I thought the debate went very well. Justin Wolfers is a nice guy with a sense of humor. I particularly liked his joking around before the debate when Gene Epstein asked him how he met his American wife. I also liked his Aussie humor about New Zealand. At one point in the debate, I almost summed up my case by saying “It’s the vibe.” Fans of the Australian movie The Castle would get the humor, and I’m guessing that Justin is a fan.
At the end of my rebuttal to Justin Wolfers, I said that one never knows what the tone of the person on the other side of the issue will be and I thought he was a perfect gentleman who didn’t pull any cheap or nasty debating tricks. So we really did have a contest of viewpoints.
The on-line version of the debate will be out next week and so I will have more to say about it then. I want to point out one thing that I mentioned but wish I had emphasized in my summary at the end.
Justin said that we should choose the least-cost way of getting a given amount of safety. Previous to that, I had focused on how incredibly costly the lockdowns are. One of the people on the chat room, Mark Bahner, who often posts comments on EconLog, asked “What if we could get the desired level of safety with everyone who works close to others wearing masks and gloves? Would that be acceptable.” I missed most of the chat comments but because I noticed Mark’s name, I read his comment and I asked his question of Justin. If I recall correctly, Justin answered that yes, that would do it, if we could trust people to comply, but he wasn’t confident that we could. I should have pushed him on his lack of confidence in people, especially compared to his overconfidence in government.
What I wish I had emphasized in my summary is that requiring gloves and masks, with substantial resources devoted to enforcement is way less costly than shutdowns. I might balk at the amount of enforcement required to get, say, 90% compliance. But my gut feel is that for the first two months, which seem to be the crucial time period in most people’s estimates, compliance would be above 70% without much enforcement.
READER COMMENTS
Dylan
Apr 22 2020 at 9:20pm
It was a good debate, I found both of you to offer valuable insights and reasonable points. Being a regular reader here, you were probably at a bit of a disadvantage from my point of view, because I felt I was pretty well acquainted with most of your main points already, and had already had time to process and think of my objections. While Justin offered some points I had not yet considered.
I had to leave a few minutes early, but there were a couple of points that were mentioned by each of you that I think the other didn’t seem to adequately respond to.
1. You brought up the Zingales study that you had written about here recently on the statistical value of a human life, and the weird results we get when we start taking that to really weird numbers. Would have liked to hear Justin address this better.
2. For you, I don’t recall you really acknowledging Justin’s point about if you assume that much of the social distancing that we have right now would happen government mandate or not, then you really can’t attribute all or even most of the economic costs to the mandate. This point of his could have been made stronger by referencing the study that looked at the economic impact of different cities responses to the Spanish flu, that found that the cities that had stronger restrictions experienced shallower downturns and quicker rebounds. I’ve not looked at the study directly, just summaries, so not sure if it holds up…but it does provide some support that the lockdown may actually be better for economic activity in the medium and long-term than no lockdowns.
Michael
Apr 22 2020 at 9:31pm
I haven’t seen the debate, but given David’s brief review of it here I will watch it when it is available.
I think Justin’s point that you raise in #2 sort of cuts both ways. One could argue that there is no need for mandated lockdowns at all if the public will (mostly) lock themselves down by choice anyway. But Justin’s take also seems valid.
Dylan
Apr 22 2020 at 9:53pm
Yeah, that was basically Justin’s point, that if you were going to take credit for the benefits of people voluntarily social distancing in terms of preventing lots of excess deaths, then you can’t also claim the economic damages compared to a world without coronavirus.
I think that David would reply that the voluntary choices will tend to do things in a more efficient way than brute force government mandates, meaning you get more benefits for every $ of cost. That’s something I’m generally inclined to agree with, but I think the externality argument needs to be better addressed before I’d be confident in that.
David Henderson
Apr 22 2020 at 10:11pm
Actually, Dylan, that’s how I did reply. I said that you couldn’t just assume that lockdowns just a more extensive form of social distancing. People who choose social distancing want to achieve things while doing it. Lockdowns don’t let them. I forgot to mention this in my closing but I definitely mentioned this in the Q&A.
And thanks for the note re Zingales. I agree that he definitely didn’t reply to the reduction ad absurdum.
Kevin Dick
Apr 23 2020 at 12:06am
I think comparisons to the Spanish flu are not very useful in this context. As pointed out on Marginal Revolution a couple of days ago, the two diseases are different in a very crucial way:
Spanish flu was most dangerous to health working age people.
Covid-19 is most dangerous to the non working age and those with underlying conditions.
This makes the cost-benefit of general lockdowns vastly different.
Not to mention that the IFR estimates for Covid seem headed to a much lower value than Spanish flu. Though even if they were identical, the different demographics of the fatalities would almost certainly lead to very different optimal responses.
Alan Reynolds
Apr 23 2020 at 9:34am
The only fatality rate estimate I have seen for the 1918-19 H1N1 “Spanish” flu were for a case fatality rate, not an infection fatality rate (which would require knowing the number of mild cases cured at home). The usual case fatality estimate is 2.5% for Spanish flu. But a H1N1 virus also caused the 2009-10 pandemic is now a part of the seasonal flu, yet nobody thinks that particular flu virus is much more deadly than a bad flu season like 2017/18. Of course we didn’t have herd immunity, vaccine or Tamiflu in 1918, or antibiotics to cure related bacterial pneumonia.
Michael
Apr 23 2020 at 12:42pm
New New York antibody data suggests that 21% of NYC residents have been exposed, along with 16.7% of Long Island residents, 11.7% of Westchester resistents, and 3.6% of residents in the rest of the state. Overall estimate for the state is anout 14% exposed.
NYC data imply an IFR (not CFR) of 0.82. LA county data suggest a bit lower 0.36%.
That would place this as much more severe than the seasonal flu (not talking 2018 Spanish flu here). The seasonal flu has a CFR of 0.1%, but an IFR much lower than that (we have asymptomatic flu cases too).
https://twitter.com/foxjust/status/1253353204911411202?s=20
JFA
Apr 23 2020 at 1:50pm
Thanks for sharing that. It’ll be interesting to see whether that study holds up once its methodology and (at least) summary data are released. Two similar studies from California have not held up under scrutiny, but I’ll withhold any pre-judgment until the study is released.
Michael
Apr 23 2020 at 3:34pm
At this early stage, it seems reasonable to think of that number more as a rough ballpark estimate as opposed to a precise measurement, given potential methodological issues.
Larry Campbell
Apr 23 2020 at 12:44am
Haven’t heard the debate, so thanks for the taste of it. My sense of the issue, though, comes down to this:
We can suppose two general situations in the face of a pandemic — one in which the state intervenes only in positive ways, for example to aid in the distribution of needed infrastructure; and one in which it intervenes also in negative ways, to enforce “social distancing”, for example, by mandating anything from business closures to, in effect, house arrest. In the first case, we have harm resulting from the disease itself directly, but also from the economic and social damage resulting from voluntary distancing. In the second case, the idea is that there is supposed to be some lessening in the harm from the disease, but at the cost of some increase in the harm from enforced distancing. The issue here, then, is whether, in the second case, the reduced harm from the state mandates compensates for the increased cost of those mandates. In both cases, the disease itself causes harm, both directly and indirectly, and in both cases, that harm is mitigated by social distancing of one degree or another. The question is simply whether state mandated social distancing is effective in reducing the direct harm from the disease over what would naturally occur anyway, enough that it at least balances the indirect harm resulting from a broad-brush state shut-down of economic activity over what would occur naturally. I think, for this disease, the case that it does not is strong and growing.
Dan Hill
Apr 23 2020 at 12:12am
I feel bad taking sides against a fellow Aussie, but I disagree with his fundamental premise:
Why assume that safety is the fixed parameter? Why not look for the safest way of avoiding a given fall in output, especially if small increases in safety lead to massive economic losses? Or as it might have played out in The Castle:
Better still, let individuals make the safety/cost trade-off for themselves. “But voluntary distancing won’t work perfectly!” That’s always the argument for nanny state big government. As if government ever delivers perfection.
And now I have to go watch The Castle. Again. Good thing I’ve got plenty of time on my hands 🙂
P.S. If you like Michael Caton (Darryl Kerrigan), you really should check out Last Cab to Darwin. More of a dark comedy but very moving.
Michael
Apr 23 2020 at 6:23am
Here you are lampooning a weaker argument in favor of lockdowns while ignoring the stronger one. In this context, individual personal decisions about risk affect not just the individuals who make those decisions, but everyone.
robc
Apr 23 2020 at 10:01am
That latter is not true. If you choose to self-isolate, then short of me breaking into your house to force interaction on you, there is nothing I can do to increase your risk.
If you choose to interact AT ALL, however, yes, you are taking on part of the risk based on others too. But you are choosing to take on that risk via your voluntary interactions.
Mark Z
Apr 23 2020 at 10:57am
It clearly is true. That the realization of the risk is conditional on one’s behavior doesn’t mean the cost doesn’t exist. If I emitted fatally poisonous gas outside your house, which you can avoid by not going outside, it is obviously true that my decision to do so affects you: it forces upon you the previously nonexistent tradeoff between going outside and staying alive. The fact that you only die if you choose to go outside mean an external cost isn’t being imposed on you. Consider what it would mean if we applied this reasoning to traditional infringements: if someone threatened to physically harm you if you went outside, and you went outside and they accordingly harmed you, would we say, “well, you voluntarily chose to go outside, so the responsibility for the harm is all on you?”
AMT
Apr 23 2020 at 9:41pm
robc,
It’s a “negative externality.” Look it up. It might help you for reading an economics blog…
Mark Bahner
Apr 23 2020 at 12:40am
YES! YES! YES! 🙂
As I told you in my email, David, I feel like for more than a month, I’ve been in some Twilight Zone episode! It’s like I can see something–something incredibly obvious–but no one else seems to see it. Finally, two other people have seen it. Even someone who is not on “my side.” 🙂
As I’ve written before, I’ve purchased 15 pair or washable cotton gloves for $13. I’ve bought 3 washable masks, I think for another $15. I bought a balaclava (aka, “head sock”) for $10. And I bought a baseball hat with a plastic face shield for…I think it was $15 with shipping. So that’s like $50.
And people could buy a respirator hood from Amazon for $23, that would take the place of–possibly be even better than–the balaclava/”head sock” and the baseball hat with the face shield.
Amazon respirator hood with face shield (please do not buy unless you really need one!)
So we’re talking about maybe $50 per person. But let’s call it $100 per person. There are 330 million people in the U.S., so at $100 per person, that’s $33 billion.
But we don’t even need to spend anywhere near that! New York and New Jersey have had almost half of the U.S. fatalities, even though they only have one-tenth of the U.S. population. And even the fatalities in *those* states are concentrated around NYC.
For the life of me, I just cannot understand why no one seems to recognize that there is a monumentally less expensive alternative to save lives than “shelter at home.”
David, if you or any economist who reads this would like to do a paper with me along the lines of, “What would be the least expensive way to achieve the same results as ‘shelter-in-place’?” please let me know in the comments.
We need to stop this madness (“shelter in place”). Or at the very least, make sure it never happens again. (Like this fall or early next year.)
AMT
Apr 23 2020 at 11:42am
This is all very good, in theory. I’m just wondering if we would actually be able to supply enough for the whole population, given what we have all heard about the persistent PPE shortages, for medical providers alone. “please do not buy unless you really need one!” basically says it all…
And are washable and reusable items like cotton gloves just as effective as disposable latex gloves or N95 masks? If they were, wouldn’t hospitals be using them to save costs and deal with the shortages they are experiencing?
Mark Bahner
Apr 23 2020 at 8:35pm
It’s trivial to provide what I’m talking about for the entire U.S. population. But it’s also completely unnecessary. As I pointed out, nearly half the deaths in the U.S. have occurred in two states with only one-tenth of the U.S. population. Further, most of those deaths were concentrated around NYC. And to go even further, I doubt more than a couple dozen were in people under 20 years of age. So the real need is far, far less than the theoretical idea of buying all those things for every man, woman, and child in the U.S. (Although even getting those things for every man, woman, and child is an absolutely trivial matter…compared to shutting down the economy, anyway.
No, it doesn’t say it all! Your saying that it “basically says it all,” shows that you don’t understand at all.
What it says is that there *other* things that are nearly as good, and leaves the respirator hood for people who truly need that extraordinary and very specialized protection. That respirator hood is a “positive pressure” hood. It’s made with a hose connection to deliver a positive supply of filtered air inside the hood. I don’t think even ER doctors wear that kind of protection. (Note: I have no medical background, but I’m a retired environmental engineer who specialized in air pollution, with particular expertise in particulate matter and particulate matter filtration.)
Before I mentioned the respirator hood, I mentioned a balaclava/”sock sock”. Here are examples:
Balaclava
Head sock
Buy them! Wear them! Even if you really don’t need them. (Which you probably don’t.)
And I mentioned wearing a washable face filter *underneath* the balaclava/”head sock”:
Face filters, including washable filters
Buy those too! Wear them underneath the balaclava/”head sock”.
And I mentioned a baseball cap with a plastic face shield…but there are other hats with face shields that are even better:
Hats with face shields
Buy those too! If you wear: 1) a hat with a face shield, 2) a washable face filter, 3) a balaclava/”head sock” over the top
No, they’re not. Only an ignorant fool (like Donald Trump***) would claim they are. But the threat to you is not even close to the threat to medical workers, who are often dealing with patients *without face mask or other protection* talking and coughing right in their faces, as the doctors lean over the patients to treat them.
If you buy and wear all four of the things I told you to buy and wear, you’ll be extremely safe from catching COVID-19, unless you’re doing something like hugging an infected person or standing right in front of someone who is coughing. And you can buy all of those things for approximately $50.
P.S. You could modify the hat with the plastic face mask to attach cloth so that the bottom of the face mask lies on your shoulders all the way around, just like “positive pressure” respirator hood, and you’d be even safer. The risk would probably be reduced by close to 90 percent, compared to if you didn’t wear those things.
AMT
Apr 24 2020 at 7:15pm
I have a couple Trump quotes for you, and some basic logic.
January 22: “We have it totally under control. It’s one person coming in from China. It’s going to be just fine.”
February 26: “The 15 (cases in the US) within a couple of days is going to be down to close to zero.”
Ok. Pretty much no cases in America. Therefore, no reason for Americans to take any precautions! Not even in New York!
“I don’t know how exponential growth works.” – Mark Bahner
It’s “you’re.” And maybe the guy who literally doesn’t understand exponential growth understands things better than me. Or maybe the self-admitted medical non-expert is randomly pulling “the risk would probably be reduced by close to 90 percent, compared to if you didn’t wear those things” out of the air. I’m not a medical expert either, so I guess I’ll just counter by saying “False, they would only reduce the risk by 10%,” based on the equivalent, no evidence whatsoever.
I’m aware that Scott Alexander and Robin Hanson have both posted on how reducing the viral load exposure could have an impact on severity. I agree that it is likely that covering your mouth and nose with something probably helps some amount, but it’s highly uncertain how much, and it might matter more for illness severity than whether or not someone gets it at all (and subsequently carries it and exposes others). Further, it’s not clear how different viruses and in particular, this novel virus, would work along those dimensions. e.g. If you get a blood transfusion of one unit of HIV contaminated blood, are you “less” likely to have HIV afterwards than someone who receives 2 transfusions, or does it not really matter at all if a certain threshold is met? If you are aware of studies that provide that information, then that would be nice.
So , maybe I “don’t understand at all,” because I should have known where you pulled the “probably” 90% risk reduction from. Which doesn’t even specify if it’s the probability of actually contracting the disease, or the amount of virus that one might be exposed to (which may or may not imply the same thing).
Mark Bahner
Apr 24 2020 at 9:22am
I had a detailed reply to AMT that I think was blocked because I included several hyperlinks. The hyperlinks including important information about the availability of personal protective equipment for COVID-19. Please post the comments as quickly as possible.
Thanks!
Michael
Apr 23 2020 at 6:32am
I assume that if this was attempted, there would be shortages and higher prices. (As you implicitly acknowledge by suggesting that no one purchase these items unless they really need them.
Are you prepared to ban (or harshly limit) travel into and out of NY, NJ, and any other hotspots that may appear elsewhere? And do it suddenly without warning, such that an exodus of infected and contagious people doesn’t flee the region, taking Covid-19 with them?
On this, we agree. But I don’t see a realistic way to do this short of having a robust testing and tracing regime that allows authorities to find and contain early outbreaks as they happen, thus limiting the population of people who are infected and contagious.
Alan Goldhammer
Apr 23 2020 at 10:49am
I will look forward to the broadcast next week!!! One key point about David’s comment on occupational wearing of gloves and masks. Yes, this does make sense for the workplace but most of this is already required in the meat processing industry and we are seeing huge closures of large plants. It’s far likelier that the viral spread in the community is causing this than work on the line (I will add the caveat that we don’t know what line workers do on their breaks; perhaps they take off masks and gloves which would be bad practice). It’s not just wearing of masks and gloves which add some protection but adding worker distancing in as well will be important. I was reading one paper today suggesting that shift personnel sizes be reduced so that distancing can take place and adding a second shift for workers would help in this regard.
I like Mark Bahner’s suggestions and I put my money up to support small businesses. I regularly purchased bow ties from Lisa Eaton in Maine who always had nice fabrics. At the request of local EMS folks she pivoted and now is making very nice cotton masks (I already bought five!). Here is the LINK. I don’t have any financial interest in her business other than as a customer. The masks fit very nicely and I wore my lacross stick mask to Safeway this AM.
Mark Bahner
Apr 23 2020 at 5:00pm
I don’t know what goes on in the meat packing industry either. But here’s how I *picture* it from other industrial situations I’ve seen.
The shift ends, the workers come into a locker room. They remove the aprons and gloves and other work clothes, and then go in underwear to long wooden benches in front of their lockers. Probably, they then go to common-room showers, perhaps with short partitions between the showers. They use common soap bars. They come out, grab a towel and dry off, and then go back to the long benches in front of their lockers. They dress in their street clothes, sitting close to other people who are also getting dressed. They talk together as they’re dressing. Then maybe they go out and have a beer or two (or three) before heading home. Some percentage of them, particularly the younger ones, may get a ride home in the truck of someone they live near.
On weekends, a fair percentage of them may socialize with other people at the plant (since the plant would typically be the largest employer in a relatively small town or city). The kids of neighbors who worked at the plant would play together. The kids probably wouldn’t even recognize they were sick.
Comments are closed.