Economist at Rethink Priorities, Farmed Animal Welfare Team
samaramendez
Karma: 111
[Linkpost] Inconsistent evidence for price substitution between butter and margarine: A shallow review
New research plans from The Humane League Labs
Paid Fellowship at The Humane League Labs
I would be happy to speak with your further, especially to further the cause of getting more economists/social scientists hired! I’ll send you a PM.
Thanks for sharing this article and opening this discussion.
I’m also interested in hearing the general survey replies to understand what specifics the movement as a whole is looking for.
Anecdotally, I’m the economist at The Humane League Labs, and I tend to get a broad range of questions about cost-benefit analysis, consumer preferences (and how to change them), market structures, impact analysis, etc. What I actually work on tends to be causal inference, which is a place where economists’ skills could be very helpful for the movement.
I have my PhD, and I know that my education level plus my research interests plus my alignment with the cause were the main factors in my hiring. I believe that high-level undergrads or masters students would be well equipped to answer general cost-benefit analysis and causal inference questions, though more specialized questions like firm competition models and counterfactual analysis using structural models would likely not be covered in their training. I think that there is a lot of work for people with broad backgrounds and interests to do.
I’ve spoken to a few EAA researcher departments who are looking to hire economists, and behavioral and agricultural backgrounds come up most often.
It’s great to see the magnitude of the change, and it will be interesting to see how it moves as the main cage-free commitment deadlines get closer.
Thanks for your comment! It’s not much of a comment-generating piece, although I do hope that people might use the data for other projects. I am always excited to see data publicized here as well.
[Link] US Egg Production Data Set
That’s right, I remember that result in Veganomics as well. Looks like I need to revisit the literature.
The wealth relationship is very interesting here, not what I would have expected. I would be interested to see if the same relationships hold over different countries with different income distributions.
Thanks for sharing!
Sure, I didn’t discuss the connection between your demand decrease and the remaining people’s elasticities. Let’s use Brian’s example of the supermarket supply system to illustrate that connection.
When you stop buying chicken, demand in the store has decreased. Perhaps it decreases to the point where the store decides to put the chicken on sale. Other shoppers who still buy meat will see the sale price and change their purchases according to their price elasicities, which are defined as the percent change in quantity demanded given a 1% change in price. Using the specific example of the paper’s elasticity, a 1% decrease in price would cause a shopper to buy 0.68% more chicken. In the next ordering cycle, the inventory manager orders less chicken because you decided not to buy chicken and because the other shoppers purchased some of the excess stock at the sale price. We could call this a new equilibrium, and we see that equilibrium quantity demand has shifted. I won’t make an assumption on whether the equilibrium price is lower in this little supermarket scenario (only because I’m not sure how supermarkets actually set their prices).
These causal pathways are always a little tricky because you can always talk your way out of any scenario, but the underlying mechanism behind a demand shift in a larger, more abstract market could work like this.
This hypothetical supermarket brings up your second point. You’re right for calling me out on my statement about not making a difference—your non-purchase of meat in a grocery store does have the probability of setting off a chain reaction in the supply chain that reduces the number of chickens killed in expectation. I should have been more clear in my previous comment, because I was certainly thinking of a larger-scale aggregate market like, for example, the entire US market for chicken. In this case, you are one of ~300m people in the market for chicken, and your one-time non-purchase would have a small effect that would be hard to estimate (this is a more precise statement of what I meant).
I do believe that you save may chickens, in expectation, with a non-purchase. And further, I do believe that your continued non-purchase of chicken and your entry into the alternatives market will, in expectation, save more chickens. I also believe that your non-purchase of chicken will likely lead you to change other purchasing decisions according to your ethics, and I believe that your ethical decisions will, in expectation, influence the people around you. (This is also something I’d like to study formally in the future.) So I believe, in expectation, you can do a lot of good!
Thanks for the clarification, I understand your question now. You’re asking about estimating the size of a demand shift that results from one economic agent leaving the market, as opposed to an elasticity. I believe we’re asking the same question; with elasticities, I want to investigate the underlying mechanism that takes us from your leaving the market to a shift in the demand curve. Whereas you would like to know the end result only.
The answer to this question (the effect on aggregate demand of one person leaving the market) may be difficult to estimate correctly. The best approximation I can think of in the existing literature is the effect of food scares on demand. For example, this paper on news coverage of salmonella outbreaks and this paper on media coverage of the BSE (mad cow) outbreak on demand for British beef both show the impact on demand of media coverage of food scares. Of course, they have to use a media coverage as a proxy for people leaving the market, since there’s no way to collect data in that way.
Circling back to your friend’s question, I do think that it’s unlikely that you or another individual leaving the market would affect price enough to trigger increased purchases in the people remaining. I suspect that you leaving the market would also have a very small effect on aggregate quantity demand, in the same way that your marginal non-purchase of chicken will be unlikely to change the way that the supermarket orders their products. Similarly, just you entering the market for meat alternatives won’t cause a large increase in demand. I do think your discussion with your friends will plant some seeds, though.
Many great points already about how to respond to your friend. I’d like to expand a bit on a few.
Shaybenmosche mentions the connection between two markets that illustrates interesting subtleties. Yes, you leaving the market reduces demand for meat and likely lowers the price which may in turn increase demand for the remaining people in the market. But you entering the market for meat alternatives increases demand and price there, which will drive producers to innovate and/or enter the alternatives market.
Cole points out that a shift of the demand curve leads to a reduction of price and quantity, which follows from the fact that you as a new vegetarian have left the market for meat. An interesting extension of this idea is that demand for meat may be more inelastic after your exit (given that remaining consumers are more inelastic in their demand than you). This is not necessarily great news for the animal advocate who hopes that increases in price (due to policy or otherwise unrelated to the price changes in your scenario) will reduce meat consumption and therefore animal suffering. But it does somewhat undermine your friend’s argument—the reduction in price that occurs because you left the market will probably affect the inelastic meat eaters less than your friend imagines.
Saulius mentions the Andreyeva et al (2010) paper, which is a good meta-analysis of the food elasticity literature. The answer to your question about cumulative elasticity is mostly here (they find the mean of elasticities that are estimated using a variety of methods), although I think you’re looking for aggregate elasticity, which is roughly the shape of the aggregate demand curve with respect to price. The main difference between the two concepts is that aggregate generally refers to summing over people or products, while cumulative seems to sum over time (though I’m less familiar with this literature). There are many subtleties about the different methods used to estimate elasticities that I’d love to discuss, but won’t here because this comment is already long.
Following up on Saulius’ reply, we generally prefer less-aggregated elasticities when it’s possible to estimate them finely, because the aggregate elasticity is sweeping a lot of important dynamics under the rug. I suspect that’s probably what you’re interested in as well—given the different behavior patterns of demographic groups of people, what is the overall effect of a price change in chicken on quantity demand for chicken? If I’ve misinterpreted your question, please let me know! The good thing about estimating elasticities for demographic groups is that the researcher can combine them after the estimation procedure to understand the overall effect of some change.
As the Andreyeva et al paper shows, a lot of research has estimated price elasticities for food, and other work has estimated income elasticities of demand for meat. Less well-studied (because the models get complicated quickly) are cross-price elasticities for different types of meat and their alternatives. This is on my short-list of projects to work on next.
Thanks for opening up the discussion here!
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.
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!
Thanks for weighing in!
I read your comment to mean that you skimmed the butter and margarine studies that we considered in our brief analysis, but please correct me if I misunderstood.
Yes, I agree that there is very little consideration for price endogeneity in the butter and margarine studies that we found. I could have made that more clear in our report. I suppose that oversight is due to the studies generally skewing older and using older methods. One of the meta-analyses we reviewed (Auer and Papies, 2020) included price endogeneity in their analysis, and they find a positive association between overlooking endogeneity and elasticity size. I do think there are other methodological issues that might be biasing these results—so accounting for price endogeneity may not remove all variation due to methodological issues—but I agree that including strong supply-side instrumental variables would be a nice standard practice.