The EPA didn't inspect 1,328 hazardous-waste sites in five years
Summary
An EPA Inspector General review of enforcement data for the country's largest hazardous-waste generators found that from 2020 through 2024, the EPA and states failed to inspect 19 percent of them even once -- 1,328 facilities producing tons of waste the agency itself calls capable of causing substantial health and environmental harm. The gap is worse in some states than others: Connecticut inspected just 36 percent of its large generators, Maine 46 percent. The report documents the shortfall in detail but makes no recommendations and requires no response from the EPA.
One in five of the riskiest waste generators, unchecked for five years
Large quantity generators, or LQGs, are the top tier of hazardous-waste producer under the Resource Conservation and Recovery Act⧉ (RCRA), the 1976 law that built the federal hazardous-waste regulatory system: any facility generating at least 1,000 kilograms of non-acute hazardous waste a month, or more than one kilogram of acute hazardous waste -- the kind that can be fatal to humans even in low doses⧉. Per 2023 biennial reports⧉, LQGs nationwide produced over 31 million tons of hazardous waste in 2022 alone.
The statute Congress wrote to govern this waste says plainly why the inspections matter: "the placement of inadequate controls on hazardous waste management will result in substantial risks to human health and the environment," and if that management "is improperly performed in the first instance, corrective action is likely to be expensive, complex, and time consuming."⧉
Nationally, the found the and authorized states cumulatively inspected 5,499 of the 6,827 LQGs reviewed -- 81 percent⧉, covering generators responsible for 94 percent of the non-acute hazardous waste and 95 percent of the acute hazardous waste in the dataset. That headline number reads better than the uninspected count underneath it: 1,328 facilities that met the bar for the country's strictest hazardous-waste tier went the entire five-year cycle without a single compliance check by anyone.
View data as table
| Connecticut | 36% | alternative inspection plan |
|---|---|---|
| Maine | 46% | standard plan; "area for improvement" |
| Illinois | 48% | alternative inspection plan |
| Texas | 55% | alternative inspection plan |
| New Hampshire | 58% | alternative inspection plan |
| Louisiana | 59% | alternative inspection plan |
| Massachusetts | 62% | alternative inspection plan |
| Maryland | 63% | alternative inspection plan |
| Arkansas | 64% | standard plan; "area for improvement" |
| Oklahoma | 66% | alternative inspection plan |
Where you generate hazardous waste changes your odds of ever being checked
The lets states pick between a standard inspection plan -- committing to inspect every LQG within five years -- or an -approved alternative plan that redirects some inspection resources elsewhere. Thirty-eight states chose the standard plan; fifteen of them actually hit 100 percent coverage⧉. The other 23 fell short, and two -- Maine at 46 percent and Arkansas at 64 percent⧉ -- fell far enough to land in the 's own "area for improvement" category, its lowest performance tier.
The 15 states running alternative plans did worse on average: 65 percent inspection coverage combined⧉, against 91 percent for standard-plan states. Connecticut inspected just 36 percent of its 146 LQGs⧉ -- the lowest facility-count coverage rate of any state or territory in the entire review, standard plan or alternative. Coverage isn't just about how many facilities get checked, either: Kansas inspected 82 percent of its LQGs by count, but those inspections covered only 8 percent of the state's hazardous waste tonnage⧉, because the one LQG Kansas never got to was its largest.
When the EPA leads an inspection, it hits harder than states do
The report also tracked what happens after an inspection finds a problem. From 2020 through 2024, inspectors ran 8,800 inspections of 5,452 LQGs, and 4,661 of those inspections -- 54 percent -- turned up at least one violation⧉. Agencies logged 23,404 violations and took 5,156 enforcement actions in response, but only 861 of those actions -- 17 percent -- were formal, the category that can carry a monetary penalty; 535 formal actions actually did.
View data as table
| EPA-led | 38% | of the EPA's 530 total enforcement actions nationally |
|---|---|---|
| State-led | 14% | of states' 4,567 total enforcement actions nationally |
| Jointly led | 44% | of 59 total jointly led enforcement actions nationally |
Who leads the inspection changes the odds of real consequences. The EPA directly led just 8 percent of all inspections nationally -- states led 88 percent, joint inspections the remaining 4 percent⧉ -- but -led inspections produced 23 percent of all formal enforcement actions and 28 percent of all penalties. EPA-led inspections carried a median penalty of $18,750, against $11,999 for state-led inspections⧉ -- about 56 percent higher. The doesn't inspect more; when it does show up, it treats what it finds as more serious than states typically do.
Serious violators face a tighter track: the 's own Hazardous Waste Civil Enforcement Response Policy⧉ says "significant noncompliers" -- chronic or recalcitrant violators, or those posing an actual health exposure risk -- should get a formal response within 360 days. 395 of the 5,452 inspected LQGs, about 7 percent, were designated significant noncompliers, and only 279 of those were resolved inside that window⧉. Nationally the resolution rate for that category runs 65 percent, but it swings by region -- EPA Regions 1, 7, and 10 fell below that average, while Region 8 resolved 100 percent of its cases in time⧉.
This isn't new, and nothing in the report requires it to change
This isn't the 's first time flagging inspection shortfalls. A prior OIG report from December 2021⧉ found LQG inspection rates by authorized states fell 47 percent during the first year of the COVID-19 pandemic, after the itself suggested states swap in-person inspections for off-site reviews. That earlier report carried five recommendations, and the closed out corrective action on all of them. This one carries none: the states outright that its "goal for this report is to share information about enforcement trends," not to "identify the root causes of observed trends" or make recommendations, and that "a response to this report is not required."⧉
The 's own fiscal year 2025 budget for running the entire RCRA program was about $121 million, plus $97.5 million in grants passed through to authorized states⧉ to run their own pieces of it -- the same funding structure, split the same way between direct federal oversight and state delegation, that produced this five-year gap.
- 1,328 of 6,827 large hazardous-waste generators -- 19 percent -- were never inspected at all between 2020 and 2024, despite an goal of inspecting every one at least once in that window.
- Where a facility sits changes its odds of ever being checked. Connecticut inspected just 36% of its LQGs, the lowest of any state reviewed; Maine and Arkansas landed in the 's own "area for improvement" tier at 46% and 64%.
- The inspects far less than states do, but hits harder when it shows up: 8% of inspections, 23% of formal enforcement actions, 28% of penalties, and a median penalty 56% above the state-led median.
- The documented all of this without making a single recommendation. No corrective action is required, and no one is on the hook to close the gap the report describes.
Figures are drawn from the 's full 27-page report (No. 26-E-0025, April 28, 2026), not only its two-page "At a Glance" summary, to source the state-by-state tables and the -led/state-led enforcement comparison. The report explicitly states it does not identify root causes for the patterns it documents and that state-level enforcement comparisons require deep contextual understanding of each state's resources and industrial composition -- this piece presents the 's own findings and cautions as given, without asserting causes the report itself declines to assign. California is entirely absent from the dataset (excluded by the for a known data-quality issue in 's own tracking system), so its LQGs are not represented in any national or state-comparison figure here. Separately, the report's own narrative names Regions 1, 7, and 10 as falling below the national 65% significant-noncomplier response-rate average, though its own regional appendix table appears to show two additional regions also below that mark -- an inconsistency in the source document itself, reproduced here as the report states it rather than independently reconciled.
Sources(3) ▾
- U.S. Environmental Protection Agency, Office of Inspector General, Evaluation of Trends in Resource Conservation and Recovery Act State-Level Enforcement Data (Report No. 26-E-0025, full report) (2026-04-28) — The 's full 27-page evaluation of state-level RCRA hazardous-waste inspection and enforcement data for large quantity generators (LQGs), 2020-2024. Fetched directly from epa.gov and read page-by-page. Used for every page-anchored figure, table, and quotation in this piece: the national inspection-coverage totals, the state-by-state tables (Table 1, standard inspection plans; Table 2, alternative inspection plans), the national enforcement-outcome statistics (Table 3), the -led vs. state-led enforcement comparison (Table 4), the RCRA budget figures, and the report's own scope note that it makes no recommendations. A Wayback Save Page Now request for this URL succeeded on 2026-07-17. epa.gov · original document
- U.S. Environmental Protection Agency, Office of Inspector General, Evaluation of Trends in Resource Conservation and Recovery Act State-Level Enforcement Data -- At a Glance (Report No. 26-E-0025, 2pp. summary) (2026-04-28) — The 's own two-page executive summary of the same report, released the same day. Its 'What We Found' paragraph states the top-line -led-vs-state-led enforcement comparison ( led 8% of inspections but produced 23% of formal enforcement actions and 28% of penalties, with a higher median penalty) in a single condensed passage that matches and confirms the detail given in the full report's Table 4 and surrounding text. Used here as a corroborating source for that headline comparison. epa.gov · original document
- Cornell Law School, Legal Information Institute (U.S. Code), 42 U.S.C. § 6901 -- Congressional findings (Resource Conservation and Recovery Act) (1976-10-21) — The founding congressional findings section of RCRA, the 1976 law that created the hazardous-waste regulatory scheme this report evaluates. Used to source the statute's own stated purpose -- protecting health and the environment from improperly managed waste -- as the baseline against which the 's inspection-gap findings are read. law.cornell.edu · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
Every large quantity generator of hazardous waste in the country -- the manufacturers, hospitals, and industrial sites producing enough dangerous byproduct to fall under the 's strictest reporting tier -- is supposed to be inspected at least once every five years, per the EPA's 2021 Compliance Monitoring Strategy for the RCRA Subtitle C Program⧉. The EPA's own Office of Inspector General⧉ checked whether that actually happened for 2020 through 2024, across 6,827 large generators in 53 states and territories. It found 1,328 of them -- 19 percent -- were never inspected at all in that five-year window.