Meat processing is one of the most concentrated sectors of the entire food system. Fromartz describes the efforts of one Joe Cloud in Harrisonburg, VA, to break into the business.
Cloud is riding a wave of consumer demand for meat from local farms, which has burgeoned along with the rash of deadly E. coli food poisoning incidents, hamburger recalls and undercover videos about grossly inhumane practices at a few large plants. Prominent chefs, who work with farmers and processors like T&E to get high-quality meat, have also championed the products.
For farmers, the sales are alluring; they make more money per animal when they sell direct, even if these channels represent less than 2 percent of all meat sales. It's also a way to escape the conventional system of meat production, since Virginia cattle typically are raised in-state for a year before being shipped to feedlots in Nebraska, Kansas and Texas to be fattened up and slaughtered -- and then shipped back as meat.
"Every step of the journey, someone has their hand in your pocket," said Jeff Lawson, who raises cattle and sheep at Green Hill Farm in Churchville, Va., a few miles outside Staunton. "If I could sell every animal I raised through Joe Cloud to get to your dinner table, I would. Any farmer would."
8 comments:
I suspect scale comes into this two ways, one economic and one epidemiologic.
Small scale operators process fewer units with which to cover their fixed costs that come as steps, such as USDA inspector costs. We have a public policy dilemma; either we taxpayers shift the burden of regulatory costs from us onto the regulated industry, which seems good economic sense but makes small scale processing less economic, or we pay for inspection as a public good.
From the epidemiologic perspective, due to differences in batch size contamination of product at a given from small scale processors is less likely to be detected than product from a large scale processor. Larger batch sizes means that more people are exposed, increasing the likelihood of problem detection. As a result of this detection bias, I suspect the increased confidence in small scale processors may be misplaced.
Thanks for the thoughtful comment on issues of scale.
There is one funny thing in your last paragraph that probably didn't come out the way you intended. You say that larger processors have larger batch sizes, so that more people are exposed. Paradoxically, you take this as evidence that larger processors are safer?
I think you may have hoped that something about detection in larger batch sizes would advance your argument. But, it does not suffice to say that a large contaminated lot would have a greater detection probability than a small contaminated lot. The large contaminated lot also endangers more people.
The real issue is whether a particular serving of meat is more likely to be contaminated if it comes from a small lot than a large lot. What is the evidence on that question?
Do I think larger processors are safer? Not having done a thorough search of that literature, I don't know. It seems to me that whatever their size any significant lapse puts them at serious risk of going out of business. Your question is an important one. I suspect the type of processing has far more impact than size of processor, especially for things like chicken.
I was thinking of this more as a detection probability problem. Given a low attack rate and all other things being equal, a problem is more likely to be detected from a large batch than a small batch. Over the long run, that would leads to the collective public consciousness that smaller producers are safer than the big ones. We see this already in the public perception that food prepared and eaten in the household is safer than food prepared and eaten in commercial establishments. My recollection (haven't looked at these papers for awhile) is that the reverse is true.
Processors, whether large or small, do not pay for USDA inspection, unless they work overtime or want their meat graded. So inspection costs are the same (none) for large or small operations unless they choose to work more than 40 hours per week.
As far as sampling and batch size, the establishment determines how much product the sample represents. The sample can represent 10 lb or 10,000 lb of product - it's up to the establishment. E. coli 0157:H7 is sporadic and present at such a low level that finding it is difficult even in a small lot of product.
From the USDA 2007 budget request:
USDA’s Food Safety and Inspection Service (FSIS) conducts mandatory inspection of meat, poultry, and processed egg products to insure their safety and proper labeling.
The FY2007 budget proposes a $987 million program level for FSIS, of which $124 million is funded by existing user fees, and $863 million by appropriation.
2011 USDA budget request summary:
User Fees for Meat & Poultry Inspection — The FY 2011 budget includes proposed user fees for meat and poultry inspection. Plants that have sample failures or require additional inspection activities due to a pattern of regulatory non-compliance would be charged a performance-based user fee. Also proposed is a flat fee for facility applications and annual renewal activities. This fee is intended to cover the increased costs above those basic inspection services provided to meat, poultry or processed egg product establishments. This proposal will require Congress to pass legislation authorizing the fees.
Surprised no one has brought up the longtime mantra "You can't test your way to safety" (Though it does help verify and keep folks honest... IMHO)
Post a Comment