The Wrong Fight About Line Speeds

On February 18, 2026, USDA’s Food Safety and Inspection Service proposed permanently raising maximum line speeds at poultry slaughter establishments operating under the New Poultry Inspection System, from 140 birds per minute to 175. The pork rule went further, removing the line-speed cap entirely. The public comment period closed April 27 with more than 42,000 submissions. A final rule is pending.

The political fight is shaped almost entirely around line speed. The National Chicken Council says US plants need 175 to compete with Brazil and Belgium. UFCW says faster lines hurt workers. Most of the press coverage frames it the same way.

USDA’s own study says something different.

The Poultry Processing Line Speed Evaluation (PULSE) Study, conducted by UC San Francisco and posted in January 2025, examined musculoskeletal disorder risk in poultry workers across plants operating at the standard 140 bpm and at the 175 bpm waiver speeds. Its central conclusion was that increased evisceration line speed was not directly associated with elevated musculoskeletal disorder risk. Industry groups led with that finding. The administration cited it when announcing the rulemaking.

But the study also identified the variable that did track musculoskeletal risk: piece rate — the number of chicken parts handled per minute by an individual worker. Piece rate is a function of staffing, not line speed. A line running at 175 bpm with adequate staffing can produce a lower per-worker piece rate than a line running at 140 bpm staffed thinly. Forty percent of workers across all studied establishments reported work-related pain — at both speed tiers. What predicted which workers were at greater risk wasn’t which speed their plant ran at. It was how many parts per minute they personally handled.

Piece rate is not in the proposed rule. The rule lifts line-speed limits and explicitly removes the requirement that participating establishments submit worker safety data. FSIS notes in the rulemaking text that the agency “does not have statutory authority to regulate worker safety.” The lever the study identified as most predictive of harm sits outside the rule’s reach.

Checking the framing against enforcement data

The cross-agency dataset I published earlier this month (source: fastdol.com) covers 76,310 US employers with federal enforcement records from two or more agencies. Filtering to NAICS 311615 — poultry processing — produces 242 plants with 75 recorded fatalities. The fatality rate is roughly five times the multi-agency baseline and twice the broader food-manufacturing rate.

What’s striking in the joined data is not the fatality count alone but the National Labor Relations Board footprint that runs alongside it. Across the same 242 plants, there are 1,407 unfair labor practice and representation cases. These aren’t OSHA records. They’re filings from workers — through unions or directly — alleging retaliation, refusal to bargain, or interference with organizing.

The plants where workers are dying are also the plants where workers are filing the most labor-practice complaints. Two different agencies, different statutes, different filing mechanisms — and they land on the same plants.

Foster Farms is the cleanest example. Across four California plants, the company has 10 OSHA-recorded fatalities and 103 NLRB cases — roughly 25 NLRB filings per plant. For comparison, Tyson Foods, with more than four times as many establishments in the dataset, averages about 4 per plant. Pilgrim’s Pride, with 30 plants, averages 11. Foster’s ratio is an outlier even in an outlier industry.

That correlation is consistent with the PULSE study’s read of where the operative risk variable lives. Understaffed lines produce both the per-worker piece-rate pressure that drives injuries and the kind of management practices that generate unfair-labor-practice charges. The NLRB data doesn’t measure staffing directly, but it tracks the downstream consequences of it.

What the rule does not address

The proposed rule does not address staffing density. It increases the permitted speed of the line and ends the worker safety data submission that the previous waiver structure required. Whether or not the rule is finalized in its current form, the debate has been about the wrong variable. The agency’s own contracted study identified piece rate. The cross-agency enforcement data corroborates the staffing- pressure framing. The rulemaking engages with neither.

For workers’ comp underwriters, P&C analysts, and researchers covering food production: line speed alone is an incomplete predictor. The ratio of staff to processing throughput at a specific establishment is the variable that matters, and public data doesn’t reliably surface it. NLRB filings are the closest available proxy. Plants with elevated NLRB activity — holding throughput roughly constant — are the ones worth a second look.

The full poultry processing dataset will be available on Kaggle, Hugging Face, and Zenodo this Friday with the same parent-company rollup applied.

Methodology

76,310 US employers with federal enforcement records from two or more agencies (OSHA, WHD, MSHA, EPA, NLRB). Poultry subset filtered to NAICS 311615. Fatality counts reflect OSHA-recorded fatality investigations only. NLRB case counts include both unfair labor practice cases and representation cases. Findings reported descriptively. Not legal, financial, or underwriting advice.

Full methodology: fastdol.com/methodology.

Aggregate analysis of public federal enforcement data. Not a statement about any specific employer’s current operations or future risk.

FastDOL aggregates enforcement records from 17 federal sources into queryable employer profiles with entity resolution across agencies. Free tier: 50 lookups per month. fastdol.com/docs