Sunday, 3 November 2013

What's for dinner?

As I suspected, the reason Eric hates the idea of GMO labelling is that he doesn't respect the eating preferences of others.

In his made up example, a new and obvious sticker is placed on supermarket apples notifying the presence of a chemical that Eric believes harmless. Sales of the stickered apples are likely to fall because not everyone will be as clever as Eric. Many doofuses (doofi?) will take fright at the multi-syllabic chemical and not buy those apples.  The chemical really is harmless, so there would end up being "too few" sold. We wise policy analysts should obviously protect those doofi from themselves by not requiring disclosure of the chemical.

The situation for GMOs is very different though. Instead of one harmless chemical on an apple, we have thousands of potential applications of lots of different GM techniques. The number of potential side effects from consuming GMOs is unknown. There aren’t nearly enough genes to do all of the things we know are done within the human body, so its pretty obvious that interactions between genes matter  But there are only so many of this vast multiplicity of risks that can be tested, and any test has some probability of type I and II errors, not to mention the whole broken science system and lots of crap published problem.

Put it this way: if there was any peer-reviewed scientific literature establishing that GMO crops in general have no unintended negative impacts then I think we'd have heard about it by now. The vague generalisations and opinions Eric is quoting fall well short of actual scientific proof, which is pretty ironic given the "bad science" slur that kicked off this discussion.

Let's face it: there is risk involved in eating GMOs and it's value is subjective. I am apparently more risk averse than Eric, and there must be others like each of us.

Feed that back into Eric's apple example and what do you get? Well, the doofi are me and other relatively risk averse people, and suddenly we're on the receiving end of the paternalism. Eric wants to stop us knowing where the GMOs are. Because he thinks knows that we over-estimate the risk.

This all seems very strange. Eric is not usually a meddlesome guy, yet here he is arguing against allowing people access to information that they want and that doesn't seem costly to provide. As far as I can tell, his sole motivation is to stop other people acting in accordance with their own preferences over what they eat.


  1. If every post on carbon taxes had to refute the dozens of "Oh, there's no link between CO2 and warming" folks, we wouldn't ever get anywhere. I'm taking the GMO consensus on it as given. If you want to start from the assumption that GMOs are really dangerous and risky, then your policy conclusion will differ from mine: I say that putting the label signals that the stuff is dangerous rather than just saying "hey, this is in here." If it is dangerous, then the label makes sense. If it isn't, it doesn't.

    Would you support warning labels on taps in towns with fluoridated water?

  2. I often read food labels. Maybe you don't, but I like to know how much fat and sugar and salt is there, and how many numbers are present. Its entirely my business how I process that information and what decisions I make afterwards.

    The contents of these food labels is already mandated. Should we rescind those rules? Hint: if you are not the main shopper, check with that person before answering.

  3. Step away from GMOs for the moment. Suppose there's some food processing technique that I hated. It scared me, and I thought it likely unethical. Ah, got one. Non-Kosher food preparation techniques. If it's not Kosher, it freaks me out (hypothetically).

    I propose mandatory food labelling letting everybody know whether a particular food is non-Kosher. I know that people can voluntarily put up Kosher labelling, but that's not good enough. I want each and every food item to specify whether it is Kosher.

    Ok, fine. So we add that to the label.

    Now somebody else wants each and every product labelled to say whether it's Halal. Now there's some but not complete overlap between Halal and Kosher standards. So we need to have two warning labels: Not Halal and Non-Kosher.

    Now somebody else wants mandatory labelling of whether things aren't organic. Somebody else wants mandatory labelling of things that aren't vegetarian. Next, the Paleos want mandatory labelling of what things aren't Paleo. And so on down the line.

    Each and every person who tries to keep Kosher (or Halal), or who values vegetarian food, or organic food, or goes Paleo, is entirely sincere in their beliefs about what's good to eat. I respect their beliefs. And I respect yours about wanting to avoid GMO products: entirely your call. But in the absence of some strong argument about riskiness, your mandatory labelling regime makes as much sense as any of mine, above.

    Now, riskiness. If you're taking Heinemann over FASNZ, and you're not expert in the field, what makes you side with him over them?

    I get disagreeing with consensus. I put more weight on some parts of the Austrian canon than do other economists. But I do so from within that field, knowing it reasonably well, and knowing which bits I want to take and which I don't. And I also get that sometimes, conveyed expert consensus can be wrong. Sometimes I'm happy to disagree with what seems the given science consensus - but I'm nervous about it unless I can make sense of the techniques used in that discipline. I've read a ton of the papers on light drinking and pregnancy and I'm happy to disagree with the standard "There's no safe level" line as consequence. But that's because population epidemiology papers are well within the standard stat techniques we use all the time - I can grok them.

    When I hear scientists talking about GMOs rather than activists, I hear the exact same weariness that I hear when scientists talk about the fluoride debate. While they can't rule out that some new crazy crop might come out from GMO techniques (or from other mutagenic but non-gene-splicing techniques like irradiation), there's no particular risk from GMO products. And whenever I hear the anti-GMO folks, I can almost hear the tinfoil hats crinkling.

    The bar for mandatory labelling can't just be "some people really care about it". If it were, then we'd need the mandatory non-vegetarian warnings. There's a good case for mandatory labelling of products that could have come in contact with nuts or other allergens that have a strong chance of killing a non-trivial number of people accidentally. The costs there seem low relative to the potential gains.

    And I do worry that big warning labels help stoke irrational fears around GMOs. It's those fears that fuel the protesters who go and rip up field trials, that scare the hell out of people in developing countries who could be really helped by modified crops like Golden Rice that can help prevent blindness.

  4. And read this.

  5. Thanks for elaborating Eric but I don't find your slippery slope argument at all convincing because I've never heard of popular support for these causes and I suspect you are just inventing them out of thin air. Neither is it surprising or news that scientists who earn their income from GM work are positive about GMOs. What does surprise me is that you care so much about what I and others choose to eat or not eat.

    1. I don't care what you choose to eat, John. I do care that big GMO warning signs over a bin of apples at the supermarket further inflames scientifically groundless fears about GMOs, reducing their adoption, and consequently worsening outcomes. In the absence of some provable "this item is riskier than another item, and by enough to justify mandatory warning labels", there's no difference between my proposed mandatory labels and mandatory GMO labels.

  6. This is getting very silly Eric. You quite obviously do care about what I and other people eat or you would not be referring to "reducing...adoption" of GMOs. The prospect that people might not want to eat GMOs is precisely the "worsening outcome" you are trying to avoid.

    GMO labels would help people make their own choices, and the ultimate reason you don't want labels is that you disapprove of those some of those choices.

    To make this paternalism palatable, it needs to be dressed up in a way that diverts attention from the core issue of who eats what. Hence the argument that adding something to a label signals risk, even though the existing mandated labeling scheme doesn't fit that story at all. And the idea that big warning signs would be compulsory because that helps to make a request for basic information sound extreme. And the fiction that we are not talking about GMO labeling at all, but rather about opening some flood gates behind which stand lots of other information requests that you are making up as you go along.