Professionalism/Kenneth Kendrick and Peanuts

Kenneth Kendrick is the whistleblower in the 2009 salmonella outbreak from contaminated peanuts. Kendrick disclosed information about factory conditions at the Peanut Corporation of America that prompted the FDA to recall over 3,600 peanut products.

Background
From September 2008 to April 2009, a salmonella outbreak occurred across 46 states, resulting in 714 illnesses and 9 deaths. The outbreak was first traced back to peanut butter produced in a Georgia facility owned by the Peanut Corporation of America (PCA). After whistleblowing efforts by Kenneth Kendrick, the outbreak was linked to unsanitary conditions at PCA plants in both Georgia and Texas, resulting in a recall of over 3,600 separate peanut products, including those of large-scale manufacturers such as Keebler and Whole Foods.

In 2006, Kendrick worked as an assistant manager for the PCA plant in Plainview, Texas. During his four months there, he used the Texas Department of Health anonymous reporting service to cite rat and bird feces contaminating the peanut products, but his concerns were ignored. In 2009, after hearing reports about the Georgia PCA plant, Kendrick decided to publicize his case. Working with the Government Accountability Project (GAP) and STOP Foodborne Illness, Kendrick told his story on Good Morning America. Because of Kendrick's efforts, Food and Drug Administration (FDA) investigators were able to tie salmonella cases to PCA products from both plants, prompting more product recalls.

Case Outcomes
Kendrick has reported many personal repercussions from whistleblowing, stating "Me coming forward has pretty much ruined my life." His difficulties have have ranged from professional, including being fired from his job in 2009, to personal, including depression and the dissolution of his marriage from the stress of public scrutiny. Kendrick's experience is similar to that of other whistleblowers, many of whom, according to GAP, are negatively affected despite "putting ethics into practice."

Beyond Kendrick, the case had ramifications for PCA. PCA, which had about 300 employees, went bankrupt in February of 2009. They have also been the target of multiple civil suits since the outbreak.

Food Safety and Modernization Act
In early 2011, the FDA's Food Safety and Modernization Act was signed into law following a series of deadly food-borne illnesses, including the 2009 salmonella outbreak. It was designed to better enable the prevention of food-related illness through easier recalls and mandatory safety standards for both produce and processed foods. A smaller clause furthers this objective by granting increased protection to whistleblowers. Under the new law, employees are not allowed to be fired or denied promotions for reporting unsafe conditions and the burden of proof shifts from whistleblowers to employers in cases of wrongful termination. Critics have stated the law may lead to general complaints instead of reliable reports of violations; however, this remains to be seen after rules take effect in spring of 2014.

Litigation
Investigations into PCA led to the indictment of four PCA officials in early 2013. After FDA inspections of both PCA plants found unsafe conditions, the Federal Bureau of Investigation joined the case. Investigators found multiple instances of criminal activity, including obstruction of justice, conspiracy, and fraudulent products introduced to interstate commerce. On multiple occasions, PCA officials falsified safety certificates and sold products after they had tested positive for salmonella. These findings led to 75 counts of indictment, including felony charges, which could give PCA officials up to 20 years in prison.

Peanut Corporation of America
Founded in 1977, the Peanut Corporation of America supplied peanuts, peanut butter, peanut meal, and peanut paste for large snack distributors, including Kellogg, Sara Lee, Little Debbie and others, as well as the government, for schools, disaster relief, and the military. PCA sustained strong profits by buying "the cheapest peanuts they could find", relying on minimum wage labor and cutting costs wherever possible. The CEO and owner of PCA, Stewart Parnell, had grown up in the low-cost peanut industry and ran PCA's three processing facilities in Plainview, Texas; Suffolk, Virginia; and Blakely, Georgia from a converted garage next to his home in Lynchburg, Virginia. The company had a history of food quality issues, repeatedly failing the inspections of prospective buyers and facing a lawsuit in 1990 after regulators found unacceptably high levels of aflatoxin, a toxic mold, in a shipment to the American Candy Company PCA officials repeatedly shipped product without waiting for confirmation of salmonella lab results, falsified certificates of analysis, and lied to customers and the regulatory agencies about their test results and quality protocols, even after the nationwide outbreak of salmonella. They also failed to register the Texas plant with both national and state regulatory agencies, avoiding routine health inspection and remaining under the radar of government scrutiny.

Food and Drug Administration
A 1990 Food and Drug Administration inspection revealed that a PCA shipment had been contaminated with aflatoxin, a toxic mold, resulting in a recall and lawsuit. Eleven years later, a Nov 15, 2001 FDA inspection again finds aflatoxins in PCA peanuts, noting “ill-repaired equipment” and other concerns at the plant.> The FDA returned to inspect the Blakely plant in June of 2008 after a Canadian distributer refused a PCA shipment containing metal shipments; however, all concerns were corrected on site. In January of 2009, more than four months after the beginning of the salmonella outbreak, the CDC found a contaminated container of King Nut peanut butter sourced from the Blakely, GA PCA plant and the FDA began their investigation which would eventually expand to all three PCA facilities and revealed PCA’s unlawful and unethical practices.

State Regulatory Agencies
Critics argue that the PCA salmonella outbreak demonstrates a failure of state regulatory agencies, which play a major role in registering food producers and inspecting facilities. Salmonella contaminated products managed to make it to market from both the Georgia and Texas facilities, demonstrating a failure of state regulatory bodies both in the rigor of their inspections and their capacity to keep track of food producers.

Georgia Department of Agriculture
The Blakely plant inspections conducted by the Georgia Department of Agriculture (GDA) from 2006 though 2008 (the period when the contaminated products were processed) revealed no major issues at the plant. In fact, Former employees describe the plant as “filthy and nasty” and yet the GDA repeatedly rated the facility as “exceeds expectations/excellent” and “superior”. With approval from the state regulatory agency, PCA continued uninhibited until the FDA investigation forced them to seize operations.

Texas Department of Health and Department of Agriculture
The Texas plant, where Kendrick briefly worked, was operating completely under the radar of state regulatory agencies. The Plainview facility received organic certification from the Texas Department of Agriculture but never registered as a food producer with the Texas Department of Health(TDOH). As a result the plant went four years without regulatory inspection. After the PCA incident the TDOH conducted a statewide investigation and found 355 food producers who also had been operating without the proper licensing in Texas.

Ethics Considerations
The PCA salmonella outbreak resulted from a system-wide failure. This US food safety system is a complex network of players aimed at keeping consumers safe from tainted food. Several cognitive biases and other effects contributed to the failure of the system. While many may have been in play, this case exemplifies risk compensation, the bystander effect, and the tuneout effect.

Risk Compensation
The consequences or risks from selling salmonella infected peanut butter are obvious: people may be harmed and the company will lose sales and possibly go bankrupt. The purpose of regulation, like the FDA’s, may be viewed as increasing the consequences by imposing further penalties for risk taking. This nudges a company to assume less risk. Yet the more pernicious effect of risk compensation is likely present. Risk compensation is a well-accepted theory that when the consequences of behaving riskily are reduced, risky behavior increases. People have some sort of internal risk assessment that is adjusted when the variables change. In the PCA case, because the regulated corporation knew it had the safety net of the regulators, it stopped policing itself as well. The risk to the public of receiving contaminated food wound up close to where it started, and the effectiveness of the regulators in the system was reduced. It can be argued that Kendrick himself displayed risk compensation when he learned Nestlé had refused PCA peanuts. He assumed other companies would follow Nestlé so that even if he left PCA, the public would be safe.

Bystander Effect
When a system has many players, sometimes each player pulls a little less weight than would normally be expected. This is known as the bystander effect, also known as diffusion of responsibility. In the PCA case, the abundance of regulators (the Texas Department of Health, Texas Department of Agriculture, and FDA) encouraged each to act as a bystander. Each regulator may have assumed the others would pick up the slack. Rather than diffusion of responsibility, this case might have been simple confusion of responsibility. Both would be flaws in the system.

Tuneout Effect
Sometimes people have the tendency to pay less attention to warning signs the more frequently they are set off. The tune out effect occurs when cautions of risk become less potent the more they are issued. It is a corollary of normalization of deviance, which is where deviant behavior becomes more accepted over time. The effect is similar to the tendency of people to rarely treat car alarms as an actual car burglary. False positives can be especially detrimental to the efficacy of a warning system. In the case of PCA however, true positives (warnings of actual health concerns) were treated as false positives and ignored. Over time, the regulators and other bodies may have tuned out the warnings from PCA, including Kendrick's. In the mid 1980s, some companies gave the PCA plants failing grades and refused to do business. In the early 1990s, lawsuits were brought against PCA for toxins in their product, and the FDA was aware. Upon inspecting the plant, one owner said “It was just a time bomb waiting to go off, and everybody in the peanut industry in Georgia, Virginia and Texas--they all knew.” The tune out effect partially explains why although the regulators and PCA knew of the risks, they did nothing, even after Kendrick reported violations the first time, in 2006.

This same effect occurred in other ethics case studies, for instance the Challenger accident. Repeated concerns over the O-rings were raised, but each successful launch reassured executives. By the tenth launch, the concerns fell on deaf ears; the executives had tuned them out. Similarly, concerns had been raised about PCA in the past, but no one had died of salmonella thus far. By the time Kendrick came along, the regulators were tired of hearing of the risks. Kendrick's concerns had faded to background noise. In these cases, unfortunately, the more the concerns were raised, the more they were tuned out, and the less effect they had.

Conclusion
A number of cognitive biases probably played a role in the case of PCA and Ken Kendrick. What is the general takeaway from a case of the failure of big system?

With large systems, the effects can be faraway or distant. A small decision by one player can lead to large consequences for others, with many players involved in the propagation. Stewart Parnell decided to ignore a few Salmonella tests and people he did not know died as a result. Cognitive biases played a role in allowing this decision to propagate. With these large systems, subconscious cognitive biases may compound into bigger problems that are not realized until catastrophe strikes. When scaling from individual decision-making to large social structures, these biases deserve extra attention.