FastDOLz/cross-agency-federal-violations
Cross-Agency Federal Violations This is a mirror. Canonical home: https://www.fastdol.com/datasets/cross-agency-federal-violations License: CC BY 4.0 Version DOI: 10.5281/zenodo.20031853 Concept DOI: 10.5281/zenodo.20031852 Download CSV: https://www.fastdol.com/datasets/cross-agency-federal-violations/data.csv Visit the canonical page for the full schema, methodology, BibTeX citation, and most recent version. Cross-Agency Federal Violations Database US employers… See the full description on the dataset page: https://huggingface.co/datasets/FastDOLz/cross-agency-federal-violations.
Cross-Agency Federal Violations
This is a mirror. Canonical home: https://www.fastdol.com/datasets/cross-agency-federal-violations
License: CC BY 4.0 Version DOI: 10.5281/zenodo.20031853 Concept DOI: 10.5281/zenodo.20031852 Download CSV: https://www.fastdol.com/datasets/cross-agency-federal-violations/data.csv
Visit the canonical page for the full schema, methodology, BibTeX citation, and most recent version.
Cross-Agency Federal Violations Database
US employers cited by two or more federal agencies — workplace safety, wage theft, environmental, labor, and more — joined into a single dataset.
What this dataset is
This dataset contains 76,310 US employers with federal enforcement records from two or more agencies, joined into a single row per employer. The goal is to surface cross-agency patterns that single-source searches miss.
Each row aggregates an employer's accumulated record across OSHA (workplace safety), WHD (wage and hour), MSHA (mine safety), EPA (environmental), and NLRB (labor relations), plus a curated parent-company rollup, FastDOL risk scoring, and SAM.gov debarment status.
Source: fastdol.com.
What's in the data
- Employer identity: name, city, state, ZIP, NAICS classification, parent company
- OSHA: inspection count, violations, penalties, fatalities, severe violator program flag
- WHD: wage cases, total back wages, employees affected
- MSHA: mine safety violations and assessed penalties
- EPA: inspections, formal actions, non-compliance quarters, penalties
- NLRB: total cases, ULP cases, representation cases
- FastDOL artifacts: risk tier (LOW / MEDIUM / ELEVATED / HIGH), risk score, agency violation count, peer percentile
- Debarment: SAM.gov exclusion flag
Findings to explore
The dataset surfaces patterns at the employer level that aren't visible in single-agency searches.
Cross-agency stacking matters. The number of agencies citing an employer correlates almost linearly with worker fatality rates. Employers cited by five federal agencies are roughly 4x more likely to have a fatality on record than those cited by one.
Specific agency combinations are stronger signals than counts alone. Employers with records at OSHA, EPA, and NLRB simultaneously — just 1,526 employers in the dataset — have a worker fatality rate of 13.4%, more than 4x the base rate.
Parent-level rollup reveals hidden risk. 96 parent companies in the dataset have worker fatalities recorded at five or more of their locations. Across all locations operated by these 96 parents (~25,000), the fatality rate is 5.0% vs. the 3.0% base rate.
EPA non-compliance tracks workplace safety. Employers out of compliance with the EPA for four or more quarters have a 7.8% fatality rate vs. 3.0% for everyone else.
Suggested use cases
- Workers' comp and casualty underwriting (cross-agency exposure)
- Investigative journalism on multi-agency violators
- Academic research on cross-agency enforcement
- ESG analysis of corporate compliance footprints
- ML feature engineering using risk scores and agency counts
- Policy research on federal enforcement gaps
Loading the data
When loading in pandas, read naics_code as a string to preserve formatting:
import pandas as pd
df = pd.read_csv('multi_agency_violators.csv', dtype={'naics_code': str})Methodology
Federal enforcement data aggregated from OSHA IMIS, WHD WHISARD, MSHA, EPA ECHO, NLRB case database, and SAM.gov. Entity resolution uses normalized employer name, state, and ZIP. Parent-company rollup uses a curated seed table augmented with OSHA ITA company-name data. Stratified sample: all multi-source employers plus a 20% sample of single-source employers.
Counts reflect what federal agencies investigated and recorded; reporting practices vary by operator and industry.
Findings reported descriptively. Aggregate analysis of public federal enforcement data; not a statement about any specific employer's current operations or future risk. Not legal, financial, or underwriting advice.
About FastDOL
FastDOL aggregates federal enforcement records from 15 US agencies into queryable employer profiles with entity resolution.
- Source: fastdol.com
- Methodology: fastdol.com/methodology
- API: fastdol.com/docs — free tier available
- Mirror on Kaggle: Cross-Agency Federal Violations Database
Citation
FastDOL (2026), Cross-Agency Federal Violations Database, fastdol.com