Asbestos (United States, 1930s to present)

Asbestos disease was reported early and the cancer link was recognized by the 1960s; thousands of firms were sued, Johns-Manville filed bankruptcy in 1982 and created a trust, a 1991 court struck down EPA's broad ban, and in 2024 EPA banned only the ongoing chrysotile uses.

The record

What was known and when (attributed, contested in part)

• Egilman, Bird and Lee (2014), who reviewed confidential corporate documents and litigation testimony, report that in 1933 UK factory inspector Merewether argued asbestos use should be prohibited if it caused disease at "slight" exposure; that some companies, including MetLife, reportedly understood by the 1930s that the original exposure limit did not protect against asbestosis; that industry-sponsored animal tests in the 1940s found cancer in mice; and that by the mid-1940s the international scientific community had recognized an asbestos and cancer link. They argue that producers delayed lower exposure limits and later built a "state of the art" defense for litigation, and that the cancer link was not a legitimate scientific debate. This is an academic paper with a stated position. The industry's side of these claims was not available in the sources we opened. (Egilman et al., Int. J. Occup. Environ. Health 2014; opened)
• The same paper says Irving Selikoff's 1964 conference drove wider publicity, and that industry defense arguments treated 1964 as the date the medical community accepted the link. It reports the Braun and Truan study (1956 to 1958), sponsored by a miners' association, published no significant lung cancer excess while a private report stated otherwise. (Egilman et al.; opened. We did not open the underlying documents.)
• OSHA exposure limits, as listed by the same paper: 5 fibers per cubic centimeter (emergency standard, 1971), 2 (effective 1976), 0.2 (1986), 0.1 (1994). (Egilman et al.; secondary, the Federal Register notices not opened)
• EPA lists lung cancer, mesothelioma and asbestosis as health effects, and notes symptoms "may take many years to develop following exposure"; smoking worsens risk of lung disease. (EPA, Learn about asbestos; opened)

Johns-Manville, the trust, and Section 524(g)

• Johns-Manville filed Chapter 11 in August 1982. The Christian Science Monitor reported in 1988 that directors cited auditors who said potential liabilities exceeded net worth, that it was then a $1.8 billion building-products company, and that about 16,000 cases had been filed by 1982 with tens of thousands more expected. Manville's CEO W. Thomas Stephens said the company would acknowledge past conduct and focus on compensation, and that it had divested all asbestos products by December 1985. (Christian Science Monitor, Oct. 3, 1988; opened)
• The reorganization funded a trust with $815 million in cash and notes from Manville and its insurers, plus $75 million a year in stock for 24 years starting in 1991, and a backstop of up to 20 percent of net earnings. Projected total was more than $2.5 billion over 26 years. Shareholders would be diluted by as much as 98 percent. (CSM 1988; opened)
• Trust official David Austern told the Senate Judiciary Committee on September 25, 2002 that claims processing began in November 1988, that about 500,000 claimants had been paid and over $2.9 billion distributed through August 31, 2002, and that claimants then received a pro rata payment of 5 percent of each claim's liquidated value. Austern was the trust's general counsel, so this is testimony by an interested party. (Senate Judiciary testimony; opened)
• Section 524(g) of the Bankruptcy Code lets a court confirming a chapter 11 plan issue an injunction channeling asbestos claims to a trust. The trust must assume the debtor's liabilities, be funded by the debtor's securities and future payment obligations, and own or be entitled to own a majority of the voting shares of the debtor or its parent. The court must find that the debtor faces "substantial future demands" whose number and timing cannot be determined, and that at least 75 percent of the voting claimants approve. A court-appointed legal representative must protect future claimants. (11 U.S.C. 524, Cornell LII; opened. The statute's enactment year was not in the text we read.)
• GAO (2011) reported about 100 companies had gone bankrupt at least partly due to asbestos liability; 60 trusts had been set up since 1988 with about $37 billion in assets. (GAO-11-819; opened, summary page)
• RAND (2005) estimated about 730,000 individuals had brought claims through 2002 against some 8,400 business entities, with $70 billion spent by defendants and insurers, "more than half" consumed by litigation expenses of both sides and 42 percent going to claimants. (RAND MG162; opened, summary page)

Corrosion Proof Fittings v. EPA (5th Cir. 1991)

• EPA's 1989 rule prohibited manufacture, importation, processing and distribution of most asbestos-containing products in stages. The Fifth Circuit (opinion by Judge Jerry Smith, panel of Brown, Smith and Wiener) decided on October 18, 1991, to "GRANT the petition for review, VACATE the EPA's proposed regulation, and REMAND." The court applied the Toxic Substances Control Act standard that EPA use "the least burdensome requirements" to protect adequately against unreasonable risk, with substantial evidence review. (947 F.2d 1201, copy at law.resource.org; opened, last 4,000 characters not read)
• The court cited EPA's own estimates: cost about $450 to 800 million; benefits of 202 or 148 lives saved depending on discounting. Cited costs per life saved ranged from $6 to 8 million for gaskets and $7 to 8 million for paper products to $43 to 76 million for asbestos pipe, with a high of $74 million. It said the ban on original-equipment brakes "appears reasonable." The holding on banning products no longer produced applied only to products not being made, imported or processed on July 12, 1989. (947 F.2d 1201; opened)
• This is a ruling that EPA's rulemaking record did not meet TSCA's standard, not a finding that asbestos is safe. The Federal Register notice of 2024 says the court "vacated and remanded most of the 1989 rule," and that certain products and new uses remained banned (tool-extracted from the notice; we did not read the full relevant paragraph).

The 2024 EPA rule

• EPA issued a final rule under TSCA section 6(a) on chrysotile asbestos. It was effective May 28, 2024, and EPA determined chrysotile "presents an unreasonable risk of injury to health," citing mesothelioma and lung, ovarian and laryngeal cancers. Three companies own eight chlor-alkali facilities that use asbestos diaphragms. Manufacture, including import, of diaphragms was prohibited at the effective date; processing, distribution and commercial use are prohibited five years after, with extensions possible to 8 and 12 years. Oilfield brake blocks, aftermarket automotive brakes and linings, other vehicle friction products and other gaskets are also prohibited, as are sheet gaskets in chemical production. Converting chlor-alkali plants is estimated to cost $2.8 to $3.4 billion. Monetized benefits quoted for some uses were small, about $6,000 a year at a 3 percent discount rate; EPA did not quantify others. (89 Fed. Reg., March 28, 2024, GovInfo; opened, first 100,000 of about 120,000 characters)
• The rule covers chrysotile, the form still imported. It did not by itself ban all asbestos in the United States; legacy uses (asbestos already in buildings) and other fiber types are outside what we read. The "Part 2" rule on legacy uses was not opened.
• On June 16, 2025, EPA asked the Fifth Circuit in Texas Chemistry Council v. EPA to hold the challenge for six more months while it prepares a notice-and-comment rulemaking to revisit parts of the 2024 rule (workplace controls for sheet gaskets in non-titanium dioxide production, whether permanent controls could replace a ban, and feasibility of alternatives), estimated at about 30 months. The rule stays in effect meanwhile, and the court had granted a 120-day abeyance on February 14, 2025. (Bergeson and Campbell, June 28, 2025; opened. A law firm client alert; the docket was not opened. Status after June 2025 unknown.)

Current U.S. imports and use

USGS reports apparent consumption of 110 metric tons in 2024 (estimate), supplied entirely from stockpiles, and says the chloralkali industry has made up 100 percent of fiber consumption since 2015 at the latest. Imports 2020 to 2023 came from Brazil (91 percent) and Russia (9 percent); reported shipments from China and Poland were likely misclassified. EPA lists aftermarket auto brakes and linings, oilfield brake blocks and sheet gaskets as the only known asbestos-containing imports. How USGS computes apparent consumption was not clear from the page and we do not rely on its yearly values. (USGS Mineral Commodity Summaries 2025, Asbestos; opened)

Deaths per year: sources disagree

The figures count different diseases, different years and different methods and are not comparable. The advocacy sources are on one side; no industry-side estimate was found. We could not open the IHME, McCormack or ICOH originals. (EWG Asbestos Nation; ADAO; CDC MMWR 53(28); all opened)

What this case shows

1. Workers and families could not see what employers and insurers might have known. Egilman et al. argue from internal documents that exposure limits stayed high for decades. If true, those exposed had little means to vote or bargain on risk. These are documented allegations from litigation material; we did not open the primary documents.
2. Latency means the people who pay are not the people who decide. EPA says symptoms may take many years to appear. Decisions about use were made decades before disease.
3. The compensation system was built by court and statute, not by a public vote. The Manville trust and Section 524(g) channeled claims to trusts. The trust official's own figures show a 5 percent pro rata payment in 2002. Claimants and future claimants had a representative, but the payment formula was set in the plan.
4. Regulators acted slowly and were reversed. EPA's 1989 rule lost in 1991 on cost and process grounds. A broader ban did not follow, and a narrower chrysotile rule came in 2024, 35 years later.
5. Litigation costs were large. RAND found over half of $70 billion went to litigation expenses. That is a cost to all parties, including those who were sick.

The strongest objections, and our answers

1. Objection: Companies acted on the science of the time. Mitigation: The industry's own account was not available to us. Manville's CEO said in 1988 that the company would acknowledge past conduct and had divested asbestos by 1985. We do not infer intent from outcomes and report that Egilman et al. is an argument from documents we did not read.
2. Objection: The Fifth Circuit was right, since the 1989 rule cost too much per life saved. Mitigation: The court's cost-per-life figures ($6 million to $74 million) came from EPA's own analysis, and the standard was set by Congress in TSCA. We report the holding and the court's statement that some brake bans looked reasonable. We do not say what the right number should be.
3. Objection: Bankruptcy trusts protect future claimants. Mitigation: Agreed; Section 524(g) requires a representative for future claimants and a 75 percent vote. GAO reported 60 trusts with about $37 billion. The Manville trust paid about 500,000 claimants more than $2.9 billion by 2002, at 5 percent of liquidated value then. Both facts are shown.
4. Objection: The 2024 rule proves asbestos is now banned. Mitigation: It bans listed ongoing uses of chrysotile, with phase-in of up to 12 years for some chlor-alkali plants. Legacy use is outside what we read. EPA has also told the Fifth Circuit it plans to revisit parts of the rule.
5. Objection: Death estimates are inflated by advocacy groups. Mitigation: The two estimates we found come from advocacy organizations, and we say so. CDC counts only asbestosis in the report we opened. No neutral all-cause estimate was located.
6. Objection: Litigation was driven by lawyers, not harm. Mitigation: RAND's figures on litigation cost support part of this. RAND did not state a share of claimants who were not sick in the page we read, so we do not claim that.

Open questions: what would change our minds

• The enactment year and legislative record of Section 524(g) were not in the sources we opened.
• The primary documents behind Egilman et al. (Metropolitan Life, Raybestos-Manhattan and others) were not opened; the claim of industry knowledge rests on that paper.
• We did not find an industry-side account of what firms knew, or the Federal Register notices for OSHA limits.
• The docket in Texas Chemistry Council v. EPA was not opened; status after June 2025 is unknown.
• The full text of the Federal Register notice (last 20,000 characters) was not read, including the legacy-use discussion.
• How USGS computes "apparent consumption" was unclear.
• The IHME, McCormack (2012) and ICOH originals were not opened. A neutral estimate of U.S. asbestos deaths would help.
• Manville Trust payments after 2002 and the current payment percentage were not found.
• The last 4,189 characters of the Fifth Circuit opinion were not read.
• Whether the number of claimants not sick can be established from RAND's full report is unknown; only the summary page was read.

Sources

Evidence status and correction history

Key claims

Correction history

Found an error or a better source? We keep every correction on the record with its date and reason. Nothing here is quietly rewritten.