What do farmer's market chickens, motorcycles, and unpasteurized milk have in common?

A friend sent me a link to a new study in the journal in the Journal of Food Safety. The study shows that 90% of the chicken sold at a farmer's tested positive for Salmonella.  By contrast, only 52% of non-organic grocery store chickens and 28% of organic grocery store chickens tested positive for Salmonella.  In addition, the study found that for another illness-causing bacteria, Campylobacter, 28% of farmer's market chickens were positive but only 8% of non-organic grocery store chickens and 20% of organic grocery store chicken.  So regardless of whether you buy conventional or organic chicken at the grocery store, it is likely safer than that bought at the farmers market (at least the farmers analyzed in this study).   For one bacteria (Salmonella), organic is safer, for another (Campylobacter), conventional is safer. 

Why is this result interesting?  Because the findings are likely to be strongly at odds with most people's beliefs.  I suspect (but do not know for sure) that if asked, most people would say they think foods from farmers markets are safer than from grocery stores.  They would also likely assert organic is safer than conventional.  Yet this evidence (and other studies like it) is at odds with people's beliefs.  

I don't have a problem with people eating at farmer's markets.  Go for it!  But, ideally one should act knowledgeably, knowing full well the risks they're undertaking.  And I fear all the hype often causes people to mis-perceive the true benefits and risks of conventional, organic, and local foods.  

A similar problem exists with unpasteurized milk (or raw milk).  Although it is illegal in many states, many people want to buy unpasteurized milk.  Again, I say go for it (as long as they are two consenting adults; kids may be a different story at least if they're not your own).  But, let's not be glib about the safety risks.  Sure, it might be possible that pasteurization kills some healthy bacteria but it is certainly true, and scientific studies clearly show, that pasteurization kills illness-causing bacteria.  

So, why do we have government regulations that ban unpasteurized milk but promote farmer's markets?  Maybe the risks are larger or are more well known in one case (raw milk) than the other (farmers market meat).  One of the proper roles of government, I believe, is to provide objective-science based information.  What people do with that information is up to them.  But, it does bother me a bit when certain foods attain a moral status that causes people to under-estimate risks and over-estimate benefits.  Kahneman talked about this problem in his book Thinking Fast and Slow: something that seems good is therefore perceived unrisky and vice versa.  It also troubled me that many calls for food policies by food activists seem to be based on inaccurate perceptions of risks and benefits.  

What does this have to do with motorcycles?  Regulations in many states don't allow people to ride without helmets (helmet-less riding is banned) .  Clearly, riding a motorcycle without a helmet is risky.  How much riskier is it than eating farmers market chickens or drinking unpastuerized milk?  I don't know.  Strangely, in Oklahoma, we allow motor cycle riding without a helmet.  But, sales of raw milk in grocery stores is banned (my understanding is that it can be bought direct from the farm in OK).  So, people are presumed smart enough to weigh the risks of riding a motorcycle without a hat but not smart enough to buy raw milk from a grocery store?  Seems like a consistent paternalist would outlaw both.  Or a consistent libertarian would make both fully legal.  Either way, shouldn't we all want the best information to make choices?  

What is Natural Food Anyway?

At little over a month ago, I discussed some of the ongoing legal challenges that are swirling around "natural" claims on foods.  One of the big challenges is that the word "natural" is nebulous and is vaguely defined by regulators.   

I thought I'd try to shed a little light on the subject by making use of the survey project I just started and asking consumers what they think the word means.  In June, I added two questions to the survey.  The first question listed 10 statements and individuals had to place them in a box that said "I believe foods containing this ingredient are natural" or one that said "I DO NOT believe foods containing this ingredient are natural."  The order of items was randomized across respondents (sample size is 1,004, demographically weighted to match the US population, sampling error is about +/- 3%).  

The results indicate that most people think added cane sugar, salt, at beet sugar are "natural" but HFCS, sodium chloride, and biotechnology are not.  Interestingly, salt and Sodium Chloride are the same thing!  Yet, using the technical/scientific name reduces the % perceiving salt as natural from 65.6% to 32%!

Processed foods are seen as least natural.  "Processed food" is also a vague term.  Is cheese a processed food?   

The second question I asked was the following, "Which of the following best fits your definition of 'natural food'?"  I gave four options, and here is the % of respondents choosing each option.

The majority of respondents thought that the best definition (at least among the four I included) was, "fresh foods with no added ingredients and no processing."  

I suspect many of the foods sitting on a grocery store shelf that use the word "natural" do not meet this definition consumers found most descriptive.  

Country of Origin Labeling for Meat

About a decade ago, the US Congress passed mandatory country of origin labeling for a variety of food products including beef and pork.  At the time, we did some research on the costs of the law and the demand responses that would be required to offset those costs.  

In the intervening years, the law was implemented, the US was taken to court by Canada and Mexico, and the US labeling law was deemed to be an illegal trade barrier by the World Trade Organization (WTO).  However, rather than dropping the law altogether, lawmakers have doubled down and made them even more onerous and costly in an attempt to comply with the WTO ruling (you can read the current regs here).    ​

There is a key disconnect that is driving much of this debate.  When you ask people on surveys if they want to know the origin of their meat products, almost all say "yes."   But, when you look at the data on whether people read origin labels or whether demand for meat has been affected by the origin laws, a much different story emerges.   ​

Given that backdrop, I found the recent editorial by the president of the National Cattlemans Beef Association interesting (the Kansas Study to which he refers is here, and as you can see, I was a co-author on that publication)​.  Here is an excerpt:

It seems as if the first thing that is said whenever COOL is brought up is, “I am proud of the cattle my family raises,” and that is absolutely correct. I too am very proud of my family’s operation and all the work my wife and I, with our children and grandchildren, do to produce great beef. But a mandatory labeling program run by the federal government is not the way I want to showcase my product and add value. Labeling programs can work - just look at Certified Angus Beef or Safeway’s “Rancher’s Reserve.” These are marketing programs that are run by individuals with a specific interest and that is to promote and sell more beef to put on dinner tables across America. That is why these programs are successful. Additionally there is a tremendous amount of time and effort that goes into marketing these programs to the consumer.. But slapping on a label that says where this product was born, raised and slaughtered does not achieve the same result. In fact, a study by Kansas State University conducted in November of 2012 titled Mandatory County of Origin Labeling: Consumer Demand Impact made some key findings on this subject. The study found that mere country-of-origin information has not impacted consumer demand for beef or other covered products, and in fact, that many consumers are unaware labeling information exists. This is the issue with allowing the federal government to mandate a marketing program - it is not in their wheelhouse. Marketing at its very core relies on the distinction of one product from another. Neither USDA, nor any other government agency, can make that distinction based on origin labeling.

The Food Demand Survey (FooDS)

For a number of years, I've thought about creating a monthly survey to track consumer knowledge, concerns, and preferences for various food-related issues.  After no small amount of effort, and thanks to the funding from the Willard Sparks Endowment and DASNR and the assistance of Susan Murray, that vision has now become a reality.  

The inaugural issue is now up online, and we will to follow with regular monthly releases.

Of course, this initial issue can't report changes , but that information will come.

For those who might be interested, the purpose of the project is to provide timely information on:

• Indices of consumer sentiments on (or beliefs about) the safety, quality, and price of food consumed at home and away from home.
• Indices of consumers' anticipated demand for various meat products consumed at home and away from home.
• Awareness of food-related issues or events that could affect demand.
• Emerging policy or marketing issues.

It is envisioned that such data could be used by analysts to:

• Construct and analyze trends in beliefs, demand, and awareness
• Forecast changes in consumption
• Compliment (i.e., merge with) existing sources of secondary data (e.g., USDA disappearance or scanner data) in food demand analysis

Some of the motivations for starting the project include the following.

• Although scanner data is available to analyze immediate past behaviors, it is inherently backward-looking.  A consumer survey can be devised to be forward looking, potentially providing better forecasts.  Moreover, analyzing demand using scanner data is tricky due to issues of supply shifts, endogeneity, unobserved quality variation, promotions, etc that can be overcome with a well-designed survey.
• Current meat demand indices are aggregate, quarterly, assume a constant demand elasticity, and attribute all price/quantity changes to shifts in demand; a survey is more rapid and can better isolate demand-side issues.
• Existing surveys of consumers (i.e., panel diaries or home scanning data) only focuses on at-home food consumption; away from home food consumption now accounts for just under half of all food expenditures.
• Although some marketing companies routinely track eating intentions and awareness of food issues, the data is proprietary and is not publically released in any uniform fashion.  Moreover, their survey questions are not always designed using state-of-the-art techniques in consumer research.

Why I'm an Economist and Not a Psychologist

​This quote from Michael Moss's book Salt, Sugar, Fat accurately sums up one of the main reasons I see economic analysis as preferable to psychological explanations (and it is one of the main reasons I often prefer non-hypothetical economic experiments to hypothetical surveys).

Pg. 150: “There is not a lot to be gained from asking people why they like something because they don’t bloody know.”  - Fancis McGlone, former Unilever scientist

​