Santa Clara University

An Ounce of Prevention

Tackling Food Safety by Targeting Substandard Providers

The food industry, like any other, has its share of bad actors — companies that cut corners on health and safety practices and hope to get away with it. Given the potential harm that an unsafe lot of food could cause, businesses and government agencies have a strong interest in weeding out unsafe suppliers.

There’s been little academic research on the subject up to now, but S. Andrew Starbird, Professor of Operations and Management Information Systems at the Leavey School of Business, has been looking into it and has some ideas.

“Research has shown that shipping contaminated food is like placing a big bet,” he said. “If you want to scare the gamblers out of the market you have to increase their downside risk by increasing the chance they are going to lose or the amount they are going to lose.”

To help large-scale food buyers scare away those gamblers, Starbird has developed a model to help calculate what level of risk an unsafe food supplier is willing to take in a large contract, where routine testing of the product could result in the entire lot being rejected.

Professor Drew Starbird

His model is explained in a paper titled “Testing Errors, Supplier Segregation and Food Safety,” which was published last year in the journal Agricultural Economics.

Unsafe suppliers, he said, are those who haven’t adopted the most sophisticated food-safety methods, which means that more of their product is probably substandard or contaminated.

Relying on government inspections to identify those producers is problematic. In the meat industry — an area of significant health concern — the majority of U.S. packing plants don’t get inspected in any given year. And if people get ill from eating tainted food, it can be difficult to pinpoint which of the foods they ate caused the problem, and beyond that, whether the food contamination occurred at the production, distribution or retail level.

“It is better to scare unsafe producers out of the market before the contaminated food even has a chance of making someone sick,” Starbird said.

A strict standard of traceability, which would allow a lot of food to be traced back through the supply chain, with details of its handling shown at every step, would improve accountability and quality, Starbird said. In fact, many of the large-brand, high-profile food businesses insist on it and have the buying power to enforce it.

But, he said, “Producers don’t support traceability, in part because other parties can be made to pay for at least some of the cost of a mistake. It goes to a question of human nature. People don’t like to be held responsible for a failure — they like to be anonymous.”

Using Starbird’s model, a large-scale buyer of food could design a contract and testing system where the expected cost of shipping contaminated food is so high that unsafe suppliers abandon the market and safe suppliers take over. The model depends in large part on how much uncertainty is too much for an unsafe producer. Testing errors, sampling errors, errors in traceability all affect the producer’s uncertainty.

The U.S. Dept. of Agriculture partially funded Starbird’s research with a grant and has been talking with him since the publication of the paper about how his findings could be put into general use.

“The key concept,” he said, “is to make the unsafe food suppliers face uncertainty. That, in and of itself, is often enough to either deter them from selling or make them work harder to deliver a clean product.”

WEEDING OUT THE BAD GUYS: Andrew Starbird has developed models that can help large-scale purchasers of food craft proposals that discourage unsafe providers from bidding on a contract.

Printer-friendly format