An appellate court one level below the Supreme Court of the United States, and highly respected in the field of administrative law, recently held that the Occupational Safety and Health Administration’s (OSHA) interpretation of a statute of limitations is wrong, contrary to its “clear” language, “unreasonable,” productive of “absurd” consequences, and that it would be “madness” for OSHA to attempt to avoid the court’s ruling by amending its regulations. So what does OSHA do? Instead of appealing to the full court of appeals en banc, bringing a test case to another court of appeals, seeking review by the Supreme Court, or asking Congress to amend the statute, OSHA amends its regulations with the aim of achieving the same result that the court condemned as “absurd.”
This behavior deserves to be reined in. Fortunately, the U.S. House of Representatives took a crucial step to doing that when, on March 1, 2017, it passed a resolution of disapproval under the Congressional Review Act and sent the resolution to the U.S. Senate.
In 2006, OSHA issued to Volks Constructors a citation alleging that, contrary to OSHA’s recordkeeping regulations, certain injuries had not been recorded on its logs. The practical problem for the company (and for me as its attorney) was that nearly all of the allegations were stale by years—many by nearly five years. Memories had faded, documents were lost or scattered, and, worst of all, the company’s previous recordkeeper had died, making it practically impossible to defend in any economical or fair way. For example, according to OSHA’s definition of “restricted” (in 29 C.F.R. § 1904.7(b)(4)(ii)), we’d have to reconstruct the movements of particular employees during certain weeks as far back as five years to determine whether they had then regularly, for example, climbed a ladder more than once a week. We therefore asserted as a defense that the citations were issued after the six-month statute of limitations for OSHA violations had run.
OSHA took the position that, inasmuch as it had a regulation requiring that injury logs not be discarded for five years, it could issue a citation for up to five and a half years after a failure to log an injury. OSHA’s lawyers relied on a 1993 decision by the independent Occupational Safety and Health Review Commission that had upheld that so-called “discovery rule” theory. They must have known, however, that that decision rested on thin ice, for in 1994 the U.S. Court of Appeals for the District of Columbia Circuit had issued a decision in 3M Co. v. Browner, 17 F.3d 1453 (D.C. Cir. 1994), pointedly disapproving of the “discovery rule” theory on which the 1993 Commission decision rested. So OSHA’s lawyers switched theories in mid-litigation, this time to a “continuing violation” theory. They won on that theory before the Review Commission, and Volks appealed.
In AKM LLC dba Volks Constructors v. Sec’y of Labor, 675 F.3d 752 (D.C. Cir. 2012), the D.C. Circuit reversed. It relied on Supreme Court case law construing a nearly identical statute and held that the six-month statute of limitations for OSHA cases meant what it “clearly” said—that OSHA must issue a citation within six months of a violation, not within five and a half years. The court observed that OSHA’s view would lead to the “absurd consequence” that the limitations period “could be expanded ad infinitum if, for example, [OSHA] promulgated a regulation requiring that a record be kept . . . for as long as [OSHA] would like to be able to bring an action based on that violation. There is truly no end to such madness.” The court also stated: “Nothing in the statute suggests Congress sought to endow this bureaucracy with the power to hold a discrete record-making violation over employers for years.”
Instead of seeking further review from the D.C. Circuit or the Supreme Court, or bringing another case in another circuit, OSHA did what the court said it could not do—it tried to reach the same rejected result by merely amending its regulations. The amended regulations purported to “clarify” that the duty to have correct logs continued throughout the five-year retention period.
The major flaw with that approach is that it rests on a fiction—that the D.C. Circuit’s ruling had been based on a misunderstanding of OSHA’s regulations. In fact, the ruling did not rest on OSHA’s regulations at all. It rested on the wording of an act of Congress, which OSHA could not change by administrative fiat. OSHA decided to follow instead a minority opinion in the Volks case, the concurring opinion of Judge Garland. The concurring opinion reasoned that the wording of OSHA’s regulations did not expressly impose a continuing duty to keep logs accurate for the five-year retention period.
OSHA’s decision to amend its regulations also created at least two practical problems, neither of which it acknowledged. First, the amendment would have perpetuated the staleness problem presented by what would, in effect, be a five-and-a-half year statute of limitations. OSHA reasoned that staleness would not be a practical problem because the facts would be shown by medical records, not testimony. In doing so, OSHA ignored rulemaking comments filed by the National Federation of Independent Business (NFIB) (for which I was counsel) that disproved that assertion. We showed that several common but key recordkeeping facts (such as those involving work restrictions, work-relatedness, and the purpose of certain medical tests) are almost never addressed by medical records; establishing them would require employers to use fallible memories to recall events years in the past.
Second, OSHA ignored the fact that, by amending its regulations to impose a continuing duty to record, it necessarily imposed on employers a duty to continuously examine injury logs for errors and omissions. The concurring opinion by Judge Garland, on which OSHA otherwise rested so heavily, reasoned that a requirement to update a stored log would “obligate an employer to constantly reexamine injuries and illnesses.” The NFIB’s comments showed that the burden on the economy of a daily duty of reexamination would be immense—up to $2 billion for a single unrecorded case—all for what OSHA itself estimated would cause only a 1 percent increase in compliance. The amendment was therefore either an attempt to effectively extend a statute of limitations without owning up to its consequences or an unjustifiable imposition of massive burdens on employers and the economy for little gain.
Supporters of OSHA’s position appear to be purveying something of a myth in its defense—that OSHA’s policy had gone unquestioned for 40 years. This is untrue. First, the term “policy” gives the matter more solidity than it deserves. OSHA was following just a prosecution practice, one not based on the words of any statute or regulation. Worse, the practice rested on a legal theory (the so-called “discovery rule”) that had been wrested from a different and inapplicable legal field—medical malpractice law. Not only was OSHA’s reliance on the discovery rule challenged in several cases in the early 1990s, but OSHA lawyers knew in early 1994, when the D.C. Circuit emphatically declared the theory inapplicable to civil penalty cases, that they could no longer rely on it. And in 2002, the Supreme Court in the Morgan case held that a statute of limitations using nearly identical wording did not support a continuing violation theory either.
In fact, there was never a realistic chance that another court of appeals or the Supreme Court would uphold OSHA’s view that it could cite employers for as long a period as its regulations required that an injury log be retained, plus six months. The reason is simple: Judges intensely dislike the idea that agencies may effectively rewrite statutes of limitations by manipulating the wording of their regulations. OSHA’s lawyers told the D.C. Circuit in the Volks case that if OSHA extended or abolished its retention period, the statute of limitations would be effectively extended or abolished with it. Unsurprisingly, the Volks court called that idea “madness.” Add to that the Supreme Court’s holding in Morgan that nearly identical language in a statute of limitations did not support a continuing-violation argument, and OSHA’s theory was destined to lose. Now with the U.S. Court of Appeals for the Fifth Circuit in the Delek decision having recently embraced the statute of limitations rationale of the Volks decision, the Senate should follow the House’s lead and put this issue to bed once and for all.