problem with that logic is that OSHA’s budget has been chronically underfunded for decades. At current staffing levels, a typical facility can expect an inspection once every 165 years. Self-reported incident investigations, conducted by the facility that just had the accident, are by definition not independent.

Oklahoma Attorney General Gentner Drummond endorsed the rollback on February 20, 2026, calling it a correction of “federal overreach” harmful to oil, gas, and consumers. Oklahoma has a significant oil and gas chemical processing presence and a consistent history of opposing federal environmental oversight.

Who Wins, Who Pays

The financial structure of this rollback is worth being precise about.

Who benefits directly:

• Large chemical manufacturers who would have faced RAGAGEP gap analysis costs
• Refiners and petrochemical processors who handle Program Level 3 substances
• Trade associations like the ACC and the American Fuel and Petrochemical Manufacturers (AFPM), which lobbied explicitly against the 2024 SCCAP rule

Who pays the external cost:

• Communities within the “vulnerability zone” of covered facilities, defined as the area where a catastrophic release could cause death or serious injury. EPA’s RMP data shows these zones overlap with millions of residential addresses, disproportionately in lower-income communities and communities of color.
• Workers inside the facilities, whose process hazard analyses will no longer need to document safer technology alternatives
• Local emergency responders, who will have less standardized information about what chemicals they might face in a release event

The third-order effect on insurance: Chemical plant operators carry liability coverage for accidental releases. Underwriters price that coverage based on perceived accident risk. Weaker regulatory requirements create the appearance of lower risk, which suppresses insurance premiums in the short term. This creates a moral hazard: the mechanism designed to align corporate financial incentives with safety outcomes (insurance pricing) is distorted when regulators remove the transparency data actuaries rely on.

What the EPA Actually Says

Credit where it is due: the EPA’s stated reasoning for this rollback deserves to be stated accurately, not strawmanned.

The agency argues that the 2024 SCCAP additions were:

1. Duplicative of existing OSHA PSM standards that govern similar safety requirements
2. Disproportionately burdensome relative to the marginal safety benefit they provided
3. Prescriptive in a way that reduced facility flexibility to achieve safety outcomes through site-specific means

These are legitimate arguments if they are true. The challenge: the EPA’s own data on RMP facilities shows declining accident rates over the history of the program, which industry uses as evidence the rules worked well enough at their mid-2000s level. But declining accident rates in a regulated industry can reflect either that the rules are strong enough or that reporting thresholds are too high to capture near-misses and minor releases. The RMP’s self-reporting structure has known gaps. The Union of Concerned Scientists (UCS) and environmental justice advocates have repeatedly noted that the program captures only the accidents that result in formal notifications, not the full universe of chemical releases.

The boring hypothesis is that EPA administrators genuinely believe OSHA’s PSM provides adequate parallel protection. That is plausible. The cynical hypothesis is that the rulemaking timeline (finalize before 2027 deadlines make compliance mandatory) was engineered to benefit industry. That is supported by the explicit timeline. The two hypotheses are not mutually exclusive. Regulatory capture often looks like sincere belief in the positions of the people who fund your confirmation.

What Comes Next

The proposed rule is not yet in effect. The timeline:

1. Federal Register Publication: Pending (as of February 21, 2026). Publication starts the public comment clock.
2. 45-Day Comment Period: Open to any member of the public, environmental group, or affected community. Comments are legally required to be considered before finalization.
3. Virtual Public Hearing: Scheduled approximately 14 days after Federal Register publication.
4. Finalization Target: Before May 10, 2027 (when surviving 2024 SCCAP provisions would require compliance).
5. Legal Challenges: Expected in the D.C. Circuit Court. Every major RMP rulemaking since 2017 has been litigated.

If you live near a covered facility, you can look up your facility’s RMP plan through the EPA’s Risk Management Plan database (rmp.epa.gov). The database shows what chemicals are stored, in what quantities, and what the estimated “worst-case scenario” vulnerability zone looks like for your address. Under the 2026 rollback, those plans will be less rigorously scrutinized and the community notification requirements that fill them with meaningful data will be weakened.

If your state’s attorney general is not already filing comments or joining a coalition opposing this rule, that is a question worth raising directly with their office.

The NIH, meanwhile, announced on February 4, 2026, the opening of an East Palestine, Ohio health research office, three years after the February 3, 2023 derailment, to begin a five-year, $10 million study of long-term health outcomes from that accident. The formal opening ceremony was held on February 18, 2026. The federal government is funding research into the consequences of chemical exposure while simultaneously proposing to weaken the regulatory environment that governs it.

The Bottom Line

The 2026 EPA RMP rollback is not a story about regulatory philosophy. It is a story about who bears the cost of chemical accidents and who profits from avoiding the precautions designed to prevent them.

The RAGAGEP gap analysis meant: if your plant’s safety practices are a decade out of date, you must at least know that. The EPA is proposing to eliminate that requirement.

The STAA meant: before you operate an extremely hazardous process indefinitely, you must at least document whether a less dangerous alternative exists. The EPA is proposing to eliminate that requirement.

The third-party audit requirement meant: if your facility causes an accidental chemical release, an independent examiner will review what went wrong, not your own safety team. The EPA is proposing to eliminate that requirement.

What remains are the documentation requirements that existed before Bhopal taught the world their insufficiency. The community notification systems will be “simplified” to two data points. The voluntary compliance culture remains the weakest link in chemical safety.

It took a disaster that killed tens of thousands of people to build this architecture. The 2026 proposed rule is dismantling it in a 45-day comment period that most affected communities do not know is happening.

The comment period is a legal right to be heard. The EPA is required by law to respond to substantive public comments before finalizing this rule. That is not nothing.