Earlier this week, Tyson Foods indicated that it will soon replac over a dozen federal food safety inspectors in its Holcomb, Kansas beef plant with company employees, pursuant to regulatory waivers that will also allow the company to raise line speeds. However you feel about the current beef inspection system, this is the wrong way to reform it.

USDA has issued its waivers pursuant to a rule whose purpose is “to permit experimentation so that new procedures, equipment, and/or processing techniques may be tested to facilitate definite improvements.” 9 CFR s. 303.1(h). The waivers apply to the Holcomb plant’s regulatory obligations in areas including inspection staffing, handling of bruised parts, line speed, and microbiological testing. In news reports, Tyson has indicated that it intends to employ “vision systems and machine learning in beef carcass inspection” at the Holcomb plant. Tyson’s waiver application, however, makes no mention of such technologies, at least not in portions of that document available to the public. Rather, it argues that waivers should be available to allow cattle slaughter establishments to operate under an inspection system pilot—“HIMP”—that USDA developed for poultry and swine slaughter establishments in 1997.

“HIMP” stands for HAACP-based Inspection Models Project; HAACP for Hazard Analysis Critical Control Point.

USDA has expanded this acronym within an acronym pilot inspection program for poultry and, under the Trump Administration, it has sought to do the same for swine slaughter inspection, although its rulemaking has been challenged in court. Labor advocates have opposed the HIMP program because increased line speeds make workers unsafe. Unsafe workers tend to make unsafe food.

Food safety advocacy groups like mine have also opposed HIMP expansions because USDA has failed to set up a scientifically valid experiment to evaluate how the inspection reforms will actually affect food safety. In the case of swine slaughter, because USDA has no applicable Salmonella performance standards, it has lacked a critical yardstick by which to make an apples-to-apples comparison of HIMP versus traditionally inspected plants.

This measurement problem is even more exacerbated in the case of the Holcomb plant’s regulatory waivers. Intuitively, removing government inspectors and line speed caps will negatively affect food safety, all else equal; USDA has conceded as much. But Tyson and USDA have done little to assure consumers that the waivers will yield the “definite improvement” required by USDA regulations. Indeed, Tyson’s waiver application claims only that the “requested waivers will not adversely affect product safety.”

Unfortunately, this has become a common refrain at USDA. The agency justified its recent swine slaughter inspection rule not as an improvement to food safety, but as a way to boost efficiency without a “significant” impact on food safety. The same rationale has motivated line speed waivers for poultry plants, with Under Secretary Mindy Brashears recently heralding in this publication a study that found “the presence of Salmonella or other indicators of process control, such as non-compliance records for regulations associated with process control and food safety, are not significantly increased in establishments with higher line speeds.” (emphasis added)

There are reasons to doubt whether USDA’s reforms are so benign, not least of which is the agency’s reluctance to collect and share data about the performance of plants participating in its New Poultry Inspection System and hog HIMP pilot, and the fact that the sole author of the study on poultry line speeds, Louis Anthony Cox, Jr., is a notorious industry shill, as explained in David Michael’s book The Triumph of Doubt: Dark Money and the Science of Deception. As Michaels explains:

Not just his science, but Cox’s integrity was called into question by the FDA following his efforts on behalf of Bayer to defend the use of an antibiotic in poultry production that the FDA believed would increase the development and spread of antibiotic-resistant campylobacter infections in humans. Taking an action that almost never occurs, in 2005, President George W. Bush’s FDA commissioner actually excluded Cox’s testimony from the proceedings; the agency found that he “intentionally misquoted published articles,” and “Dr. Cox’s credibility was such that his testimony was so unreliable that it was inadmissible.”

Yet, even assuming for the sake of argument that the agency is correct, that waiving regulatory requirements leaves us more or less as safe as we were before, the consequences for consumers are still untenable.

In the United States, progress on reducing foodborne illness has largely stalled in recent years. In fact, the most recent CDC data shows an upward trend in reported foodborne illnesses caused by pathogens including Campylobacter, for which USDA suspended verification testing in poultry in 2018, and Shiga toxin-producing E. coli (STECs), for which cattle are the primary reservoir. New culture-independent diagnostic testing (CIDTs) may account for some of the increase in reported foodborne illness cases, but CDC researchers have made clear that “identification of infections that might not have been detected before the adoption of CIDTs cannot explain this overall lack of progress.”

Fortunately, we know how to get out of this rut, particularly when it comes to meat and poultry. As the industry is fond of pointing out, raw meat is not sterile. Despite the antimicrobial sprays applied in the slaughter and processing facilities, low levels of pathogens persist. Sometimes not so low levels persist. Because not everyone follows safe handling and cooking instructions with the precision of a NASA scientist, and not everyone has the immune system of LeBron James, people get sick. To make meat and poultry safer for consumers, the industry needs to do more on the farm to reduce the food safety risks. This is not controversial. USDA’s own Dr. Brashears has written that “[p]re-harvest reduction of E. coli O157 colonization will require targeted intervention strategies and should reduce contamination of carcasses thereby enhancing public health.” Yet USDA offers precious few incentives for industry to invest in these “pre-harvest” strategies.

That needs to change, ideally with legislative reform. Even under the current law, however, USDA has ample flexibility to make meat and poultry safer. Instead of simply asking companies to maintain the status quo in exchange for regulatory waivers, why not require them to adopt better food safety practices, and to demonstrate progress with transparent, quantitative criteria? Why not require companies to make public the information that they submit in support of waiver applications? USDA officials are quick to claim that they lack authority to go on farm, or that they cannot divulge “confidential” information, but companies do not have to participate in these “experimental” programs. So while USDA cannot require a company to say, vaccinate cattle in feedlots for E.coli—a practice that, incidentally, would make our lettuce safer as well—USDA can take into account a company’s vaccination program when it determines whether that company’s plant is a good candidate for a regulatory waiver. And it can tell companies that anything they submit to USDA in support of a waiver application will be disclosed to the public.

Almost always, a more transparent food system is a safer food system. Unfortunately, however, we are very short of what most knowledgeable, disinterested observers would consider an optimal level of transparency in our food system. One need only glance at the copy of Tyson’s regulatory waiver application made available to the public—blackened with extensive redactions—to appreciate that fact. But it doesn’t have to be this way. Safer meat and poultry is achievable, we just need leadership that is willing to stand up to industry and ask for it.