He wrote to me:

There’s two things about the labeling debate that really bother me:

First, we have to concede that not everything can be labeled.  If so, the burdens would almost instantly put huge swaths of businesses out of business.  My dad, a dentist, does not have to label every instrument to describe where the metal came from, which machines made it, etc.  So the question is: where do we draw the line on what should be labeled?  My view is, if there is scientific evidence suggesting a plausible connection to harm, then requiring labeling makes sense.  But the view of the food activists is that they should just know everything, regardless of evidence, irrespective of cost.  So everyone should pay high costs because of their fears, which have no basis in evidence or fact.

On related matters, here is Mark Bittman on his ideal food labels, serving up a rather ambitious proposal.

On one specific point, he wants to levy a black mark against companies which treat their workers poorly.  On the contrary, that is a sign the product likely comes from a poor country and probably you are doing the world more good in buying it and, in the longer run, bidding up wages and working conditions in that country.  It helps other people more to buy from China than Portland, even though workers fare much better in the latter locale.  This difference in perspective is a simple illustration of how “ideal” food labeling can rather rapidly go wrong, especially when it is tangled up with the desire to make expressive statements about what one wishes to affiliate with or not.

A further question: at which margin do consumers stop paying attention?  When was the last time you read your new iTunes “I agree” contract before clicking?  Attention is scarce, so we need to pick and choose priorities.

What about the cost of producing such complicated labels and the enforcement of their veracity?  Food supply chains these days are often quite complicated.  Do you need to monitor how the fish sauce or oyster sauce in your composite food product was produced?  Bittman writes:

These are not simple calculations, but neither can one honestly say that they’re impossible to perform.

That is setting a rather low bar, and vaguely at that.  Most bad economic policies would meet that standard.  I would rephrase it: first figure out how many small and poor and foreign farmers this labeling proposal would put under — and then get back to us with a proposal.

Here is my earlier survey post on Bittman and evidence relating to GMOs.  And here Jonathan Adler offers an excellent analysis, including freedom of speech issues.