This post is written for a more general audience & therefore some things in it may be obvious to EAs. And the tone is a little bit more strident than is the norm here. But I think the two arguments I cover are important to any EA interested in animal suffering.
Faintly are heard the ploughmen at their ploughs,
But not an eye can find its way to see.[1]
– John Clare
Of all the positions I hold strongly, the one that says it’s morally right & good to practice vegetarianism & veganism is, in a way, I think – & I hope this doesn’t reek too pungently of hubris – the easiest one to argue for. That’s because most people already have these intuitions, though they apply them inconsistently. They feel that we have some duties & obligations towards our pets & other companion animals, for example. They would also most of them admit that what goes on in factory farms today is less than perfect. And then, having not quite forgotten this state of affairs, when they happen to be used by someone, they complain of having being treated like an animal, which of course suggests that animals are treated in a way that we find utterly unacceptable (if they are in fact the sorts of creatures that can be mistreated).[2]
That said, as with everything else, so with vegetarianism & veganism: there are arguments for & against it & some of those are good & some are bad. I am going to describe two arguments against moral vegetarianism & explain why I think they are inadequate. What they have in common is that they argue not that reducing animal suffering is not a worthwhile end but that vegetarianism (& veganism, but I’ll stop saying that) is an ineffectual means of achieving that end. They do this on empirical consequentialist[3] grounds, giving essentially economical reasons for why it is so.
The Argument from Price Elasticity
The first argument goes as follows. The law of supply & demand says that when supply exceeds demand, prices go down. Take cattle ranchers, for example. They already maximise production in the area they have – they are not going to be producing less cattle if demand drops. Instead they will sell it at a lower price. But if they make any profit at all from their cattle, they will raise as much cattle as they can. Hence people going vegetarian won’t reduce meat consumption, because meat eaters will make up for the lost quantity demanded by buying more meat at cheaper prices.
In economic terms, this argument makes the claim that the price of meat is perfectly elastic, that is, the price will adjust as demand changes. But from what I can tell from the literature, the price of meat is relatively inelastic, though it depends a little on the region & the type of meat.[4][5] That makes sense in the developed world at least (but likely elsewhere, too) as meat, for meat eaters, is something of a necessity; and moreover, if the price of meat fell, most people wouldn’t eat more meat for the simple reason that they already eat about as much meat as they want to.
It’s true that the cattle rancher maximises production in the area they have & that they’re unlikely to start producing less cattle when demand drops. But at some point the price of purchase isn’t going to cover the costs of production anymore, especially taking opportunity costs into account.[6] So the whole ranch will have to fold & the rancher will need to find something else to do with the land, e.g. lease it to some developer.[7] And there will always be some cattle ranchers whose profit margins are very low; these will be the first to go when demand goes down.[8]
There are also strong regional differences that we wouldn’t expect to see if this argument were true. For instance, in India, which has a strong vegetarian culture, meat consumption per capita is one quarter of what it is in neighbouring Pakistan, even though GDP per capita is much higher in India. So the difference from many Indians eating less meat isn’t made up by remaining Indians eating more of cheap meat. Though I’m sure there are many confounds here.
The Argument from Supply Chain Buffers
The second argument begins with the observation that, when one buys a piece of meat, whatever animal it came from has already suffered & died for it – whatever good or bad comes from the purchase must be good or bad for future animals. In other words, if there’s any effect at all, it must be an effect on future meat production. But we know that any one individual purchase is exceedingly unlikely to have that kind of an effect.[9]
Now you may argue that, sure, the chance that a single purchase will affect future production is very low, but when it happens (as it will) the effect will be very large indeed. But I can refine my argument. I can say, as Mark Budolfson does[10], that we know something about meat-producing supply chains. We know that meat producers always produce a bit more than is normally sold (which goes to waste or is repurposed when it expires on the supermarket shelves) & moreover generally that there are small but reliable buffers at every step of the supply chain. This introduces more than enough noise in the signal to make any one purchase invisible to the producer’s eyes.[11] The signal is “absorbed” by the buffers. Put differently, in deciding not to buy a piece of meat, one has no greater chance of making a difference than does the person who casts a vote in a U.S. presidential election from a state like Vermont or Wyoming.[12]
So goes the argument from supply chain buffers. I have tried my best but am forced to admit that not only am I not able to refute this argument, but I’m not even sure I understand it. To see why, imagine for example a scenario where the populace of some city or country decides, for cultural reasons, one by one to stop buying meat. If the inefficacy argument were true, the first person’s doing so would not affect meat production, because the signal would be lost in the noise & absorbed by supply chain buffers. But the same is true for the second person, & the third, & the fourth & so on. Gradually consumption would drop towards zero but meat producers, having not at any point in time discerned a significant drop in demand, would blissfully go on producing the same amount that they always produced.[13]
But this is absurd. The argument proves too much. I don’t think the inclusion of supply chain buffers adds anything to it. It just shifts it sideways, so to say. Because it’s clear, I think, to those who manage each stage of the supply chain roughly how much of the buffer is used or sold at that point & how much of it goes to waste, in the same way that the store manager knows how much meat they’ve sold each day & how much they’ve had to throw out.
Say a supermarket expect to sell around 100 sirloin steaks in a day, perhaps 90 on a bad day & 110 on a good day. Even if they stock 110 steaks, which of course is more than they sell on average (giving them a reliable buffer), there will nevertheless be a threshold below which they decide that too many steaks go unsold & they need to reduce their stock. This doesn’t change if their desired buffer is zero percent, ten percent or 100% because the quantity stocked is still a product of the expected quantity sold. I’d assume that each preceding step of the supply chain operates similarly.
Final Thoughts
McMullen & Halteman[14] mention some further reasons to believe that moral vegetarianism is effective:
When you buy a chicken, you are in a way buying more than a chicken. That’s because it took a captive hen to breed that chicken. So by reducing demand for chicken you are also reducing the need for egg-laying hens. The same goes for most other animals sold for meat.
By buying vegetarian, you encourage grocery stores & restaurants to offer more vegetarian options, which in turn may encourage other people to eat vegetarian food more often.
Meat is relatively cheap because the meat industry is large & benefits from economies of scale. The more people buy vegetarian, the smaller the meat industry gets & the larger the plant-based food industry gets, meaning that, all other things being equal, the price of meat will increase & the price of plant-based foods will fall.
We are fewer than 8 billion people in this world but every year we slaughter 80 billion non-human animals for meat. If that’s wrong, it’s a wrong of truly staggering proportions. It’s also one that is growing & will continue to grow as humans across the world get more prosperous. And it’s not something that technology will solve (though technology can & does help); what’s needed is nothing less than a great moral revolution on the order of abolitionism. That, if for no other reason, is why these questions matter.
Gallet, C. A. (2010). Meat Meets Meta: A Quantitative Review of the Price Elasticity of Meat. American Journal of Agricultural Economics, 92(1), 258–272.
Selvanathan, E. A., Jayasinghe, M., Hossain, M. M., & Selvanathan, S. (2020). Modelling the Demand for Meat in Bangladesh. In Science and Technology Innovation for a Sustainable Economy (pp. 135–151). Springer International Publishing.
McMullen, S., & Halteman, M. C. (2018). Against Inefficacy Objections: the Real Economic Impact of Individual Consumer Choices on Animal Agriculture. Food Ethics, 2(2–3), 93–110.
Budolfson, M. B. (2018). The inefficacy objection to consequentialism and the problem with the expected consequences response. Philosophical Studies, 176(7), 1711–1724.
It’s almost as if the argument is that the demand signal is a kind of emergent phenomenon – it’s there in the aggregate but invisible at the individual level. That makes me wonder whether there are general preconditions for emergence &, if so, what those preconditions are.
McMullen, S., & Halteman, M. C. (2018). Against Inefficacy Objections: the Real Economic Impact of Individual Consumer Choices on Animal Agriculture. Food Ethics, 2(2–3), 93–110.
Two Inadequate Arguments against Moral Vegetarianism
Link post
This post is written for a more general audience & therefore some things in it may be obvious to EAs. And the tone is a little bit more strident than is the norm here. But I think the two arguments I cover are important to any EA interested in animal suffering.
Of all the positions I hold strongly, the one that says it’s morally right & good to practice vegetarianism & veganism is, in a way, I think – & I hope this doesn’t reek too pungently of hubris – the easiest one to argue for. That’s because most people already have these intuitions, though they apply them inconsistently. They feel that we have some duties & obligations towards our pets & other companion animals, for example. They would also most of them admit that what goes on in factory farms today is less than perfect. And then, having not quite forgotten this state of affairs, when they happen to be used by someone, they complain of having being treated like an animal, which of course suggests that animals are treated in a way that we find utterly unacceptable (if they are in fact the sorts of creatures that can be mistreated).[2]
That said, as with everything else, so with vegetarianism & veganism: there are arguments for & against it & some of those are good & some are bad. I am going to describe two arguments against moral vegetarianism & explain why I think they are inadequate. What they have in common is that they argue not that reducing animal suffering is not a worthwhile end but that vegetarianism (& veganism, but I’ll stop saying that) is an ineffectual means of achieving that end. They do this on empirical consequentialist[3] grounds, giving essentially economical reasons for why it is so.
The Argument from Price Elasticity
The first argument goes as follows. The law of supply & demand says that when supply exceeds demand, prices go down. Take cattle ranchers, for example. They already maximise production in the area they have – they are not going to be producing less cattle if demand drops. Instead they will sell it at a lower price. But if they make any profit at all from their cattle, they will raise as much cattle as they can. Hence people going vegetarian won’t reduce meat consumption, because meat eaters will make up for the lost quantity demanded by buying more meat at cheaper prices.
In economic terms, this argument makes the claim that the price of meat is perfectly elastic, that is, the price will adjust as demand changes. But from what I can tell from the literature, the price of meat is relatively inelastic, though it depends a little on the region & the type of meat.[4][5] That makes sense in the developed world at least (but likely elsewhere, too) as meat, for meat eaters, is something of a necessity; and moreover, if the price of meat fell, most people wouldn’t eat more meat for the simple reason that they already eat about as much meat as they want to.
It’s true that the cattle rancher maximises production in the area they have & that they’re unlikely to start producing less cattle when demand drops. But at some point the price of purchase isn’t going to cover the costs of production anymore, especially taking opportunity costs into account.[6] So the whole ranch will have to fold & the rancher will need to find something else to do with the land, e.g. lease it to some developer.[7] And there will always be some cattle ranchers whose profit margins are very low; these will be the first to go when demand goes down.[8]
There are also strong regional differences that we wouldn’t expect to see if this argument were true. For instance, in India, which has a strong vegetarian culture, meat consumption per capita is one quarter of what it is in neighbouring Pakistan, even though GDP per capita is much higher in India. So the difference from many Indians eating less meat isn’t made up by remaining Indians eating more of cheap meat. Though I’m sure there are many confounds here.
The Argument from Supply Chain Buffers
The second argument begins with the observation that, when one buys a piece of meat, whatever animal it came from has already suffered & died for it – whatever good or bad comes from the purchase must be good or bad for future animals. In other words, if there’s any effect at all, it must be an effect on future meat production. But we know that any one individual purchase is exceedingly unlikely to have that kind of an effect.[9]
Now you may argue that, sure, the chance that a single purchase will affect future production is very low, but when it happens (as it will) the effect will be very large indeed. But I can refine my argument. I can say, as Mark Budolfson does[10], that we know something about meat-producing supply chains. We know that meat producers always produce a bit more than is normally sold (which goes to waste or is repurposed when it expires on the supermarket shelves) & moreover generally that there are small but reliable buffers at every step of the supply chain. This introduces more than enough noise in the signal to make any one purchase invisible to the producer’s eyes.[11] The signal is “absorbed” by the buffers. Put differently, in deciding not to buy a piece of meat, one has no greater chance of making a difference than does the person who casts a vote in a U.S. presidential election from a state like Vermont or Wyoming.[12]
So goes the argument from supply chain buffers. I have tried my best but am forced to admit that not only am I not able to refute this argument, but I’m not even sure I understand it. To see why, imagine for example a scenario where the populace of some city or country decides, for cultural reasons, one by one to stop buying meat. If the inefficacy argument were true, the first person’s doing so would not affect meat production, because the signal would be lost in the noise & absorbed by supply chain buffers. But the same is true for the second person, & the third, & the fourth & so on. Gradually consumption would drop towards zero but meat producers, having not at any point in time discerned a significant drop in demand, would blissfully go on producing the same amount that they always produced.[13]
But this is absurd. The argument proves too much. I don’t think the inclusion of supply chain buffers adds anything to it. It just shifts it sideways, so to say. Because it’s clear, I think, to those who manage each stage of the supply chain roughly how much of the buffer is used or sold at that point & how much of it goes to waste, in the same way that the store manager knows how much meat they’ve sold each day & how much they’ve had to throw out.
Say a supermarket expect to sell around 100 sirloin steaks in a day, perhaps 90 on a bad day & 110 on a good day. Even if they stock 110 steaks, which of course is more than they sell on average (giving them a reliable buffer), there will nevertheless be a threshold below which they decide that too many steaks go unsold & they need to reduce their stock. This doesn’t change if their desired buffer is zero percent, ten percent or 100% because the quantity stocked is still a product of the expected quantity sold. I’d assume that each preceding step of the supply chain operates similarly.
Final Thoughts
McMullen & Halteman[14] mention some further reasons to believe that moral vegetarianism is effective:
When you buy a chicken, you are in a way buying more than a chicken. That’s because it took a captive hen to breed that chicken. So by reducing demand for chicken you are also reducing the need for egg-laying hens. The same goes for most other animals sold for meat.
By buying vegetarian, you encourage grocery stores & restaurants to offer more vegetarian options, which in turn may encourage other people to eat vegetarian food more often.
Meat is relatively cheap because the meat industry is large & benefits from economies of scale. The more people buy vegetarian, the smaller the meat industry gets & the larger the plant-based food industry gets, meaning that, all other things being equal, the price of meat will increase & the price of plant-based foods will fall.
We are fewer than 8 billion people in this world but every year we slaughter 80 billion non-human animals for meat. If that’s wrong, it’s a wrong of truly staggering proportions. It’s also one that is growing & will continue to grow as humans across the world get more prosperous. And it’s not something that technology will solve (though technology can & does help); what’s needed is nothing less than a great moral revolution on the order of abolitionism. That, if for no other reason, is why these questions matter.
Clare, J. & Farley, P. (2011). John Clare: poems. London: Faber and Faber.
I owe this observation to Korsgaard, Getting Animals in View.
Specifically they concern act consequentialism. Other variants, like rule consequentialism, escape unharmed from these arguments.
Gallet, C. A. (2010). Meat Meets Meta: A Quantitative Review of the Price Elasticity of Meat. American Journal of Agricultural Economics, 92(1), 258–272.
Selvanathan, E. A., Jayasinghe, M., Hossain, M. M., & Selvanathan, S. (2020). Modelling the Demand for Meat in Bangladesh. In Science and Technology Innovation for a Sustainable Economy (pp. 135–151). Springer International Publishing.
McMullen, S., & Halteman, M. C. (2018). Against Inefficacy Objections: the Real Economic Impact of Individual Consumer Choices on Animal Agriculture. Food Ethics, 2(2–3), 93–110.
ibid.
ibid.
Kagan, S. (2011). Do I Make a Difference? Philosophy & Public Affairs, 39(2), 105–141.
Budolfson, M. B. (2018). The inefficacy objection to consequentialism and the problem with the expected consequences response. Philosophical Studies, 176(7), 1711–1724.
ibid.
ibid.
It’s almost as if the argument is that the demand signal is a kind of emergent phenomenon – it’s there in the aggregate but invisible at the individual level. That makes me wonder whether there are general preconditions for emergence &, if so, what those preconditions are.
McMullen, S., & Halteman, M. C. (2018). Against Inefficacy Objections: the Real Economic Impact of Individual Consumer Choices on Animal Agriculture. Food Ethics, 2(2–3), 93–110.