Blog

Thinking about risk

Consider this passage from a recent New York Times article

“Mr. Portier, who led the center when the revision process was initiated, said he believed parents should have been presented “with enough information to say caution isn’t ill advised, because we really don’t know, and there are enough indicators to say we should be cautious.””

The quoted former CDC official is espousing a form of the precautionary principle.  What do you think he's referring to? GMOs?  A new pesticide?  Food irradiation?

Nope. He's talking about cell phones.   The article describes some squabbles at the CDC on whether using cell phones cause cancer (the same World Health Organization group that says bacon and the weed-killer glyphosate may be carcinogenic  have also said that cell phones are a possible carcinogen), and how to communicate with the public on the issue.

So, here he have an issue for which there is apparently some scientific uncertainty, for which some government officials want the public to proceed only with caution, and the the public response?  A big shrug.  

Why is it that people think about the risks surrounding cell phones so differently than they do the risks surrounding GMOs, glyphosate, irradiation or many other food and agricultural technologies?  One could write a whole paper on that topic.  In fact I have (along with Jutta Roosen and Andrea Bieberstein).

There are a variety of reasons.  For one, people tend to conflate benefits and risks.  If something is beneficial then, people tend to think of it as less risky (even though we can imagine some very beneficial products that are also risky).  People directly see the benefits of using cell phones every day and thus they are perceived as less risky than, say, a pesticide that they have never heard of and scarcely can imagine  using.  Then, there is the old risk perception literature that originated with folks like Paul Slovic that is still relevant today.  The idea is that risk perceptions aren't driven by objective probabilities of possible bad outcomes but by how familiar or unusual a product seems and by how much control we believe we have over the risk.  Cell phones seems relatively safe because they're now quite familiar and because we decide whether to pick it up or turn it off.  Many food and agricultural technologies, by contrast, seem foreign and have secretly been slipped into our food supply (or so the story goes; ever notice now many of the top-selling food books use words like "hidden" or "secret" in the subtitles?).  

Whether there are good reasons for these psychological biases is less clear, particularly when they run at odds with the best scientific evidence we have on relative risks.  I for one, am perfectly at ease eating a tortilla made from Roundup-Ready corn while chatting on my cell phone.  The biggest risk is probably getting salsa on my iPhone.      

Country of Origin Labeling Conspiracy

I've seen the following meme going around on social media.    

I don't know whether the source is actually March Against Monsanto, but whoever put it out is obviously trying to stoke paranoia without any context or background.  And, like most good lies, the meme contains an element of truth.

First, it is true that both Houses of Congress repealed mandatory country of origin labeling (MCOOL).  This happened about a month ago.  What the meme doesn't revel is why.

MCOOL has been controversial since it's inception more than a decade ago, and the policy was fought by the largest beef and pork producer organizations.  Moreover, our trading partners were less than happy with the law.  Canada and Mexico filed suit with the World Trade Organization (WTO) claiming the law represented a non tariff trade barrier.  The USA lost the first round and several appeals (here's the timeline from the WTO).  What the meme doesn't reveal is that if Congress didn't repeal MCOOL then Canada and Mexico could slap on more than a billion dollars in retaliatory tariffs; my understanding is that these tariffs could have been on any products, not just meat.

What's not true about the meme?  

Well, despite the existence of MCOOL, our research shows most consumers didn't know where their beef or pork was coming from anyway; the vast majority of consumer's weren't checking the label.  More importantly, it simply isn't true that "now you will not know which country your meat comes from" simply because the government doesn't require the information.  If consumers really want the information and are willing to pay to have it, why wouldn't a retailer voluntarily advertise origin?  The restaurant chain Wendy's, for example, advertised for a while it only used North American beef.  Lots of small producers sell the animal products they raise at farmers' markets and at local restaurants.  There are thousands of products that voluntarily (without a government mandate) use  the "Made in the USA" logo to try to garner more sales.   There are lots of things I want (a new BMW; a private jet; perhaps origin information on meat), but just because I want something doesn't mean the government should mandate that it be provided. 

For more background, here's a recent report on issues surrounding MCOOL by the Congressional Research Service; here's another recent report summarizing the research on the subject prepared by my friends Glynn Tonsor, Ted Schroeder, and Joe Parcell for the USDA Chief Economist.

Year End Review 2015

Thanks for reading my meandering thoughts on this blog.  It's been a great year in 2015.  There were about 150 posts over the past year.  That's a little under three posts a week, and it's down about 30% from last year.  The drop can be explained by two things: 1) I focused quite a bit on writing my new book, Unnaturally Delicious, out on March 22 2016, and 2) Twitter has emerged as a better spot for pithy comments or for forwarding short information.

The good news is that the posts on this site have received around 98,000 page views over the last year, which is up about 88% from 2014.   

The top 15 most accessed posts this year have been:

1. Food Demand Survey (FooDS) - January 2015 (this issue contained the first results on consumer preferences for DNA labels; no other post was even close in terms of page views)

2. DNA labels

3.  Why are Beef Prices so High? (originally posted in 2014)

4. Ted Talks and GMOs (originally posed in 2013)

5.  The Organic Food Subsidy Myth (originally posted in 2012)

6.  Organic vs. Conventional Crop Yields (originally posted in 2014)

7.  How to Remove Gluten from Bread

8.  How Will Avian Influenza (Bird Flu) Affect Egg and Turkey Prices?

9. Might Consumers Interpret Labels as A Warning Label? from (2014)

10.  Who are the Vegetarians (from 2014)

11.  Taleb on GMOs

12.  Is the Growth in Agricultural Productivity Slowing?

13.  Do People Really Want to Express and Opinion on GMO Labeling?

14.  Red Meat and Cancer

15.  Why Are Beef and Pork Prices so High? (from 2014)

 

What is a GMO anyway?

Yesterday on Twitter, Nathanael Johnson asked a good question, and got lot of good answers.

I couldn't figure out how to embed the whole Twitter conversation but there were scores of interesting responses.  The discussion is related to another important one: A GMO isn't a single thing, it's many, many possible things.  But, Nathanael's question is deeper, and philosophical.  Is this thing we've called "GMO" something that's only in our heads or is it something that exists independent of our minds.  Another way to look at it: if a Martian were to travel to earth and look at what's our dinner plate and is growing on our farm fields, could they - without knowledge of our history or social baggage - identify a class of things called "GMO" that would match up with the class of things we call a "GMO"?

Lettuce a bigger environmental threat than beef?

On Monday a press release began making the rounds claiming that "Vegetarian and 'healthy' diets are more harmful to the environment" and the story has generated provocative headlines like, "Lettuce three times worse than bacon for the environment."  The results are based on this paper published in Environment Systems and Decisions by Michelle Tom , Paul Fischbeck, and Chris Hendrickson.

Here's a portion of the abstract:

“This article measures the changes in energy use, blue water footprint, and greenhouse gas (GHG) emissions associated with shifting from current US food consumption patterns to three dietary scenarios, which are based, in part, on the 2010 USDA Dietary Guidelines . . . . This study finds that shifting from the current US diet to dietary Scenario 1 [same mix of foods but fewer calories] decreases energy use, blue water footprint, and GHG emissions by around 9%, while shifting to dietary Scenario 2 [holding calories consumed constant but shifting away from meat to fruit and veg] increases energy use by 43%, blue water footprint by 16%, and GHG emissions by 11 %. Shifting to dietary Scenario 3, which accounts for both reduced Caloric intake and a shift to the USDA recommended food mix, increases energy use by 38 %, blue water footprint by 10 %, and GHG emissions by 6 %. These perhaps counterintuitive results are primarily due to USDA recommendations for greater Caloric intake of fruits, vegetables, dairy, and fish/seafood, which have relatively high resource use and emissions per Calorie.”

So, what's going on here?  Didn't the authors of the latest (proposed) Nutritional Guidelines suggest less meat eating due to "sustainability" concerns?  Haven't we repeatedly read things like this quote from a story in Time in  2008? 

“It’s true that giving up that average 176 lb. of meat a year is one of the greenest lifestyle changes you can make as an individual. ”

A few comments about this latest study and how it relates to the "received wisdom."

First, meat eating often looks bad in aggregate because the industry is so big.  In the US, we eat a lot of meat.  As a group, animal products probably represent the largest share of food expenditures of any food category.  As a result, we need to put the outcomes on some kind of units where foods can be compared on an even playing field.  There are a lot of options: lbs of food produced, dollars spent, acres used, or as the current study uses, calories.  As I've discussed before:

“. . . meat is relatively (relative to many fruits and vegetables) inexpensive on a per calorie or per gram of protein basis, although meat looks more expensive when placed on a per pound basis. If you want really inexpensive calories eat vegetable oil or crackers or sugar; if you want real expensive calories, eat zucchini or lettuce or tomatoes.”

For what it's worth, this discussion is related to the debate on whether eating healthy is more or less expensive.  As it turns out: the answer depends how you measure it ($/lb or $/calorie) .

Second, it is important to think on the margin, or think in terms of changes.  Meat is bad.  Compared to what?  As I discussed in the Food Police,  many fruits and vegetables are big users of water and pesticides.  So, if we eat less meat, what will we eat more of instead, and what are the impacts of the items we switch to?  This recent study produces results that look a bit different because it looks not just at aggregates but at marginal changes.  

Third, as this study makes clear, "the environment" is not a single thing.  It is multidimensional.  Here's a graph from the paper.

Some foods are better water users, others worse in terms of GHS emission, others produce more calories/acre and spare more land, and so on.  Very rarely are there "have your cake and eat it too" moments, and there are typically tough tradeoffs between health, environment, and taste (and even tradeoffs within each of those categories).  One problem with studies like this is that they don't count the consumer welfare cost from eating a different mix of foods or different number of calories from what they normally consume, so we don't have any sense for the tradeoff between taste and cost on the one hand and health and the environment on the other.  

The last comment is that it's tough for me to evaluate the "quality" of this paper, and one should probably be a bit careful of suffering from confirmation bias.  A lot of the assumptions driving the result actually come from other papers and analyses.  Moreover, the work was published in a relatively new journal (started in 2012) and I know nothing about it. [update: a follower on Twitter informed me that the journal has been around since the 1980's but recently changed its name]  However, it does appear that the authors' results are similar to those in another paper that the Nutritional Guidelines Committee actually discussed in their report.  The committee writes: 

 

“a report from Heller and Keoleian suggests that an isocaloric shift from the average U.S. diet (at current U.S. per capita intake of 2,534 kcals/day from Loss-Adjusted Food Availability (LAFA) data) to a pattern that adheres to the 2010 Dietary Guidelines for Americans would result in a 12 percent increase in diet-related GHG emissions. This result was modified, however, by their finding that if Americans consumed the recommended pattern within the recommended calorie intake level of 2,000 kcal/day, there would be a 1 percent decrease in GHG emissions.”

But, of course, people don't just follow the guidelines to a tee.  That's one reason why these sorts of guidelines and recommendations should consider consumers' behavioral responses to the policies in question.