How Timber Products Michigan Ran a Polluting Boiler for Decades Without Testing It

EvilCorporations.com • EPA Docket No. CAA-05-2026-0022 / Timber Products Michigan Limited Partnership • Estimated reading time: 6.7 minutes

The Non-Financial Ledger: What $112,297 Doesn’t Cover

Munising, Michigan is a small town on the southern shore of Lake Superior. It is the kind of place where the air is supposed to be clean. The whole brand of the Upper Peninsula is wilderness, cold water, pine trees, distance from the industrial Midwest. The people who live there are not, by and large, people who have the money or connections to pick up and relocate if the air goes bad. They stay. They breathe.

PM2.5 is not an abstraction. It is particulate matter small enough to bypass your nose and throat entirely and lodge deep in your lungs. Long-term exposure to elevated PM2.5 levels is linked to asthma attacks, cardiovascular disease, reduced lung development in children, and premature death. The EPA sets a reporting threshold of 10 tons per year precisely because concentrations above that level are considered a meaningful risk to public health.

Timber Products Michigan Limited Partnership’s own data shows the facility has been reporting PM2.5 emissions above that threshold since at least 2017. The boiler that was producing those emissions, Boiler No. 4, was running above its permitted emission limit the entire time. The company knew the 1995 test had already flagged a violation: that test recorded an emission factor of 0.590 lbs per 1,000 lbs of exhaust gas against a legal ceiling of 0.50. Then they simply stopped testing and kept operating.

The workers inside that facility breathed the air closest to the source. The neighborhood surrounding Highway M-28 East breathed the drift. The children who went to school downwind from a wood-fired boiler that was almost certainly exceeding its pollution limit for years are not named in this settlement document. They have no line item. They received no payment. The $112,297 fine goes to the federal government, not to any person whose lungs absorbed what the permit prohibited.

There is no mechanism in this settlement to compensate anyone for breathing air that should not have been breathed. That is not an oversight in the legal paperwork. That is how environmental enforcement in the United States works.

Legal Receipts: Their Own Words

The following quotes come verbatim from the EPA’s Consent Agreement and Final Order, Docket No. CAA-05-2026-0022, filed April 29, 2026.

“The Information Request Responses stated that no emissions testing had been performed at the Facility since February 1, 2012.”

• This is a company admission, submitted in writing to the EPA in response to a formal Section 114 information request, that testing stopped at least by February 2012. The EPA’s information request asked for all testing dating back to February 2012; the company reported nothing.
• The practical floor on how long the boiler went untested is even worse, as clarified in the next quote.

“…no performance test, or emissions test of any kind, was performed at Boilers #2, 3, and 4, between the 1995 [Performance T]est and February 2012.”

• The company confirmed in a direct email that the gap was not just “since 2012.” The three boilers covered by the 1995 test, including the boiler already known to be over the limit, were never retested between 1995 and 2012. That is a minimum of 17 years with no testing, and when the company’s own 2022 statement is read alongside the 2022 information request response, the total gap between tests runs from 1995 to the EPA-mandated test in August 2023: 28 years.
• The 1995 test had already shown Boiler No. 4 at 0.590 lbs/1,000 lbs, above the 0.50 permit limit. The company was not operating in ignorance of a potential problem. They had the data. They chose not to retest.

“Performance testing conducted in August of 2023 identified an emission factor at Boiler No. 4 of 0.677 lbs PM/1000 lb of exhaust gases, corrected to 50% excess air.”

• The permitted ceiling is 0.50. The 1995 reading was 0.590, already 18% over the limit. The 2023 reading is 0.677, now 35.4% over the limit. The pollution did not stay flat; it worsened. The years of avoided testing coincided with worsening exceedance.
• The only reason this 2023 test exists is because the EPA ordered it. The company did not conduct it voluntarily.

“The MAERS Report submitted for reporting year 2021 did not include emission data for the kilns at the Facility.”

• Kilns are listed as emission units in the facility’s own permit. Omitting their data from the annual emissions report is a separate regulatory violation from the boiler exceedance. It means the 2021 report submitted to Michigan regulators understated the facility’s true pollution output.
• Together, the two violations created a situation in which the company was emitting more than permitted and reporting less than it emitted. The gap between legal reality and disclosed reality pointed in the same direction: toward minimizing apparent harm.

Public Deception: The Gap Between the Reports and Reality

The facility’s public regulatory record presented a picture of compliance that the underlying facts did not support.

• What was reported: The company filed annual Michigan Air Emissions Reporting System (MAERS) reports for 2017 through 2021, indicating ongoing regulatory participation and apparent transparency. What was happening: Those same MAERS reports covered years during which Boiler No. 4 was operating above its permitted emission limit, a fact that went undetected because no independent emissions test had been conducted in decades.
• What was reported for 2021: The MAERS report submitted for reporting year 2021 presented itself as a complete account of facility emissions. What was omitted: The kiln emission units at the facility were left out of that report entirely, in violation of Michigan Administrative Code r. 336.202.

Profit-Maximization at All Costs: Compliance Was Optional

The decision to skip emissions testing for 28 years is not a clerical accident. Emissions performance testing costs money. Failing that test triggers a compliance obligation and potentially costly equipment upgrades. The calculation the company made, whether formally or by institutional inertia, was that not testing was cheaper than testing.

• The 1995 performance test documented that Boiler No. 4 was already operating at 0.590 lbs PM/1,000 lbs exhaust gas, against a permitted ceiling of 0.50. That result should have triggered remediation. Instead, it became the last test on record for nearly three decades.
• By 2023, when the EPA forced a retest, the emission factor had climbed to 0.677 lbs/1,000 lbs, a 35.4% exceedance of the legal limit and a clear indicator that whatever was driving the violation had not improved; it had worsened over time without intervention.
• Emissions testing at these boilers was not affirmatively required by the permit on a set schedule, but the 1995 result created a documented, known violation. The company possessed that data and chose not to act on it. The equipment upgrade (a new multi-cyclone dust collector, now required under the 2025 permit) was deferred for decades.
• The total penalty assessed is $112,297. The CAFO does not disclose the company’s revenues or the cost savings achieved by deferring dust control upgrades. The fine resolves the civil penalty liability; it does not recover the economic benefit of avoided compliance costs.

Legal Minimalism: Following the Letter, Ignoring the Point

The company’s approach to emissions reporting illustrates a recurring playbook: do exactly what the rules literally require, and nothing that the rules are obviously designed to produce.

• Michigan’s annual emissions reporting requirement under Mich. Admin. Code r. 336.202 is designed to give regulators an accurate, current picture of what is being emitted. Timber Products filed annual MAERS reports. But the emission figures in those reports were based on the 1995 test data, not current measurements. Annual filings without updated underlying data create the form of compliance while hollowing out the substance.
• The permit condition establishing the 0.50 lb/1,000 lb limit on Boiler No. 4 exists to protect air quality. There is no source in the CAFO establishing that a specific re-testing interval was written into the 1997 permit. The company appears to have treated the absence of an explicit re-testing mandate as permission to never retest, even after the 1995 test showed exceedance. The purpose of the emission limit, ensuring ongoing compliance, was undermined by the absence of any verification for 28 years.
• The 2021 MAERS report omitted kiln data. The requirement under Mich. Admin. Code r. 336.202 and AQD-013 is that once a facility triggers the reporting threshold for any pollutant, it must report for all listed pollutants and all emission units. Kilns are listed in the facility’s own permit. Their omission was not a gray area; it was a missing section of a required report.

Time as a Corporate Weapon: 28 Years Without a Test

Delay, in environmental enforcement, is not a neutral outcome. Every year without a test is a year in which pollution continues unchecked and accountability recedes further into a statute-of-limitations fog.

• The last voluntary emissions test on Boilers 2, 3, and 4 was conducted in 1995. The EPA conducted its compliance inspection in October 2021: a gap of 26 years before federal regulators set foot on the property to check.
• The NOV/FOV was issued in September 2022. The forced performance test occurred in August 2023. The final consent order was filed April 29, 2026. From inspection to final resolution took four and a half years. During that entire period, Boiler No. 4 continued operating.
• The CAFO notes that the EPA and the Department of Justice jointly determined the matter appropriate for administrative penalty assessment despite involving violations that occurred more than one year before the proceeding was initiated. That joint determination was required precisely because the delay between violation and action is long enough to raise procedural flags under the standard enforcement timeline.

Societal Impact Mapping: Who Absorbs the Cost of Dirty Air?

Public Health

The documented violations translate directly into a public health exposure that was preventable and legally prohibited.

• The facility’s own MAERS filings show PM2.5 emissions exceeding Michigan’s 10-ton-per-year reporting threshold since at least 2017. This means regulators have had five or more years of data showing the site was a significant particulate emitter, while the boiler generating those emissions was confirmed to be above its legal limit.
• PM2.5 penetrates deep lung tissue and is linked by decades of epidemiological research to respiratory disease, cardiovascular events, and premature death. The populations most vulnerable: children, elderly residents, people with pre-existing respiratory conditions, and outdoor workers in the facility’s airshed, had no way of knowing the facility’s emission controls were failing.
• The 2021 MAERS report omitted kiln emissions, meaning regulators and the public were working with an incomplete picture of the facility’s pollution output even in the most recent reporting cycle.

Economic Inequality

The structural inequalities embedded in environmental enforcement fall hardest on communities with the fewest resources to push back.

• Munising, Michigan is a small rural community in Alger County. Residents in areas like this typically lack the organized advocacy infrastructure, legal resources, and political access needed to compel enforcement action. The EPA’s inspection in October 2021 was not triggered by community complaint; it was a regulatory compliance inspection. The community was not a driver of accountability; it was simply exposed while the process worked on its own timeline.
• The $112,297 fine is paid to the federal government. None of it is directed to health monitoring, medical assistance, or environmental remediation for the people who lived and worked near the facility during the years of illegal emissions. The benefit of the settlement accrues to the regulatory system, not to affected residents.
• Workers at the facility itself, those most directly exposed to the particulate pollution from Boiler No. 4 and the kiln units, are not identified as a protected party in any provision of this settlement. Their occupational exposure during the exceedance period is not addressed.

The Settlement Isn’t Justice: What $112,297 Actually Buys

The CAFO resolves civil penalty liability for the two violations cited. It does not close the gap between what happened and what accountability should look like.

• Respondent “neither admits nor denies” the substantive violations. This is standard administrative settlement language, but its consequence is real: no public record establishes that the company admits it ran an illegal boiler for decades or that it omitted required data from regulatory reports. The legal resolution exists; the admission does not.
• The CAFO explicitly states it “resolves only Respondent’s liability for federal civil penalties for the violations specifically alleged.” Separate violations, future violations, and any state-level enforcement actions are outside its scope. It is a targeted liability release, not a comprehensive reckoning.
• The source document does not disclose the company’s revenues, profits, or the cost savings achieved by deferring dust control equipment for 28 years. Without that data, it is not possible to calculate a precise fine-to-profit ratio. What is documented is that the fine represents the assessed civil penalty; the economic benefit of avoided compliance costs over nearly three decades is not recovered or even estimated in the settlement.
• The new permit issued in April 2025 and the required installation of a multi-cyclone dust collector are compliance requirements that should have been triggered by the 1995 test result. They are happening 30 years late. The settlement formalizes future compliance obligations that were already legally required before this enforcement action began.

The Fine in Context

This Is the System Working as Intended

What happened in Munising is not a story about a system that failed. It is a story about a system that produced exactly the outcome its incentive structures predict.

• The 1997 permit established an emission limit but did not mandate a recurring testing schedule. A regulated entity with a financial interest in avoiding expensive boiler upgrades had a structural incentive to avoid voluntary retesting after an already-failed 1995 result. The rules created the space for that avoidance.
• Annual MAERS filings based on outdated test data satisfied the letter of the reporting requirement. Regulators received paperwork each year that looked like compliance data but was grounded in measurements taken before Bill Clinton’s first term ended. The system accepted it.
• The EPA conducted its first on-site inspection in October 2021, 26 years after the last test documented a violation. Federal enforcement resources are finite; small industrial facilities in rural Michigan are not the highest-priority targets. That triage logic, which is rational given resource constraints, means communities near lower-priority emitters absorb years of unverified pollution before anyone checks.
• The settlement penalty of $112,297 is the outcome of a legal process that is designed to be resolved through consent agreements rather than contested proceedings. The company cooperated, received credit for that cooperation per the CAFO’s own language, and paid a fine without admitting wrongdoing. This is the routine operation of administrative environmental enforcement, not an aberration.
• The new permit and required dust collector are the tangible remediation outcomes. They will reduce future emissions. They arrived 30 years after the problem was documented. The system got there eventually. Slowly. At community expense.

What a Legitimate Fix Looks Like

The core structural failure this case exposes: emissions permit limits are meaningless without mandatory, recurring, independently verified performance testing on a defined schedule. Without that, a permit is a ceiling that no one checks against.

The following recommendations are editorial analysis, not findings of the source document.

Regulatory Track

• Mandatory retesting intervals: The EPA and state environmental agencies should require that any permit condition establishing an emission rate limit for a combustion unit trigger a mandatory re-performance test on a defined schedule, not to exceed five years, and immediately following any permit modification or major equipment change. The 28-year gap in this case was made possible by the absence of such a requirement.
• Consequence for exceeded limits in prior tests: When a performance test documents an exceedance, state agencies and EPA should be required to issue a formal compliance order within a defined window, not merely note the result. The 1995 test should have triggered a mandated remediation timeline.
• MAERS data integrity audits: Michigan EGLE and EPA Region 5 should implement random-sample audits of MAERS filings to verify that reported emission factors are based on testing conducted within a defined lookback window and that all permitted emission units are included. Omissions like the kiln data gap in 2021 should be caught by the filing system, not discovered during an on-site inspection.

Legislative Track

• Close the self-reporting gap: Congress should amend the Clean Air Act to require that any regulated facility with permitted emission limits certify annually that its reported emission factors are based on testing conducted within a defined timeframe. Self-certification based on decades-old data should not satisfy the reporting requirement.
• Community notification requirements: Where a facility is documented to have exceeded permit limits, affected communities within the facility’s airshed should receive mandatory written notification within a defined period. Currently, the CAFO is a public document, but no affirmative notification obligation runs to residents of Munising.
• Penalty calculation transparency: Settlements should be required to disclose the estimated economic benefit of the violation to the respondent, enabling public scrutiny of whether the penalty actually eliminates the financial incentive for non-compliance.

Corporate Governance Track

• Internal compliance calendars with board-level accountability: Facilities operating under emission-limited permits should be required to maintain a documented compliance calendar, updated annually, identifying all performance testing due dates and the date of last test for each permitted emission unit. That calendar should be certified by a senior officer and retained for regulatory review.
• Proactive disclosure of known exceedances: When a performance test documents a violation of a permit condition, the operating entity should be required to self-report the exceedance to the permitting authority within 30 days and submit a corrective action plan. Sitting on a known exceedance for 28 years should not be a viable corporate strategy.

What Now? Here Is Where to Direct Your Energy

The company in this case is Timber Products Michigan Limited Partnership, operating as Timber Products Company, at Highway M-28 East, Munising, Michigan. The settlement was signed on behalf of the EPA by Carolyn Persoon, Division Director, Enforcement and Compliance Assurance Division, EPA Region 5. The final order was issued by Ann L. Coyle, Regional Judicial Officer, EPA Region 5.

Regulatory Watchlist

• EPA Region 5: The regional office with jurisdiction over Michigan. Air Enforcement and Compliance Assurance Branch: R5airenforcement@epa.gov. Track their enforcement docket for future actions involving this facility.
• Michigan EGLE (Department of Environment, Great Lakes, and Energy): The state agency that administers Michigan’s SIP-approved permitting program, including MAERS reporting and Permit to Install compliance. EGLE received a copy of the 2022 NOV/FOV and co-administers the rules violated here.
• EPA Office of Inspector General: If you believe enforcement resources are being systematically under-deployed in rural communities, OIG accepts complaints regarding EPA program effectiveness.

Grassroots and Mutual Aid Actions

• Request the full permit file: Under the Freedom of Information Act and Michigan FOIA, residents and advocates can request Timber Products Michigan’s complete permit history, all MAERS filings, and any EGLE correspondence related to the facility. Knowing what the regulators knew and when is the foundation of accountability.
• Connect with environmental justice organizations in the Upper Peninsula: Groups focused on Great Lakes environmental health and rural Michigan air quality advocacy are the organizational infrastructure through which community pressure on regulators gets built and sustained.
• Attend public comment periods on permit renewals: The 2025 PTI represents a new compliance baseline for this facility. Future permit modifications or renewals must go through a public process. Showing up to those proceedings, in writing or in person, is a documented and legally meaningful form of participation.
• Push your state legislators on MAERS audit requirements: The kiln omission and the decades-stale emission factors in annual reports point to a gap that EGLE has the administrative authority to close through enhanced audit protocols. State legislative oversight committees can pressure EGLE to act without waiting for federal action.