The debate over lockdown between David Henderson and Justin Wolfers is extremely instructive. They take to heart the message of Ronald Coase—responsibly reckoning “the total effect” of one institutional arrangement as compared to another.
Henderson and Wolfers speak of externalities. In 2006 I published an EconLib piece “Rinkonomics: A Window on Spontaneous Order,” likening the spontaneous order on the floor of the roller rink to that of the free-enterprise system. The likeness lies in mutuality and coincidence of interest:
I call that a coincidence of interest. I liken the skating to voluntary exchange:
How well does the logic carry over to COVID? Let’s try, followed by caveats and concessions.
Broadly, there are three groups of people:
Group 1: Those who think they are still vulnerable to COVID.
Group 2: Those who think they are not vulnerable to COVID. (Because tests show anti-bodies.)
Group 3: Those who think they have COVID. (If they do, they may be contagious.)
For an individual in Group 1, the collision analogy is pretty strong: In promoting my interest in avoiding getting infected from you, I also promote your interest in avoiding getting infected from me. There is a coincidence of interest, and infection tends to be avoided.
For a Group-2 individual who is right about his or her not being vulnerable, there is no colliding to worry about, if that person can neither get infected nor infect others.
For an individual in Group 3, the coincidence of interest breaks down. It is like a skater with a hard spiky helmet and gear, among a throng of roller nudists.
By no means does the mutuality logic work as well as in the roller rink: People might think they’re not vulnerable when they are; vulnerable people cannot social distance always and everywhere; we’re not entirely sure whether someone who is post-COVID cannot get it again or cannot be a spreader. (If we learn that they can get it again, then those people revert back to being Group 1 types, where the collision analogy is pretty strong.)
Most importantly, people might be contagious without knowing it, and they are a grave danger. But that last condition lasts only so long, and people who do know it show a lot of concern for others. If you wore a hard spiky helmet and gear and were skating through a throng of nudists, how would you behave?
Moreover the roller nudists will be keeping their distance, never knowing who might be spiky. They skate defensively. Indeed, the roller rink may be empty, even under laissez-faire.
The logic carries over to some extent. The externality is not a random force. The points raised here relate to a key issue in the Henderson-Wolfers debate: Assessing the additional dampening achieved by locking-down. Maybe most of the dampening that has occurred under lockdowns would have occurred without lockdown.
With improving knowledge and moral responsibility, people can limit the spread without heavy-handed government. Social distancing and mask wearing happen pervasively without coercion. Lockdowns surely achieve some benefits, but lockdowns also hamstring and decommission local processes of focal learning, adjustment, leadership, and the taking of responsibility. By and large, targeted action is best taken on a voluntary basis by people close—socially and relationally—to the targets.
Talk of “externalities” is often slippery, unruly, opportunistic. (It often has negative externalities!) Clearly there is an important truth to the idea, but I’ve never gotten a firm handle on it. If I slam into Bob, I impose harm on Bob. But mutuality makes that “externality problem” a small one. And, can we not say that a lot of the benefits that flow from private enterprise rain down on people as “positive externalities”?
Maybe it’s better to avoid “externality” talk and instead frame the issue in terms of what Coase called “the total effect.”
READER COMMENTS
Dylan
Apr 28 2020 at 11:06am
So, for a bit of disclosure, learning about externalities is probably what hooked me on economics as an undergrad studying environmental issues. I really like Coase’s focus on transaction costs, but I find libertarian leaning folk often reflexively go to Coasean bargaining without diving too deeply into what that would look like, or why examples of dealing with externalities in this way seem so rare (perhaps this is just a bias, where issues that are successfully dealt with by bargaining are invisible to us, and other externalities are easily visible?) I also realize that you didn’t explicitly suggest Coasean bargaining as a solution, but instead focus on total cost, which I think is the right idea regardless of whether you think externalities or relevant or not.
However, I think your analogy breaks down specifically because there isn’t nearly the same level of mutuality with the virus as there is in a roller rink. You mention a number of caveats, but one that you don’t cover is the asymmetric nature of the risk. If I’m young and healthy, chances are that even if I get sick, the disease will be pretty mild (that’s not always the case for sure, the only two people that I personally know that have died so far were both relatively young and seemingly healthy), but for those I infect it could be much more serious or deadly. That doesn’t seem like mutuality to me.
Honestly though, even though I think externalities are a really useful concept. I’m agreed that I don’t know if that is the best way to think about this particular problem. I think the knowledge problem that you outline here is the more difficult.
You say that the condition of being contagious and spreading without knowing only lasts so long, that’s true, but doesn’t get at the crux of it, because until we have widespread testing and a better idea of immunity, the truth is that I’m not going to have any idea which one of the three buckets I’m in. Id I knew for sure that I was the one wearing a spiky helmet, then I’d of course be super-cautious, so cautious that I’d never even get in the rink with the nudists if I had any choice. But if I think there’s a good chance I’m in Group 2, and if I’m not in Group 2, then the alternative is I’m likely in Group 1 but probably not that much at risk, then I’m unlikely to modify my behavior by the optimum amount. Which I guess does get us back to externalities.
robc
Apr 28 2020 at 1:30pm
I think the vast majority of coasean bargaining is done implicitly, and may even be credited to “politeness”.
Putting up a fence to keep your dog out of your neighbor’s yard is the polite thing to do, but it is implicit bargaining, as the more expensive option is your neighbor shooting your dog (assuming you value the dog more than the cost of the fence). For the neighbor, the cost of a single bullet is cheaper than the cost of him putting up a fence. The bargaining is therefore straightforward and never even has to take place.
Dylan
Apr 28 2020 at 3:23pm
That’s a really good observation. I’ve noted neighbor disputes where this doesn’t seem to work very well, even though it seems like transaction costs should be pretty low, but I’ve not considered how basic manners could be interpreted as implicit coasean bargaining.
Jon Murphy
Apr 28 2020 at 2:09pm
To robc’s point, there is a ton of implicit bargains going on.
But also remember that Coase’s story is one of transaction costs. Once we factor in transaction costs, whether or not an externality can be profitably bargained away becomes the question. Only if the expected net benefits were miscalculated by the parties involved to be below zero when they are, in reality, positive and the parties should have known their calculations are off is there a policy-relevant solution.
In other words, the relative dearth of certain kinds of Coasian bargains does not imply either 1) a failure of Coasian bargaining or 2) a policy-relevant solution exists. All that it does imply is that the transaction costs may exceed the benefits.
Michael
Apr 28 2020 at 12:34pm
Don’t we also (unfortunately) also have a Group 4: People who don’t take the threat of Covid-19 seriously? They are vulnerable and potentially contagious, but take no precautions.
Mark Brady
Apr 28 2020 at 9:19pm
And what about all those in Group 5: People who wittingly or unwittingly exaggerate the threat of Covid-19 for whatever reason…unreasonable fear, ignorance, inability or unwillingness to appraise news stories and narratives, and/or a propensity to use a crisis as an opportunity to perpetrate their own agenda?
Michael
Apr 29 2020 at 6:32am
Your phrasing suggests a strong prior.
Mark Brady
Apr 29 2020 at 8:20pm
And I suggest that it takes a strong prior to deny the existence of Group 5.
Daniel Klein
Apr 29 2020 at 2:43am
Thanks all for the nice comments.
Smithian propriety is a richer and larger notion than politeness or manners.
Smith expressed great faith that bottom-up processes of propriety formation among jural equals regarding practical non-political affairs tend to align to the social good (or as Coase would say, “the total effect”). That’s part of Smith’s reasoning against the governmentalization of social affairs.
He had less faith about those processes when it came to political sentiment and opinion.
robc
Apr 29 2020 at 12:57pm
Propriety is a much better term than politeness. Thanks.
Comments are closed.