Why most OSHA PELs still reflect 1971 science

This page explains what the rules and common practice call for. Every site is different, and what a survey or a filing needs at your facility is something we work out with you after we see it.

In 1992 a federal court struck down OSHA's only broad attempt to update its air contaminant limits, and the agency has never tried again at that scale. Everything about how US exposure limits work today traces back to that decision.

Meeting a PEL settles your compliance position for that substance, but it does not tell you much about whether the exposure is safe. This page explains the difference and where it comes from.

The short history

Most OSHA PELs date from 1971. OSHA adopted them in bulk, under the start-up authority of the OSH Act, from voluntary standards already in use. As a result, much of what is enforceable reflects the science of the 1960s.

OSHA knew this. In 1989 it issued the Air Contaminants Rule, which did two things:

• Set more protective PELs for 212 substances that already had one.
• Set new PELs for 164 substances that had no federal limit at all.

That made 376 chemicals in total.

In July 1992, a federal appeals court vacated the rule (AFL-CIO v. OSHA, 965 F.2d 962, 11th Cir. 1992). The court's objection was procedural. The law requires findings for each substance individually, and the agency had not made them because it tried to address 376 at once.

The consequences were as follows:

• The 212 revised limits reverted to their 1971-era values.
• The 164 new limits were eliminated entirely, and those substances went back to having no federal PEL.
• OSHA began enforcing the old values again on June 30, 1993.

The NIOSH Pocket Guide lists all 376 in Appendix G, "1989 Air Contaminants Update Project: Exposure Limits NOT in Effect."

In other words, a federal agency maintains a list of limits that the government decided were right and then lost the power to enforce.

What is enforceable today

The table below also refers to ACGIH, the professional body that publishes the threshold limit values (TLVs), and to Cal/OSHA, California's state OSHA program.

So 16 agents have modern, complete standards, while several hundred substances are still on 1971 numbers. That is the shape of the problem.

The annotated tables are worth knowing about. OSHA publishes them because it finds the comparison useful, and it states on the tables themselves that the added values carry no enforcement weight. They are the quickest way to see, substance by substance, how far a federal PEL sits from the current recommended number.

Worked example: manganese

This is the clearest case, and it affects any welding, foundry or steel fabrication shop.

In the table, mg/m³ means milligrams per cubic meter of air, and a ceiling limit is one that must not be exceeded at any time. The inhalable fraction is all the particles that can be inhaled, while the respirable fraction is only the part fine enough to reach deep into the lungs.

OSHA's 5 mg/m³ against ACGIH's respirable 0.02 mg/m³ is a 250:1 ratio, but those two numbers measure different things. One is an instantaneous ceiling on total particulate, and the other is an eight-hour average of the respirable fraction.

So take the closer, like-for-like comparison. Against ACGIH's inhalable 0.1 mg/m³, OSHA's number is still 50 times higher. NIOSH sits between the two, at 1 mg/m³ as a shift average with a 3 mg/m³ STEL, which is five times below the PEL.

However you frame it, the conclusion holds. ACGIH lowered its limit because of harm to the nervous system, and the research behind it is not thin.

→ Welding fume exposure monitoring · → OSHA PEL vs ACGIH TLV

What this costs you in practice

General Duty Clause risk, but less of it than you might expect. The General Duty Clause, Section 5(a)(1) of the OSH Act, requires employers to provide work free from recognized hazards likely to cause death or serious physical harm. Where a PEL applies and you meet it, that route is largely closed.

OSHA's Field Operations Manual (Chapter 4) says the clause "shall normally not be used to impose a stricter requirement than that imposed by the OSHA standard."

A separate OSHA regulation adds that compliance with a standard counts as compliance with the General Duty Clause for the condition that standard covers.

A citation below the PEL is rare. It requires evidence of actual employer knowledge that the standard was inadequate to protect workers from death or serious physical harm. It also goes through review by OSHA's Directorate of Enforcement Programs and the Office of the Solicitor before it can be issued. The published gap between enforceable and recommended limits does not meet that bar by itself, and that holds for the annotated tables and for Appendix G.

So the gap matters in other ways: for the two items below, and for what the numbers say about health risk as opposed to citation risk.

Lawsuits. Workers' compensation is generally the only remedy a worker has against the employer, but that does not shield everyone in the chain. In court, "we complied with a 1971 limit" is a weak answer when the health-based limit was revised in 2013 and appears in the reference your industry uses.

Substances with no PEL at all. The 164 limits that were eliminated left real gaps. Where there is no PEL, there is no compliant number to hit, and the obligation comes entirely from the General Duty Clause. Where respirators are in use, it also comes from OSHA's respiratory protection standard.

What you should do

Nothing extra in the field. This is the practical point, and it is worth stating flatly: the sample is the same sample. It is the same pump, the same filter or tube, the same laboratory and the same analysis. What differs is which limit you compare the result to.

Comparing against both the PEL and the current recommended limit costs nothing and tells you far more about where you stand. A good report does both. The PEL comes first because it determines your legal obligations and your monitoring frequency, and the recommended limits belong alongside it where there is a meaningful gap, flagged as informational and not as legal requirements.

Confusing the two numbers is how plants end up either complacent or alarmed for no reason, so know both.

→ Objective data and exposure assessment · → How to read an industrial hygiene report · → Air monitoring surveys