Thursday, 21 May 2009

Welfare and animal welfare: NZ pork edition

The Kiwi econ blogosphere, namely Paul, Matt and Brad, has been abuzz with discussion of pigs. As background, some animal rights activists broke into a pig barn in the middle of the night, frightened the animals, took a video of the frightened animals, and aired it on the news. The video ignited discussion of the desirability of farrowing crates as compared to free-range options among those who've never been in a pig barn. [Update: Of course, I'm here referring to the popular outrage expressed in the press; the econ discussion subsequent to the popular outrage has been rather good, as noted below.]

Brad Taylor
points out that domestic pig farmers are making the best of some bad press by using the video in a bootleggers-and-baptists move: they argue that they're forced to use crates despite their better wishes because of the existence of foreign competition; if only tariffs protected them, they could change their methods. Paul Walker argues that the foreign competition angle is a red herring so long as barn raised pork is the cheaper production method; even with an import ban, producers would have a strong incentive to choose the cheaper option. Matt Nolan suggests that separate markets in free-range and barn raised pork solves most issues insofar as we only count animals' utility as it enters into human utility functions; if animals have intrinsic rights then a Pigovean tax (he calls it an externality tax, but Pigovean seems uniquely appropriate) on animal suffering solves things in the absence of Coasean bargaining.

A few years back, Tyler Cowen wrote an interesting piece on animal welfare. Relevant points:
  • If animals can be shifted among sectors of varying welfare, then a tax on meat-eating is of ambiguous consequence: there's no apriori reason to expect more to shift to the higher welfare sector than to the lower.
  • If regulation specifies a minimum standard of animal welfare such that we'd all view those animals as being happier having lived and been eaten than never having lived at all, vegetarianism is strictly harmful: fewer animals get to be born (and later eaten) when the demand curve is closer to the origin than it could be if vegetarians did their part in helping to defray these animals' costs of existence.
    "Some animal advocates argue for both mandatory care standards and for taxes or restrictions on meat consumption. But in fact, if we are at a care standard where animal lives are worth living, mandatory care standards should be combined with subsidies to animal breeding, not taxes on animal breeding."
  • If regulation specifies a minimum standard of animal welfare much higher than the animal enjoying epsilon utility, there's a tradeoff between number of animals and happiness of animals (imagine Parfit's repugnant conclusion, but with the conclusion as starting point and working backwards).

I'd follow up on Matt's argument by suggesting that the two-sector model (people who care about animal welfare buy free-range; others buy barn-raised) isn't sufficient to induce efficiency where folks who care about animal welfare care not only about the animals they themselves eat (or abstain from eating if vegetarian) but also have Pareto-relevant preferences concerning the animals eaten by others. In that case, they'd be willing to pay some money to subsidize barn-pork consumers to switch to free-range. Presumably a charitable organization could achieve that end. However, if there are some people who now don't eat meat because they can't afford the free-range variety and are unwilling to eat barn-raised for ethical reasons, and if vegetarians with Pareto-relevant preferences concerning animal welfare would prefer that nobody eat meat but that any meat eaten be free-range, then the charity doesn't unambiguously solve the problem for them: the negative effects on the extensive margin, as they see it, may outweigh the positive effects on the intensive margin. Unclear that there are enough vegetarians with Pareto-relevant preferences (ie willing to put their money where there mouths aren't) for the latter bit to be a major issue, in which case the charitable organization would be sufficient, but fun to ponder.


  1. "where folks who care about animal welfare care not only about the animals they themselves eat (or abstain from eating if vegetarian) but also have Pareto-relevant preferences concerning the animals eaten by others"

    Hi Eric,

    My aim was actually to point out that if pig welfare mattered independently of human preferences on their welfare, then the allocation of pig types would be inefficient.

    This is slightly different to stating that this is the result of people caring about what other people eat ...

  2. You'd suggested that the two sector model solves things absent animals' welfare counting directly but that we get inefficiency when animals' utility counts independently; I think I've shown we can get inefficiency even if animals' utility doesn't matter independently.

  3. Perhaps academic economists should be sow-crated up until they start contributing something useful to a discussion