# kbog did an oopsie! (new meat eater problem numbers)

post by kbog · 2019-02-15T15:17:35.607Z · EA · GW · 8 comments

Edit: this is out of date, see https://forum.effectivealtruism.org/posts/YuFD4v7DFBcM57eSA/consequences-of-animal-product-consumption-combined-model [EA · GW]

The "meat eater problem" is the problem that, the more we enrich people, the more meat they will eat - which increases animal suffering.

Well I took my old analysis here [EA · GW], that was estimating the magnitude of the meat eater problem. Then I decided to combine it with the animal suffering numbers from my more recent estimate [EA · GW] of the net farm animal suffering caused by farming.

In the process I found that my meat eater calculation included a critical misreading of York & Gossard. Fish consumption does increase as African countries get wealthier, by an estimated 0.79kg annually per \$1000 GDP.

I'll suppose that 0.6kg out of this marginal increase is from farming. I'll suppose that 0.4kg of this is like catfish (because the similarly small tilapia is commonly farmed in Africa and other poor regions) and 0.2kg is like salmon.

Plugging these numbers into the animal welfare spreadsheet as a "diet" (too simple an action to merit uploading - it's just the kg quantities from the meat eater sheet, plus 0.4k for catfish and 0.2kg for salmon) I get -720 welfare day-points for the year's marginal meat consumption. This means that a \$1,000 increase in GDP per capita for a person in Africa has an animal cost that is equivalent to -2 points on a -100 to +100 scale, which is 4% of the average welfare difference between living in India and living in Canada, using my sentience weights. Things may be worse in India/Bangladesh: presumably they fall somewhere between the average relationship for Asia (2.31kg) and the relationship for the West, Africa and the Middle East (0.79kg). The authors suggest that cultural factors drive the difference.

I'm ignoring the climate change costs of meat consumption because that can and should be left as part of the more general question of the link between economic development and long run GHG emissions. In any case, GHG emissions only make a relatively small difference on any short or medium run outlook, as I argued when I posted the farming analysis.

Given these assumptions about animal farming (which point heavily towards animal charity in the first place, anyway - see Shulman's comment in the original meat eater problem thread) I think these numbers suggest general agnosticism about the direct impacts of economic growth aside from the question of wild animal suffering. I would make an exception for pulling people out of serious poverty because the wealth-welfare relationship is solidly logarithmic. Per Footnote 7, York & Gossard seem to think that neither a linear model (which they present, because it's simpler) nor a logarithmic model (which they claim to have done) is decisively a better fit for the relationship between wealth and animal consumption.

comment by Aaron Gertler (aarongertler) · 2019-02-15T18:20:54.114Z · EA(p) · GW(p)

Upvote for noticing an error in your model and announcing the update -- that's a good habit, and I like the idea of posts on the Forum which encourage said habit.

I'm sure you had a strategy in mind for the current title, but I'll still suggest choosing something more descriptive; this seems like it might be confusing to link to in a few months (especially for people who don't recognize the name "kbog").

comment by samaramendez · 2019-02-20T04:43:01.316Z · EA(p) · GW(p)

Thanks for posting this update. I'm fairly new to the forum, so I missed the original posting of the two links you've provided.

These posts seem closely related to meat consumption Kuznets curves. I am not an expert in this literature and I intend to do more reading on the topic, but a quick search found this 2013 paper, which might be of interest to you. I've only briefly looked over the paper, but their results seem to update and support the findings of York and Gossard (2004). It seems that the expected turning point of the meat consumption Kuznets curve occurs at a per capita income of US\$45,263 in the full sample of 150 countries (in their sample, only 3 countries reach this threshold).

Not that Kuznets curves should be interpreted as causal relationships, but I was naively hopeful that it might occur here.

Anyway, I'm interested to see your animal suffering extension of this idea. I think there is a small literature on animal welfare Kuznets as well, which may be interesting. Thanks for sharing your work and your updates!

Replies from: kbog
comment by kbog · 2019-02-20T06:57:44.362Z · EA(p) · GW(p)

Interesting. Y&G said that they checked for a curvillinear relationship and the results "do not suggest substantively different conclusions," which I understand to mean that there isn't good evidence for a Kuznets curve.

I did not know that India's average consumption was so low, perhaps their marginal increase in consumption is not much either.

Looking at Table 3. Am I reading this right: the relationship for low income countries is +0.0188kg (annually) per \$1 annual income? That's 18.8kg from \$1000 which is about an order of magnitude greater than the Y&G results.

Replies from: samaramendez
comment by samaramendez · 2019-02-20T15:11:37.533Z · EA(p) · GW(p)

I just had a very quick look-through of Y&G, but it looks like they tested for curvilinear (i.e., a log transformation of GDP) only. I could be missing a footnote, but I don't believe they included a second-order GDP term to test a polynomial relationship.

However, the findings of the 2013 paper largely support that, from my quick reading. The estimation of the second-order coefficient is significant but basically zero for most of the different data slices. Further, when they back out the inflection points, the income levels for the turning point of meat consumption are much higher than the turning points for other Kuznets curves (\$45K relative to \$3K-\$12K in the general environmental KC literature).

But actually now that I'm digging in to the results, I think the tables report different numbers from the text. Neither Y&G nor RC&M are forthcoming about their units, which is frustrating, but at least Y&G discuss their results clearly. I'm a bit frustrated about the write up of this paper. I believe that you're reading the results correctly, unless the authors are actually using per capita GDP in thousands like Y&G and failing to report that (although that result wouldn't make any more sense). It does seem a lot higher than Y&G.

I'm losing faith in this paper now (at least in the result discussion), but I would like to check out the literature further and see if there are any other newer papers that can provide insight into the differences.

comment by AviNorowitz (AviN) · 2019-04-14T21:23:00.934Z · EA(p) · GW(p)

Thanks for following up on this and posting a correction. I'd suggest updating your original post to include your updated fish consumption estimates. Or if that's too much work, a correction on that post with a link to this new one would be helpful. As it stands now, it still says: "I also excluded fish because there is no significant correlation between income and fish consumption in African countries." I think people are likely to find your original post when researching the meat eater problem, especially since the corresponding EA Concepts page cites it.

Replies from: kbog
comment by kbog · 2019-04-14T21:27:31.446Z · EA(p) · GW(p)

kk will add a link. But major edits would be a pain because the old post is in HTML.

comment by rafa_fanboy · 2019-02-15T21:46:38.528Z · EA(p) · GW(p)

do 3rd world countries eat factory farmed meat?

Replies from: kbog
comment by kbog · 2019-02-16T05:31:59.031Z · EA(p) · GW(p)

I'm not sure exactly, my perception is that (1) often they don't currently but the new growth is more likely to be factory farming, (2) traditional farming isn't clearly better. Farming in the West is probably covered by more welfare regulations than farming in poor countries.