A key insight of the Coase Theorem is that externalities are reciprocal. Yes, a polluter imposes a negative externality on his neighbor. But if the neighbor insists on clean air, he imposes a negative externality on the polluter. While common-sense morality may urge you to take the side of the neighbor, economic efficiency urges you to keep an open mind. If the polluter’s cost of reducing pollution greatly exceeds the neighbor’s cost of enduring pollution, the Coase Theorem tells you to tell the neighbor, “Tough luck. Suck it up.”
This Coasean insight is deeply relevant to COVID policy. It’s also been almost entirely ignored. Yes, people who don’t wear masks impose negative externalities on others. But people who insist on masks impose negative externalities, too. Efficiency requires both sides to consider the burden they’re imposing on the other.
Is the cost of wearing masks ever actually higher than the cost of enduring COVID exposure? Definitely. Suppose ten healthy young people all work in an office from 9-5 on weekdays. Once a week, an immuno-compromised senior citizen stops by for five minutes. The unmasked workers definitely impose a tiny negative externality one senior. But if you require everyone to wear masks all the time, you impose a large negative externality on all ten young workers. The efficient outcome would probably be to tell the senior to stay home if he’s nervous – not tell everyone else at the office to remain masked forty hours a week to accommodate him.
You might reply, “Forcing everyone to wear masks is inefficient, but we should still follow common-sense morality.” I’m sympathetic, but is common-sense morality really on the senior’s side? Not really, for two reasons:
1. Voluntary assumption of risk. Every job has problems, including a bundle of risks. The risks are unacceptable? Common-sense morality’s standard reply is: “If you don’t like your working conditions, quit.”
2. De minimis. Even if you don’t voluntarily assume a risk, common-sense morality says that the expected severity of harm matters. If the expected harm is trivial, you’re free to inflict it. Example: I risk your life whenever I drive in your vicinity. You don’t consent, but common-sense says I’m still entitled to drive. Why? Because the expected severity is low. You could protest, “Only because you’re liable for any harm if it occurs.” But in the real-world, imposing such liability is easier said than done. After all, a lot of people are judgment-proof. While you could heavily restrict the freedom of everyone who fails to post a $1M bond, common-sense morality strongly condemns such measures as tyrannical.
To state the obvious, I respect not only the individual right to wear a mask, but property-owners’ right to require a mask as a condition of entry. But not only do I have a strong presumption against any stronger legal support for mask-wearing. I also think that informal norms should take Coase’s notion of reciprocal externalities seriously.
READER COMMENTS
N
Feb 2 2021 at 9:21am
Interesting argument, yeah.
But the world is not like your toy example. People are complexly networked; not wearing a mask and transmitting covid (for instance, between the many young workers in your example) risks being part of a transmission chain that directly causes death and suffering many iterations away. Particularly so when a vaccine is looming and we just need time to distribute it. A broad mask mandate seems like a reasonable policy instrument in such a complex and volatile system.
Mark Brophy
Feb 2 2021 at 7:16pm
The “vaccines” attempt to stop the symptoms of the virus but do not prevent transmission of the virus to others. Essentially, it’s a fraud to call them vaccines.
JFA
Feb 3 2021 at 7:56am
“it’s a fraud to call them vaccines”
Ummm… what?
Pedantically, they are called vaccines because of Edward Jenner’s use of cow pox to inoculate people against small pox (“vacca” being Latin for “cow”).
Non-pedantically, the data showed high efficacy of preventing symptomatic cases of Covid, and yes, there is no study showing that it does or does not prevent the spread. But here’s how you would have to interpret the data to think that it doesn’t prevent spread: the symptomatic cases that would have occurred absent vaccination are turned into asymptomatic (in the sense that the person never develops symptoms) cases by the vaccine and that the number of asymptomatic cases that occur in the absence of vaccine are not reduced by the vaccine. Another way of putting this is that you would have to think that there are the same number of infections in those with the vaccine as those without but that there is just a much larger share of asymptomatic cases in those with the vaccine. Then, you would have to ignore the studies that show that truly asymptomatic cases (as opposed to pre-symptomatic individuals) spread the virus much, much less than those individuals who eventually develop symptoms.
Even if that unlikely scenario is true, preventing most serious cases and hospitalizations while also providing nearly 100% protection of Covid deaths seems like a pretty good result from a vaccine, no?
Jon Murphy
Feb 3 2021 at 9:17am
Yes, but that is true no matter what. As Bryan said, the externalities are still reciprocal. The face externalities exist do not recommend one policy over another.
N
Feb 4 2021 at 12:29pm
But when the externalities are absolutely massively higher on one side (the non-maskers), it absolutely does recommend one policy over the over. That’s my point.
Rohit
Feb 2 2021 at 9:35am
This strikes me as a utopian analysis that only works if everyone is homo economicus.
1) You don’t know nearly well enough the harms that not wearing masks places. In your example do those healthy office goers disapparate at the end of the day or do they meet others?
2) It’s way easier to make a simple, easily followable rule like “mask up” rather than let the market and common sense dictate. For one there’s almost no indication that a large % of people have a sufficiently high knowhow to make the decision. We’re reliant on experts, even if armchair, to tell us what to do.
3) When you have highly diffuse negative externalities imposed vs more concentrated ones on the other hand, it’s easy to say net harm is the same. The error bars on showing that is enormously wide though! One disutility is hospitalisation and death. The other is annoyance. There’s a difference in calculus there and this is a corner case where utilitarianism always hits societal failure mode.
If you disagree with the last point, please know my disutility from Bryan Caplan not thinking of me as a consummate genius is much higher than your indifference to an internet comment, so you should do everything possible in your power to rectify that immediately.
Tyler Wells
Feb 2 2021 at 2:38pm
I may be misunderstanding you, but I think that this is backwards. The masks and social distancing rules are there to put some of the onus of mitigation of the virus onto the young, healthy and less-risk with the goal of minimizing the inconvenience to the risk averse population. The risk averse could simply stay at home and order their essentials online but, because that would suck, they prefer to rob the children of their childhoods.
Personally, I have no problem wearing a mask to the grocery store. What makes me irate is the closing of the schools and parks.
Alabamian
Feb 2 2021 at 9:37am
Fixed it for you: If the polluter’s cost of reducing pollution greatly exceeds the neighbor’s cost of enduring pollution, the Coase Theorem tells you to tell the neighbor, “Good news – I’ll be happy to compensate you for your unclean air.”
There is a distinction between imposing externalities as an initial matter versus avoiding or regulating activities that create externalities. You characterize them as if they are the same. They aren’t. To be sure, it is always more convenient for me if I don’t have to internalize the externalities I create. But asking me to internalize the externality (or avoid the activity) is not itself an “externality.”
There is no market mechanism by which we can internalize the cost of passing COVID to someone else. See 1 above. In the absence of such markets, coasean reasoning is of limited use, and we need things like common sense morality to make sense of the situation. Mindless utilitarianism gets us nowhere since, in the absence of any sensible market, you can justify any outcome by simply manipulating the (totally made up) utility curves. Take this, for instance: “But if you require everyone to wear masks all the time, you impose a large negative externality on all ten young workers.” I think most people who have been wearing masks for the past year will tell you that this is, to use a technical term, “bullshit”. But, hey, the math checks out if you make these assumptions, so yay it works!
Overall, this is a bad take. It’s a myopic view of the world.
Alexander Turok
Feb 7 2021 at 1:23pm
Have you been wearing masks ~40 hours a week continuously, or only occasionally?
Tyler Wells
Feb 2 2021 at 10:59am
Reciprocal externalities is the right way to look at COVID response. My child benefits almost not at all from any (potential) reduction in the virus due to mitigation efforts like social distancing or masks, but has missed school, sports, birthday parties, summer camp, and so many of the activities that help young people grow up to become adults due to those efforts. Many of those who write here are elderly and rich and suffer only minimal cost from those same mitigation efforts. In fact, they inconvenience their lives less by imposing the burden of theoretical virus mitigation on the millions of low-risk individuals, like my child, instead of staying at home and isolating themselves.
Ultimately, the decision of what to do is political one and not moral. This is natural due to the imbalances in the externalities.
David Henderson
Feb 4 2021 at 11:14am
Nicely said, Tyler. My heart breaks for the two young kids next door who are very social, and very well-behaved, and have been separated from their friends for almost a year.
Dan Cohen
Feb 2 2021 at 11:22am
How do you take Coase into account when there’s somewhere with many different unrelated people (e.g. the grocery store)? The transaction costs are too high to negotiate with every single person nearby.
robc
Feb 3 2021 at 6:38am
That specific answer is easy. When the transaction costs are too high, you eliminate them altogether, by having your groceries delivered or doing pickup.
Are any major grocery chains not doing parking lot pickup?
Dylan
Feb 3 2021 at 12:31pm
Well, considering we don’t have grocery stores with parking lots in this area, I’m going to say the answer is yes.
Philo
Feb 2 2021 at 11:35am
“Efficiency requires both sides to consider the burden they’re imposing on the other.” But how useful will this consideration be, when, in the absence of markets, the burdens are so hard to quantify? Even your artificially constructed example involved a lot of arm-waving. In theory, your Coasian perspective may determine the optimal choice among possible alternative rules about mask-wearing; in practice, it will not help much.
James
Feb 2 2021 at 11:45am
Some people want to use the Coase theorem to determine what people should do, what rules people should make, etc. Since this application turns the Coase theorem from a statement of positive economics to an ethical foundation, people who want to use the Coase theorem in this way should offer some ethical argument, or just not use Coase in this way.
If I have a youtube channel where I sneeze on people and record their reactions, should I be curtailed or should my victims just suck it up? If we take the Coase theorem to be an ethical principle, the answer depends on how much money my youtube channel is making and how much harm my sneezing actually causes. My example is silly, but a little creativity should be sufficient to come up with more realistic scenarios that expose the weaknesses of trying to derive normative conclusions from a theorem that was not about norms in the first place.
Rohit
Feb 2 2021 at 11:47am
This strikes me as a rather utopian analysis that only works if everyone is homo economicus.
We don’t know nearly well enough the harms that not wearing masks places. For instance, in your example do those healthy office goers disapparate at the end of the day or do they meet others?
It’s way easier to make a simple, easily followable rule like “mask up” rather than let the market and common sense dictate. For one there’s almost no indication that a large % of people have a sufficiently high such knowhow to make the decision. We’re reliant on experts, even if armchair, to tell us what to do in a crazy pandemic.
When you have highly diffuse negative externalities imposed vs more concentrated ones on the other hand, it’s easy to say net harm is the same. The error bars on showing that is enormously wide though! One disutility is hospitalisation and death. The other is annoyance. There’s a difference in calculus there and this is a corner case where utilitarianism always hits societal failure mode.
Evan Witt
Feb 2 2021 at 1:58pm
I don’t think the argument hold’s up very well, for a few reasons.
One is that the spreading nature of a disease is relevant. The point of mask requirements isn’t just to protect that five-minute interaction with an immuno-compromised elderly person. It’s to slow the rate of spread. I couldn’t argue that I should be allowed to build a fire in at my campsite (during the dry season), with the only externality being whether one or two of my neighbor’s tents are burned. Fire (and disease) don’t work like that. Those suffering from Covid now are paying for mistakes made by people a year ago, several steps removed from them in the infection trail.
Thomas Hutcheson
Feb 2 2021 at 2:33pm
Each venue and situation is different. The “standard case,” wearing a mask when occasionally among other people indoors (e.g. shopping) seems to easily pass the net benefits test. In the situation presented, it might be better to prohibit entry by vulnerable people into the risky office environment.
It is sad that criticism of COVID-19 regulations have not been more specific and aimed at increasing their net benefits. [Of course this is partly a failure of public health professionals to have done the research to permit lower cost regulation and voluntary behavioral change.]
Nicholas Decker
Feb 2 2021 at 6:09pm
It would seem to me that a mask law is akin to a law against driving whilst drunk. It is, quite simply, people telling others that you are grossly misevaluating the risk, and you doing so is putting everyone else at risk too. Just as we might think it right to physically prevent someone from driving themselves home while intoxicated, so too should we compel the wearing of masks when you are interacting with other people.
notsocommonsense
Feb 2 2021 at 10:58pm
These two examples of “common sense morality” are far from widely held.
Many people treat their working conditions as legitimate subjects not just for individual negotiation but various types of collective action and regulation. They have a sense that (for example) legal restrictions and bargaining agreements that prohibit abusive treatment by bosses or constrain the hours they are required to work are a good thing, and are indignant and outraged when they hear of cases where people are subjected to unsafe, unpleasant, or interpersonally cruel working conditions that could reasonably be mitigated.
People do, I think, widely share your intuition that severity of harm matters, but not in a strict expected value sense. Many people share a sense that imposing small, certain costs is acceptable to reduce or mitigate small probabilities of great harm. Many people support drivers licensing (with the possibility of revocation for repeatedly or recklessly endangering others), safety standards, safety-focused laws (e.g. seatbelt laws) and insurance mandates to constrain the entitlement to drive and minimize or mitigate harms caused.
I understand that Bryan disagrees with many of these example policies, and may believe that most people are exhibiting social desirability bias rather than a rational belief that their policy preferences are welfare-enhancing. But to assert that broadly shared informal norms support his position badly misunderstands what peoples’ actual norms are. The most important one–which I think is at the root of Bryan’s attraction to unpopular positions–is that as a matter of ‘common sense morality,’ people feel they have rights not just to act without interference from others, but to constrain others from directly or indirectly causing them significant harms.
David Henderson
Feb 4 2021 at 11:21am
This is well-argued. I’m not saying that I agree with you, but you do argue it well. I’ve got to think about it further.
One correction, though. Seatbelt laws are not a good example for the point you’re making. In fact, they’re a terrible answer. Requiring people to wear seatbelts reduces the cost of driving aggressively, putting others outside the car at greater risk.
AMT
Feb 4 2021 at 1:56pm
I agree with David that the seatbelt example is not good, because that to me is entirely a paternalistic regulation with a very limited externality justification. The benefit is almost exclusively received by the person being “burdened” by being forced to wear their seatbelt. (It might be more complicated that with multiple people in a vehicle you could hurt others in the car in an accident, but that would only justify a law requiring seatbelt use when others are in the vehicle, not if you are alone, because I’m pretty sure there is almost zero chance you end up hurting someone else if you are alone.) So essentially we decide for people that it passes the cost-benefit analysis. I can see the argument it increases the externality risk to others, although I think it likely still would pass a cost-benefit analysis.
But the main justification for forcing wearing a mask is the negative externality individuals place on others (but also the benefit to the individual by reducing their own risk of getting sick).
I have no idea why people complain about being forced to wear a mask, when they have a far, FAR stronger argument to complain about seatbelt laws, which they seem to not care too much about, if at all.
Overall, I just disagree with Bryan’s point because I think “you impose a large negative externality on all ten young workers.” is incorrect, and if we are going to talk about “de minimis” anywhere in this discussion, it fits best for regarding the inconvenience of wearing a mask.
BC
Feb 3 2021 at 2:14am
Maybe the best example that externalities are reciprocal is peanut allergy, which can be lethal for young children. Even before the Covid pandemic, some parents have tried, with some but not widespread success, to ban peanuts from school lunch rooms and other public places. Their argument is that peanut particles impose an externality on their peanut-allergic children, much like a polluter releasing a pollutant. They also elicit additional sympathy because, in many cases, it may be unrealistic to expect young peanut-allergic children to be conscientious enough to avoid sitting too close to those eating peanuts or to carry epi-pens with themselves at all times. Many people, however, are quite opposed to peanut bans because they see the externality running the other way: the desire to create a peanut-free environment imposes an externality on those that would like to consume peanuts by limiting their food choices. The fact that we have a longstanding history, custom, and tradition of eating peanuts makes peanut consumption seem different from polluting: eating peanuts is “normal” behavior, unlike emitting pollutants into the atmosphere. Our custom suggests banning peanuts may infringe a person’s natural rights in a way that banning pollution does not. Because most people both have sympathy for peanut-allergic children and view peanut consumption as a part of “normal” life, they can grasp that the externalities in this case indeed run both ways.
In many other cases, people don’t recognize the reciprocity of externalities because they have little sympathy for one of the parties. For example, people easily grasp the externalities that carbon-emitters impose on others but not the reverse. In fact, limits on carbon emissions do damage the property values of fossil fuel owners in the same way that starting a fire on one’s own property can cause fire damage to one’s neighbor’s property. People that don’t recognize anti-Covid measures as imposing externalities in the same way that peanut bans impose externalities generally don’t think wearing masks and other measures are “that big of a deal”.
Having set all these precedents with anti-Covid measures, it will be interesting to see whether in the post-Covid world we will see more widespread peanut bans. Covid may have weakened the pre-Covid notion that people have a natural right to engage in “normal” activities, as defined by history, custom, and tradition — activities like eating peanuts in public.
Thomas Hutcheson
Feb 3 2021 at 5:32am
Interesting example. I still find “reciprocal” an odd word for what seemed like the idea that either party to an externality can take action to mitigate the harm.
Peanut particle pollution, however,does not seem a good match for COVID in that the harm is not general, but binary for a small group, shifting the presumed least cost mitigation strategy to the receiver of the externality.
robc
Feb 3 2021 at 2:48pm
Gravity doesn’t “seem” reciprocal either, but it is. I am pulling the Earth up with the exact same force that the Earth is pulling me down.
The reciprocal nature of externalities is exactly like Newton’s 3rd law.
LeeN
Feb 3 2021 at 9:55am
So according to your last statement, if government is considered the owner of the land they could in fact require mask wearing.
I also find it strange, as intelligent as people are here, none of them seems to have done any actual study on the efficacy that masks actually work. And that if masks do not work, forcing others to wear a mask might in fact be creating an even bigger hazard by making people feel safer than they ought to.
LisaS
Feb 5 2021 at 11:13am
I find it interesting how many of the commentators call mask wearing low cost. I would disagree with that assessment. Mask costs include increased theft and homicides, less social interaction, increased loneliness, increased fear (both of Covid and of meeting masked men in a parking garage or on a street at night. ) None of that is negligible. And if you want an explicit cost look at the rate of hearing aid replacement this year. It is far too easy to catch them on a mask string and flip then off. One has to wear them though if there is to be any hope of alleviating social isolation.
Tim Townsend
Feb 5 2021 at 11:16am
Masks are not, if at all, that effective ( pre 2020 medical literature ) and the best vaccines for flu are generally 70% effective in any given year.
John Halstead
Feb 6 2021 at 6:06am
Widespread community transmission of covid creates massive externalities for pretty much everyone in society. If people go to a nightclub and don’t wear masks that increases community spread of covid, which down the line increases the risk that old people get the virus. Similarly, widespread transmission in schools is not dangerous to children, but it is dangerous to other people because the virus predictably eventually makes it out into the much more at-risk adult population – children live with adults, as it happens.
Widespread mask wearing reduces transmission by 50%, so it predictably massively reduces these society-wide externalities.
Re – if you don’t like the risk, quit or don’t go out. Comparably, if you don’t like the risk of being killed by a drunk driver, don’t drive. Well, we still have laws against drunk driving. I could even come up with ridiculous edge cases where the drunk driving imposes lower costs than the risk of a greater road traffic incident. Much like you have done here, Bryan.
It’s hard not to see this post in the context of other libertarian thought during the pandemic. All of the rhetoric of the philosophy is that people have absolute rights not to be aggressed against (“persons are separate”). But that all goes out the window when a virus is killing hundreds of thousands of people. Instead, all of the focus is on burdens that are in general laughably small, like mask wearing.
Student of Liberty
Feb 7 2021 at 6:32am
This needs to be reminded over and over.
In the case when my conviction gives me the feeling that the economic efficiency is the right thing to look at, it is easy but there are so many cases around when an aggression on my sense of morality makes me uncomfortable that I understand why most people are – unfortunately – not ready to hear the argument and take the side of the “polluter”.
Comments are closed.