Freakonomics Radio - 409. The Side Effects of Social Distancing

Episode Date: March 19, 2020

In just a few weeks, the novel coronavirus has undone a century’s worth of our economic and social habits. What consequences will this have on our future — and is there a silver lining in this very black pandemic cloud?

After a relatively slow and scattered response to the global pandemic known as COVID-19, the U.S. has in the past week assumed what is essentially a wartime footing. The primary focus is on curtailing the spread of the virus and creating capacity to treat those who contract it. To accomplish this, we've been encouraged, all 330 million of us, to keep to ourselves as much as possible a practice known as social distancing. Schools and universities have been shut down, along with cultural and religious institutions, restaurants, and much more. The same for sporting events, theaters, conventions, and any other large public gatherings. Many office buildings have emptied out, with employees ordered to work

remotely. Travel, especially on planes and trains, is being severely diminished. All this has resulted in the biggest disruption of daily life that many of us have ever known, and it will last for weeks, perhaps months. Will it successfully contain the spread of COVID-19? We'll find out. Hopefully, it will at least be minimized. If we listen to the public health people, the virologists and the epidemiologists, and we should because they've been dreading and studying this kind of pandemic for years, they say the situation will get substantially worse in the U.S. before it gets better.

And what other effects and after effects will this social distancing produce? There will be many consequences, and certainly some unintended ones. Would anyone be surprised, for instance, to see a baby boom starting around nine months from now? Entire industries and segments of our society are being upended. The economic impact will be massive. Obviously, some people stand to be hurt badly.

Others are more protected. And some may well benefit, including those who can entertain and deliver and sell to the millions of people who suddenly have few places to go, not much to do. But the overall economic impact will be hugely negative and will likely require a massive infusion of government aid. Everything from industry bailouts to rent and tax relief, to emergency aid for laid-off workers. The stock markets fell 30 percent, with one harrowing single-day drop of 10 percent and another of 12 percent. Volatility is higher than it's been since the financial crisis of 2008. Now, to be fair, before the COVID-19 pandemic began, apparently in Wuhan, China, the U.S. markets were at an all-time high. And part of the drop should be attributed to another huge market disruption that sort of slipped in under the radar, an oil price war between Saudi Arabia and Russia.

The fact is, there are a lot of things flying under the radar right now. A lot of consequences that may come out of the global response to COVID-19. So, we thought we'd call a few economists whose past research indicates they might have some insights into the future that's being crafted right now by the extraordinary changes we're all living through. They are... Nicholas Bloom. I'm a professor of economics at Stanford University. Bloom studies economic uncertainty and the management of firms. Also... I'm Toby Moskowitz. I'm a professor of finance and economics at Yale University,

and I study financial markets and sports. And finally, My name is Marshall Burke. I am an economist at Stanford University. I'm an environmental economist, so I study how changes in the environment shape a range of human outcomes, health outcomes, economic outcomes,

livelihoods more broadly. We'll talk about the sudden spike in working from home and online learning, about the super volatile stock markets, and, believe it or not, one silver lining in the black pandemic cloud. All that's coming up right after this. From Stitcher and Dubner Productions, this is Freakonomics Radio, the podcast that explores the hidden side of everything. Here's your host, Stephen Dubner. Let's start with an overview of the economy itself in the age of covid19 here's nick bloom

so there are two things that i think are happening now first is there's clearly you know a tremendous negative shock on both demand and supply you know businesses are shutting down transportation tourism is falling so that's what's called a first moment effect. We know for sure that's bad news. But there's a second factor thrown on top of that, which is this incredible uncertainty, in particular, left tail risk. What does left tail risk mean, please? Left tail risk is very bad outcomes. So risk can, in theory, be on both the good side and the bad side.

So there's upside risk and downside risk. The COVID-19 really only has obviously downside risk. It's hard to see anything good coming out of this. And this additional uncertainty historically has turned out to be really costly for the economy because businesses pause hiring or investing. So I'd be pretty confident saying I suspect we're now already in a recession. How bad it will be is hard to tell. You know, some of the latest work in academia is about network effects. And that's Toby Moskowitz. So, you know, take the NBA who just postponed the season.

You know, the vendors, the suppliers, all the periphery industries, including people that, you know, work at the stadiums, taxis, hotels, they're all going to be affected by this. So we'll see that trickling effect happen as well. I wonder if you could talk for a minute about the difference between being right now a salaried, full-time employee versus being an hourly or non-salaried worker and how, you know,

it strikes me that that may be a huge split, bifurcation, that one class is going to do much better than the other. Am I wrong on that or right on that? You're right. So on the one hand, there's folks like, you know, me and you that are on a salary and then kind of relax in some senses. And we have all kinds of, you know, issues with our kids and health risks, but at least we're not worried about losing our income. And then on the other hand, there are people that are hourly pay that I think life is, you know, substantially harder yet still for because they've also got to go out and work, which puts them at a higher exposure risk. And when the recession happens, they especially the most vulnerable workers, have the rug pulled out from under them, a case where government and or private firms responded well to this problem?

I mean, unfortunately, when you look at recessions, generally, the lower skilled, lower paid do much worse. When have there been a good response? You know, I'm scratching my head to think about it. I mean, there was things like Ford on the $5 a day, which was famous in the 30s, stepped in to guarantee workers, you know, an honest living wage. In recent times, I mean, you know, the last 20, 30 years, actually, labor markets have generally been moving towards being more flexible. And so it's become easier for firms to, you know, lay people off,

particularly hourly workers. In fact, we're already starting to see some layoffs from this. They would be, you see it, bakeries, restaurants, things of that nature. Those are going to be the first to go. And that's something where policymakers and economists, I think, need to think about to sort of smooth out this disruption. What do you do? Well, I think one of the answers, and it seems like this is a, you know, I'm sure there's some debate, but my reading of the news has been that it's been fairly bipartisan, is longer paid leave from work. That, I think, has to help.

But that's, again, for full-time employees. I don't know what we do about, I mean, some sort of safety net for those people, I think, would be important. When there's such a high level of economic uncertainty, especially one produced by, you know, this massive disruption or shock, what does that do to monetary or fiscal policy? Does it rob it of some of its typical power? Yeah, so as we speak on March 12th, the S&P 500 fell almost 10%, which is actually the second biggest drop since World War II. And more surprising was halfway through the day, the Fed and the European Central Bank both stepped in to try and provide stimulus to the market. And, you know, it's like blowing against the wind. The market briefly went up and then just kept falling down. So, unfortunately, monetary policy

and to a broader extent fiscal policy, the government with tax and spending, I think has pretty limited effect on slowing this down. In this case, was it really the uncertainty or is it just the magnitude of, you know, the fear right now? You know, the primary reason why the Fed's move wasn't, you know, consequential is the damaging impact of COVID-19 is so large, there's not much the Fed can do. The other thing that's worth bearing in mind is, of course, before we went into this two weeks ago, interest rates were already very low. You know, we just fought a war against the Great Recession in 2008, 2009. We really hadn't reloaded our arsenal. And so, you know, the biggest meanie in sight appears on the horizon. So economists had worried about this for a while.

One of the reasons people wanted actually the Fed to put up interest rates a bit over the last two or three years so that we had some ammunition. The president, President Trump, has very vocally said that he wanted to keep rates as low as possible. Jay Powell, the chair of the Fed, reportedly had had inclinations in going the opposite direction over the past several years, but instead either kept them or kept lowering them. I mean, is this exactly the kind of instance where you wanted to keep that powder dry? Yeah. With benefit of hindsight, it was a mistake in particular to have massive tax cuts over the last two, three years because we're actually growing very fast. What would be much better is to push down the government debt so that right now, when we really need it, we could spend money.

Normally, you want to have big blowouts and recessions to support the economy and earn your savings back in the booms. And instead, we're in the hangover from spending in a boom and suddenly we're hit with a recession, there's very little money left in the bank. So the fiscal position, I think, is much more worrying because there we should have been generating a surplus and instead with a big deficit. This past Sunday, after we spoke,

the Federal Reserve cut its rates to nearly zero. It also announced it would buy at least $700 billion in government bonds, a move known as quantitative easing to try to keep markets from locking up. The stock markets were not mollified. They fell another 12 percent on Monday. Let's face it, if we're not transacting with each other for some long period of time, that will take a toll. I mean, think about the airlines. If the airlines aren't making any money for a while and can't run their routes, that's going to affect lots of businesses and all of that's going to go down for a while. Now, the thought is, of course, that as soon as everything jumps back up, the airlines can kickstart and start flying again. If that takes some time, it may take a little bit longer to

recover, but I'm hopeful that won't be the case. The last time we saw a stock market meltdown like this was after the financial crisis began to really gather steam. And what we saw was a lot of investors, institutional, but a lot of individual investors really panicked when the markets ended up falling. And many people sold low. And then as the recovery started, they ended up buying high. I think we all understand the emotional component of that, especially for people who are a little bit older and they just want to preserve the capital. As a finance person who's seen a few of these rises and falls now over the past few decades, do you have any general advice for people? Yes, I have very specific advice. Don't touch it. One of the basic findings from

economics is you can't outthink the market. Anytime people try to time the market, they end up doing far more damage than they help themselves. It's very difficult to do. As one example, I had many colleagues, these are famous economists, who said last week, I'm buying, I'm buying like crazy, this will be a blip. They're all sorry they did. And you're swimming with the sharks because the other side of that trade is guys in Wall Street that, you know, eat for lunch retail investors like us that don't really know what we're doing. So what I've always done, and I'm not the only financial economist that would tell you this, many of us would, is you have a long-term strategy, you stick to it. And you can't be blindsided or emotional about these short-term blips because you can't really do much about them.

So the best advice is actually not to look. Investing is, of course, a case-specific pursuit. If you are an older investor or if you've been putting money in a 529 plan for your kid's college tuition and they're already deep into high school, then a 30% drop in the market has different implications than if you've got a longer horizon. In any case, we should probably expect stock market volatility to continue for the near term, especially because COVID-19 has created so much volatility in how the biggest companies in the world are doing business. Apple has closed most of its retail stores around the world. Microsoft, Google, and Amazon have enacted mandatory work-from-home policies for most employees. What will the effects of that be? And will COVID-19 provide researchers an

opportunity to measure all these effects? Yes. Much as, you know, aircraft engineers investigate crash sites, economists will investigate, you know, what happened after COVID-19. Nick Bloom probably knows more about working from home than just about anyone you'll ever meet. Not just because he's a professor and not because he's lazy, but because he has studied this very question. Somewhat coincidentally, six years ago, we ran a study out in Shanghai in China, where a large online travel agent called Ctrip, which is really like China's version of Expedia, they decided to allow employees to work from home because they found office space in Shanghai was expensive. They asked 1,000 employees who wanted to work from home, and interestingly,

only 500 of them volunteered, despite the fact that employees on average were commuting 30 minutes each way. Of those 500 employees, they then randomized them by birth date. They randomized them so that the experiment would be truly an experiment and not an exercise in self-selection. And then we tracked them for nine months. And what we found were three things. Firstly, employees working from home. So these were people, I should say, were booking telephone calls and making processing data on computers. So they were kind of individual working jobs. They were 13% more productive. I mean, 13% is a huge increase. And the reasons they told us was, you know, A, it's quieter at home, so they could concentrate more. But B, actually, they just tended to work their full shift rather than spending as much time at lunch

or arriving late or, you know, taking long toilet breaks. Secondly, their quit rates halved. Many of them much preferred working from home and didn't want to leave their job. And thirdly, once you controlled for performance, since they were performing better, they actually weren't getting promoted any faster. So there is some stinging the tail that being at home seemed to reduce your ability to get promoted. It sounds like good news that productivity and happiness and all these things can increase. On the other hand, it sounds like that job that you were looking at lends itself particularly well to working from home, yes? Yes. As you say, there's a couple of major caveats. So it's really not a team job. So that's why you can be at home four out of five days a week.

The second point was that after the end of the study, they then ask employees to re-decide whether they wanted to work from home or come back into the office. And half of the employees said after spending nine months at home, they didn't like it. They felt isolated and lonely, and they volunteered to come back into the office. So for me, the warnings from the COVID experiment is, A, the type of working from home we're talking about now is very extreme. It's full-time, five days a week. You should note that less than 5% of Americans currently do that. Lots of people work from home a day a week, but very few people work from home full-time. It's kind of like comparing going to the gym sporadically with marathon training. So it's pretty extreme. And B, as you say,

we tested employees that don't need to spend time together, and most people do. And C, the COVID threat could well go on for months and months. I really worry about a big tick up in people getting depressed, mental health issues, which leads to health issues more generally because of the isolation it could lead to. My prediction is we will find that people that do routine jobs may perform okay at home, but for the majority of us, I think it's going to be pretty painful personally with all the loneliness, and I suspect will be pretty damaging for productivity,

particularly as time goes on. So I think if it's one or two weeks, it wouldn't be so bad, but if it stretches on to three to six months, I think it's going to be hugely damaging economically. As you gather that data, and I suspect a lot of companies are doing this, you can get a sense of, well, what is the productivity loss, if any, from having people work at home? Toby Moskowitz again. And this might be useful for when, you know, when you're hiring

employees, some of them want to, let's say, come to the office three days a week and work at home two days a week. Sometimes there's pushback on that, but it's not really backed by data. This will give us a chance to make that assessment. And this might be a nice way to force everybody to actually run this experiment and see what happens. I think another thing that's going to be damaged in the long run, actually, is if everyone's working from home, there's not going to be that kind of, you know, workplace discussions, coffee table discussions, lunchtime talk. And most of that, it turns out, is important for long-run innovation. So day-to-day, we can get along with, you know, if you're dealing with the same current customers or same ideas.

But when you examine businesses or scientists, or even the way I do my own research, a lot of that creativity comes from idle time and relaxed discussion with colleagues. And that's all gone. So I also worry that, you know, five, 10 years out from now, we will see this as another luring and long run growth rate because we've taken a big hit to innovation.

You know a lot about management and leadership and firms that are successful and unsuccessful. What do you think might be learned along those lines from the COVID-19 situation? So if you have a great manager that's very organized, that can deal with change, that inspires their employees, they can survive this. But I think a chaotic and disorganized firm could literally fall apart. And in fact,

already we've started to see bankruptcies start to head up. Is there anything you can point to that really good leaders do or don't do in a crisis? Well, one narrow piece of advice around working from home would be to try and regularly check in with your employees. So it's an unusual thing to do, but we're in unusual times. You could easily set up every day, you know, beginning of the day, end of the day, 10, 15 minutes, face-to-face, one-on-one Skype call, and just chew the fact. It's going to be really important for employees to feel like somebody's there that cares about them as much as they're there. It may mean that managers are going to lose, frankly, three, four hours a day on these one-on-one meetings. But I fear without that, they could A, lose contact, and B, employees could become quite seriously depressed and lonely. Well, it's the same argument that people had with

the school closing. So it's very costly to close a school. People work. Well, it's the same argument that people had with the school closing. So it's very costly to close a school. People work. Now, what do you do with your kids? The kids aren't learning. I get all that. Those are costs. I think people are making the wrong comparison, though.

You can't compare the cost of doing this versus not doing it. Not doing it's not an option. The question is, you do it now for a shorter period of time, or you do it later for a much longer period of time. All the things you're mentioning are definite costs. I just don't think we can avoid them at this point. Many people, including us on this show, have kind of complained about or agitated

against the standard setup with meetings in corporate America. We have a lot of meetings, and many people feel that those meetings are beyond not productive, but actually onerous and intrusive. Do you have any thoughts on whether this might change meeting culture in any way? Well, nobody hates meetings more than professors, that's for sure. But everybody feels this way. I'll give you an example. I had at the university today a couple of meetings scheduled that if I wasn't coming in to do this, I would have canceled completely. Instead, I said, well, let's not all get together in a room. That

seems stupid at this time. Why don't we try this on Zoom, which is just a, you know, teleconference. It worked just fine. I mean, initially there was a little awkwardness where people introduced themselves and a few of the, you know, I would call them the old schoolers said, well, this is why we don't like these things. See how awkward it is. Once we got into it, the meeting was far more efficient. And what was a lot less. Once we return or if and when we return to normalcy, do you think it's going to be hard to get people to get back on that commuting wagon? Nobody likes commuting. It's actually the number one thing on surveys that people say they hate the most. And I think the longer this goes, you'll have more requests for, hey, look, I could do this commute three days a week, but not five.

As we know, for the last 20 years, the prices of property in the center of cities have surged. Young people in particular want to live in downtown areas. If suddenly we switch to working from home, you could easily see that reversing. So I could easily see there being kind of satellite towns, but they would all be relatively cheaper and we could spread out more uniformly. So yeah, I think for the housing market, working from home is a big mass rollout to cut property price in the center of cities. I'm curious, you know, once people are ordered to work from home and then they all of a sudden they're not having to commute, they're not having to tuck in a shirt, et cetera, maybe they don't even change their underwear. I don't know. Do you think getting some of those people back to work, especially back to the commute every day will be difficult? I've talked to a lot of firms in the U.S. and internationally about

working from home. They're kind of reluctant to do it. They're nervous that they let the genie out the bottle and, you know, it's hard to reverse it. I actually think this experience is going to force a lot more serious thinking about working from home and see a big spike up, some of which will definitely be beneficial in the long run. These are just a few and perhaps the most predictable of the countless economic and social side effects we'll be seeing from the COVID-19 response. What are you seeing? Let us know at radio at Freakonomics.com. Coming up after the break, how will education work with schools and universities suddenly shifting to online teaching? And if you are in desperate need of a silver lining, well, there is one.

Clearly, the epidemic was causing an immense amount of harm on the ground, but it was also reducing air pollution. That's coming up right after this. Social distancing to diminish the spread of COVID-19 means that we're all supposed to limit our interaction with other people as much as possible. Some of us, maybe most of us, will find this incredibly hard. The late economist Gary Becker made a career of studying behaviors that most economists didn't think much about, including addiction. And what did Becker argue is the most addictive thing in the world? Other people. We are, for the most part, social animals. What kind of animals will we be without the socializing? We're about to find out.

One of the biggest changes right now, one of the biggest unplanned experiments of the COVID-19 era, has to do with remote teaching and learning. Schools and universities across the U.S. and elsewhere have been shut down. Yeah, Stanford closed. Yeah, that's what they've done here at Yale, too. Those, again, are the economists Nicholas Bloom and Toby Moskowitz. We're talking about COVID-19 as a natural experiment that will allow people like you to measure the efficacy of remote teaching. So can you envision actually doing that study or someone doing that study?

What kind of data would be needed? What kind of timeframe would be useful, et cetera? You may compare courses to, say, the courses taught in person last year. So you may say, look, we've got 10 years of basically the same course. And then suddenly this year we teach it online. What do the grades look like? What do the long-run outcomes look like? So you may say, look, we've got 10 years of basically the same course, and then suddenly this year we teach it online. What do the grades look like? What do the long-run outcomes look like?

And Stanford, for example, Econ 1, which is one of our key courses, was taught in fall, spring, and winter quarters. So you could look at the Econ 1 students that took it face-to-face last quarter versus the ones next quarter that are going to take it entirely online and see how they do in years two, three, and four. And what's your suspected conclusion? Do you think that there are some kind of antiquated inefficiencies in the system that could be, you know, eliminated? And even though it'd be a shame that it took something like COVID-19 to produce this kind

of change, that there might be a lot of benefits to some forms of remote learning? You know, I'm not that optimistic that remote learning is going to be that successful. The reason is from personal experience. I feel a lot of my value added is what I'll call the personal trainer effect. You know, giving students motivation. You're forcing people to focus for an hour and a half. For example, probably the biggest single improvement

in my teaching was the year that I banned laptops and cell phones from being used in the class. And, you know, it was miraculous. Suddenly everyone pays attention. Whereas when it's offline, it's so easy to get distracted by, you know, watching the football or seeing the news or watching the stock market. So let me pretend to be cynical for a moment and say, well, the kind of person who would measure this effect long term, In other words, remote teaching actually brings some gains and that might eliminate a lot of live professorial activity. The kind of people who would come up with that are professors.

So is this in the best interest or in the interest at all of anyone to actually study and come to that kind of conclusion? Yes, absolutely. I think, you know, as economists, you always want progress. So imagine professors can suddenly teach much more efficiently by recording a class once than having it played on video. Then, of course, we can spend our time more one-on-one with students. My experience, the highest value added is sitting down with individual students and going through their research papers. And if we could spend our hours every day on that, I think it would be much more valuable for students actually

than just reciting material they could learn out of a textbook. So many universities have already spent a lot of time and a fair amount of money putting their courses online. And then there are for-profit firms, Coursera and others, who, you know, have pretty robust platforms. And I see that the take up numbers are not insubstantial, but considering how difficult and expensive and timely it is to attend college, to go to four years worth of college, I've been shocked at how low the take up rates are for online learning. And I'm curious whether you think that this might

adjust that. Here's my view on this, and just as from 25 years of teaching myself, I find if you're just teaching facts and methods, that can be done almost as well online as it can live. You could even argue maybe even better online because you can supplement it with video materials and you can record it, right, and get it perfect. If you're trying to teach someone how to think and you're trying to teach them, let's say, how to do research or how to ask an interesting question and get a scientific answer, that's much more hands-on. So, you know, if I'm teaching basic investments, what's a stock, what's a bond, I think I can do that just as well online as I probably can in the classroom. But if you're talking about training students how to think, right, and how to really, you know, whether it's writing an English essay or writing a poem or, you know,

coming up with a computer program that, you know, does X or, you know, thinking about how to price some obscure private equity firm, that I think requires a lot more back and forth and a lot more interaction. And I suspect that the reason that we still have the university model is that's what they're trying to do. What I tell my students is, I'm trying to teach you how to think because what I teach you today may not be relevant 5, 10, 20 years from now, but if you know how to think about it, you'll be able to figure it out. You know, when LeBron James, it's a very terrible analogy, but, you know, there's a reason that

basketball players want to play in front of fans. I have to just say, whenever an economist compares himself to LeBron James, I'm all ears, okay? I wasn't going there, I swear. But making a very poor analogy, well, the analogy's okay, but it's on a very, very different scale, which is, you know, I think when you're teaching, you're reacting to your students. You see their faces, you see their questions, you see confusion. Sometimes you see a light go on. That's very difficult to do. Even though you see their faces on the flat screen, it's not quite the same. And for an athlete, it must be a hundred times like that, right? And I sort of get where that sense comes from. So now it's in my interest to say that that's the case.

If it turns out we could teach just as effectively remotely, well, that's a whole different paradigm for a guy like me. Yeah, for academics, it's been, in some sense, a real opportunity to explore other ways of doing things. That's Marshall Burke, the environmental economist at Stanford. So all the conferences I was going to go to this spring have been canceled. I've sent my lab group home. So I hold my lab meetings online.

We've had speaker candidates online. And I was skeptical that it would work well initially. And actually, I think it's working quite nicely. And I really hope that we learn that lesson. And then we do some of these substitutions. There's no reason I need to fly to the East Coast, you know, many times a year to give talks. Why can't we just do it remotely? I think this will give us the opportunity to explore the benefits of these other modes of economic production. So I think this is going to generate huge amounts of experimentation and teaching. And in the long run, some of it would be good, but it's very painful right now. You know, when we think about schooling,

especially in the early ages, socialization seems to be a big benefit, or at least that's what people say. So right now we're seeing all levels of schools being closed around the U.S. and in other places from pre-K up through graduate school. So I'm curious if you have any thoughts about what kind of unintended or knock-on effects that may have. You know, I think socialization is actually really important. It's also interesting that the American system has been criticized heavily

for doing not that well in international comparable math tests and science tests and reading tests. But Americans are very good at socializing. And I don't think it's a coincidence. There's lots of American startups and successful CEOs and reading tests. But Americans are very good at socializing. And I don't think it's a coincidence. There's lots of American startups and successful CEOs and entrepreneurs. It's part of this socialization process.

From an early age, American kids are taught to argue, to speak up. And I think that will be lost if we move entirely to online home-based teaching. Marshall Burke has already moved his own young children to home-based teaching. We pulled our kids out of school this morning, and already two hours into our homestay, they were punching each other in the face. So I am not optimistic about the livelihood improvements from these stay-at-home weeks.

We should say they are twins, right? They are twins, so they like punching each other in the face. Are they both boys by any chance? Two girls, actually. It cuts across all genders, apparently. Here's how Burke describes the work he does as an environmental economist. So I study how changes in the environment shape a range of human outcomes, health outcomes, economic outcomes, livelihoods more broadly.

Can you talk for a second about your past research on the association between climate change and violence? So one thing we have done is assemble historical data from around the world on violent outcomes. And what we find is a very strong linkage between changes in temperature and increases in various types of violence. So temperatures go up, you actually see more homicides in the U.S., you see more civil wars in sub-Saharan Africa. So this is not a new finding. It comes through very clearly in many different data sets. And can you explain just on a kind of behavioral level the mechanism by which that

happens? So there are multiple mechanisms that are consistent with this effect. I think we're still trying to understand exactly which one or which combination of them is most important. There's a physiological mechanism. So psychologists have shown for a long time that if you just put people in a room and turn up the temperature, you can piss them off, even if there's no one else in the room. But patterns of social engagement also change. So a warm evening, many more people are out on the street, and maybe there are enough different social interactions that lead to an increase in crimes. And these larger scale changes, declines in economic productivity or other outcomes is the standard explanation for why you would see things like civil war, civil conflict increase when temperatures go up.

So given all that, I am curious if you have any predictions about a similar association between violence and COVID-19? Because again, you've got the regular living patterns of many, many, many people. It's going to be billions of people around the world being disrupted, everything from going to work versus not going to work, versus how they eat, versus how they interact, et cetera. And I am curious whether that potential makes you anxious, interested to measure, et cetera, et cetera? Absolutely interesting to measure. I would think most of the measures being put in place to reduce the spread of COVID-19 are reducing social interaction. That's explicitly their goal. And so,

you know, my naive prediction would be that that would, you know, reduce at least many types of individual crime. Now, if things got bad enough where resources were constrained, could things my naive prediction would be that that would, you know, reduce at least many types of individual crime. Now, if things got bad enough where resources were constrained, could things get a lot worse? I think so. Anecdotally, a lot of people are buying weapons here in the U.S. So to me, that suggests that some people think that broader scale violence could break out. I very strongly hope that's not the case, but that's clearly a concern of some people. Okay. So the big reason we wanted to speak with you today was about this recent work you've done specifically related to COVID-19 and an unintended consequence of the lockdown in China.

Sure. So a few months into the COVID-19 epidemic in China, where it started, NASA published some really startling photos of changes in air quality over China. The Chinese had, you know, taken very aggressive action to limit the spread of the virus, and that had economic consequences. And you could actually see these from space. So many factories have been shuttered. People are not going to work. So what NASA showed was a dramatic reduction in air pollution across a lot of China. And as an environmental economist, we're very interested in air pollution, why it changes, what effect it has on economic things we care about in the world. So I and many other environmental economists immediately thought, you know, this could be in some sense a silver lining in the epidemic.

Clearly, the epidemic was causing an immense amount of harm on the ground, but it was also reducing air pollution. And we have decades of research suggesting that air pollution is really bad for health outcomes. And so any reductions in air pollution are going to be good on average for health outcomes. And so any reductions in air pollution are going to be good on average for health outcomes. There was another recent example of a sudden drop in air pollution, also in China, in 2008. Chinese officials wanted the air around Beijing cleaned up ahead of the 2008 Olympics. At that point, Beijing had some of the dirtiest air in the world. And so what we saw in late 2007 and early in 2008 was one of the most dramatic efforts to rapidly clean up air quality that we've probably seen anywhere throughout human history. They raised the price of gasoline to encourage people to not drive. They prohibited certain polluting vehicles from being on the roads at all. They shut down a range of manufacturing plants, cement manufacturer, concrete manufacturer.

They forced some of the large steel plants to either turn off or to actually relocate. And all of this had a really dramatic effect on air quality. Air quality improved by about a third in the span of just a couple months. And this was all temporary suspension then, correct? This was temporary suspension. Olympics were done and all these constraints were lifted. And once the constraints were lifted, where did the air pollution quality go back to? So it went back to its previous very high levels, unfortunately. And was that relatively short amount of time with less air pollution enough to substantially improve people's long-term health outcomes?

So it turns out it was. So you can compare how mortality changed in Beijing relative to other cities that did not see this dramatic change in air pollution. So the researchers found very large reductions in child and infant mortality and very large reductions in old-age mortality. What can you tell us about the in utero effects of air pollution? This has been studied a lot in the context of the Clean Air Act in the U.S., which led to pretty substantial improvements in air quality across the U.S. And people, again, find really large effects. So exposure to air pollution in utero leads to later life outcomes, reductions in earnings, poor health overall. So air pollution in utero leads to later life outcomes, reductions in earnings, poor health

overall. So air pollution does not have to kill you to make you worse off. So when Marshall Burke saw those NASA maps showing a decrease in air pollution across China due to COVID-19, he went to the data. And what kind of drop did he find? So the percentage drop was actually quite close to what we saw in the Beijing Olympics, maybe slightly smaller. I would say a 20% improvement in air quality. And with the Olympics, you mentioned that the drop was very variable or very concentrated on Beijing. If you were to look at all of China a couple months into COVID-19 versus before, do you see a drop pretty much everywhere? Or is it mostly in the

places where COVID-19 was particularly concentrated? So the data are still rolling in on this, but the reductions do seem to be the largest in places where COVID-19 was most serious. The satellite data indicate that. So Wuhan, you know Wuhan saw an immediate and very large decline in economic activity, and that led to a decline in air pollution. Cities in southern China have seen the largest overall decreases in air pollution, and that's because in the north, places like Beijing, actually, home heating is an important source of energy use and also a source of pollution, and that hasn't changed that much. So actually, in Beijing, you've seen next to no decrease in air pollution due to COVID-19. That's both because the air pollution and because cement factories actually don't like to turn off their blast furnaces. It's really expensive to stop them and then start them

again. So they like to just leave them on if they can and sort of try to wait it out. And that's what they've done. And so you've seen sort of no decline in air pollution around Beijing, sort of ironically. All right. So tell us what you believe will be the relationship between this reduction in pollution and mortality in China. So I calculated that this reduction in pollution will likely save 50,000 lives across China. I want to emphasize that this is, it's a prediction in a sense, it's not a measurement. It's using these estimates of the relationship between air pollution and mortality that we know from other studies. So what those estimates tell us is that

the gains are concentrated, again, among the very young and among the very old. So those 50,000, let's call them saved lives, kind of saved in quotes, because it's somewhat speculative or predictive, that in your estimate, as we speak with a moving target granted, would compare to how many lives lives lost to COVID-19 in China. That number continues to go up, but go up much more slowly. China seems to have really turned the corner on this. So it's about, you know, 20 times, 15 to 20 times higher, the deaths saved from air pollution. I want to emphasize that the 3,000 deaths are just the direct attributed deaths to COVID-19. That does not count all the, you know, the immense disruption that's happened that could have led to additional deaths, other people not being able to get treated for non-COVID-related diseases, those we have yet to really observe in the mortality statistics. And so a full accounting

will need to, you know, take those into account. And we just don't have the data to do that yet. If we were to focus for just another minute on the silver lining, I'm curious about other behavioral responses to COVID-19 in China. So air pollution fell enough to have a big positive effect. Transportation is a substantial contributor to air pollution. So did transportation fall enough to lead to maybe a big drop in traffic fatalities as well? I think that's an interesting question. And again, one that would have to be studied once we have more mortality data in, in settings in which car accidents are a large share of mortality or for certain demographics, like young, otherwise healthy demographic for which car accidents are the leading cause of death

in many parts of the world. I think you would expect a reduction in traffic deaths. I can imagine people listening to this and say, this Professor Burke sounds like a nice person and a thoughtful person and a bright person, but he's such an economist. There's a global pandemic going on and he's finding this silver lining in this gigantic black cloud that you're identifying what might be, let's say, 45,000 fewer deaths in China because of less air pollution versus the deaths from COVID-19. So I'm just curious, does it feel, I'm sorry, I don't mean to ask you to, you know, hoist yourself on your own petard, but does it feel somewhat awkward or do you feel guilty or conflicted in any way in kind of examining the specifics of that silver lining while the world is in such

shock and uproar? It feels terribly awkward. And it was a calculation I could do because I had the data, but I would hesitate to even call it a silver lining. So to step back, I did this calculation, I tweeted it out. And a lot of the response was like, well, you know, this is terrible, like, and you're not accounting for all the other negative consequences that a large pandemic like this would have on the health system. And I think that's right. And those have been the stories coming out of Italy and coming out of China, right? If you have any other health complication unrelated to COVID-19, it's going to be very hard to get treatment. The silver lining is irrelevant right now.

Let's focus on the dark cloud and getting the dark cloud under control. Epidemics are terrible. They result in an immense amount of human suffering. That is going to continue to happen in our country. And that is where our focus needs to be. And none of my silly calculations should take away from that. To me, what this calculation does highlight, though, and unfortunately, what things like COVID-19 or other dramatic events bring into focus is what's going on when we don't have pandemics. And if anything, the pandemic can help highlight the negative effects of our

everyday actions. So I guess ideally, what you'd want to think is after the recovery from this pandemic, what you'd have is policymakers look at your data and your conclusions and say, whoa, let's not wait for the next pandemic to curtail pollution. Look how many lives it saves, right? That's right. And I think the Chinese government, you've actually seen pollution levels come down pretty dramatically in the last few years. China's already making progress on this problem, and I think should be commended for that. Economists really like to think in cost-benefit terms. So let's weigh the positives and let's weigh the negatives and let's

compare them. To me, that's not obviously the right way to go in an epidemic. Let's just think of this. Let's focus on the negatives. Let's not do a cost-benefit calculation of epidemics. I think that's just the wrong framing overall, despite that being what we always want to do. But let's use the epidemic to learn about things we could do better when we don't have epidemics. I think that's the way to think about it. Let's pretend for a minute that, you know, God willing, the pandemic turns out to be less fatal, less damaging than it certainly could be, and that in a couple months' time, things seem to be getting back to what would seem to be pre-COVID-19 normal. Do you really think that people will, for instance, decide not to congregate as much? Do you think that online learning or remote working or any of these other substitutes will actually stick around? That's a great question. And yeah,

honestly, I have no idea. I mean, I'm old enough now to remember when computing started to become pretty widely available. And a lot of smart people said, well, that is the end of cities. It's the end of in-person work, period. Everybody's going to live in, you know, Fiji or wherever they want. And it just didn't happen. In fact, the opposite happened. This propinquity turned out to be incredibly valuable and urbanization has risen and risen and risen. So I'm curious about what curiosities you may have to what kind of lasting effects you may want to look for as an economist? Yeah, so economic historians, certain economic historians point to these critical junctures in history where it wasn't slow change, but something specific happens that set entire countries or entire economies on a different course. So obviously, it remains to be seen whether this specific epidemic, and hopefully it subsides and we get it under control, once it goes away, what happens? But one idea is that these sorts of large-scale events could be critical junctures that cause us to change our behavior in ways that just sort of slow-moving changes in technology or slow-moving changes in preferences never do.

Coming up next time on Freakonomics Radio, we're not really sure what's coming up next time. As with you and everyone else, our lives have been scrambled a bit. Here's hoping you and your loved ones, even your enemies, stay healthy. And if not, that they can get the care they need. Let's all keep an eye on one another, shall we? And we will be back next week with a new show. Thanks for listening. Freakonomics Radio is produced by Stitcher and Dubner Productions. This episode was produced by Daphne Chen. Our staff also includes Allison Craiglow, Greg Rippin,

Zach Lipinski, Matt Hickey, Harry Huggins, and Corinne Wallace. Our intern is Isabel O'Brien. We had help this week from James Foster. Our theme song is Mr. Fortune by the Hitchhikers. All the other music was composed by Luis Guerra. You can get Freakonomics Radio on any podcast app. We can also be found on Twitter, Facebook, and LinkedIn, or via email at radio at Freakonomics.com. Stitcher.