Paul Romer on a Culture of Science and Working Hard

From charter cities to mass testing for COVID-19, Paul Romer doesn’t always think his ideas are good—they’re just better than the alternatives.

Paul Romer makes his second appearance to discuss the failings of economics, how his mass testing plan for COVID-19 would work, what aspect of epidemiology concern him, how the FDA is slowing a better response, his ideas for reopening schools and Major League Baseball, where he agrees with Weyl’s test plan, why charter cities need a new name, what went wrong with Honduras, the development trajectory for sub-Saharan Africa, how he’d reform the World Bank, the underrated benefits of a culture of science, his heartening takeaway about human nature from his experience at Burning Man, and more.

Listen to the full conversation

You can also watch a video of the conversation here.

Read the full conversation

TYLER COWEN: Hello, everyone. Today I am chatting once again with Paul Romer, who needs no introduction. Paul, welcome.

PAUL ROMER: Good to be here.

COWEN: You have a recent article in the periodical Foreign Affairs about the failings of economics. Let me try to defend the economics profession. Tell me what you think. If I look at the big catch-up winners over the last few decades, it seems to me it’s Poland and Ireland, and they basically followed a neoliberal recipe. They more or less did what economists told them to do. What’s the failure in that?

ROMER: By the way, what about China? China caught up pretty well too, but they followed some of the basic insights from economics.

COWEN: And the Solow model.

ROMER: Yeah. But the origins of that article were that I read some books that said economists got a lot more influence and things got worse in the United States, and this was a really troubling argument for me because it’s not easy to dismiss. What I concluded in that article saying we should do a cost-benefit analysis — look at the big things that economics has done well, the things it may have done badly, and just see how it works out.

The point you’re alluding to is something that my colleague, Peter Henry, has also made, which is that one of the areas where economics may really have been helpful is in the development process or the catch-up phase of growth. So that should go on the plus side, I think, on the benefit side of the cost-benefit analysis, no question there. And I think there’s some other ones that belong there too.

My point was that there may have been some things that have also been significant negatives, and it’s time to do the numbers and see what the net is.

COWEN: So if I ask myself, “What do I think has been the biggest negative?” I suppose I would say around 2000 date, economists for the most part did not understand the importance of the shadow banking system. What seemed to be a kind of ordinary real estate bubble, like the early 1990s, was far, far worse, and we totally missed that. That seems to be a defect of institutional knowledge, but you tell me what you think the greatest problem has been.

ROMER: I think this problem is an interesting one. I put a slightly different spin on it, but I think it’s in the class of things of a failure to understand or incomplete understanding. I don’t think that’s a sign of a science that’s failed. That’s a sign of a science which is just making progress. There’s some things it knows and things it doesn’t know. So I don’t view this one as a sign of a systemic problem that we’re not doing it right, in a sense.

For what it’s worth, we can come back and talk about this, but I think the lesson from the financial crisis, which we’re learning again now, is one about the fragility of extensive interconnection. We’ve paid attention to optimize efficiency with massive reliance on specialization and these complicated supply chains.

But the growth, the proliferation of connection means that our system is more fragile than we realize. A shock comes, and things happen that we didn’t anticipate. But again, that’s part of learning about a very new type of economy which is changing in real time.

The ones that struck me as being particularly worrisome were, first, I think the negative effect that economists have had in terms of protecting competition. Through the law and economics movement, we ratified this notion that big is okay as long as you can make some case that it’s efficient.

The upshot is, is that I think because of technical economics and the arguments of economists, antitrust is much more tolerant now of dominant firms, and if we believe that competition’s good in a whole bunch of ways, this could actually be very, very harmful. So that’s one.

COWEN: Doesn’t Amazon look pretty good right now in the midst of the pandemic? Do you wish we had split it up into different parts?

ROMER: My sense is that we’d be better off if we had five Amazons instead of one. And I don’t see why we couldn’t have five Amazons if we, as voters, say, “This is the kind of society we want to live in. Let’s just aim for that.” And same thing — the more worrisome positions are those of the tech firms that are so deeply connected now to many aspects of our lives and where there’s really very little competition and a lot of opacity about what these firms actually do.

COWEN: Let me try to defend the economics profession a bit more. If we look at climate policy, a lot of economists have recommended a carbon tax — not quite a consensus, but a very common view. Now, of course we haven’t done it, but it seems to me the profession, in some manner, is essentially correct there. So you would side with the profession on that?

ROMER: Yeah. Again, in some sense, the main point of the article that I’m making is that economists need to accept that our role is that of the technical adviser. We can say, “If you apply a carbon tax, carbon emissions will go down. Here’s what other effects we think they’ll have. But it’s up to you, the voters, to decide whether you want to follow that policy or not.”

So, if the voters don’t follow us, I think, to a first approximation, that’s not really our responsibility. And what I’m critical of is this tendency for economists to assume the responsibility of philosopher king and say to voters, “Well, we know better what a society should be like, what society should do. Listen to us. We’ll tell you the way things should be. We’ll tell you what you should do.”

And, in truth, I think we get into that mode a lot more than we realize. Certainly, some members of the profession get into that mode. And I think they’ve done, really, quite a bit of harm when they did that.

COWEN: When you, as a voter, judge policies, what normative or philosophical standard do you use?

ROMER: Well, I think all of us have some notions about self-interest, and then the well-being of those around us. Everything else equal, if our position is the same, we’re somewhat happier if those around us are happier as well. Different of us have either a bigger or smaller circle of those we care about, so there’s some mixture of making sure everybody is doing okay, and then making sure I’m doing okay. That’s the first thing that I look at as a voter.

But I also look at — this is a little bit of a tangent relative to your question — but I also look at the question of, in what direction will this policy take our norms, our beliefs about right and wrong? I think those change. I think there’s some beliefs about right and wrong which are better in an objective sense, in the sense that we economists think of in terms of efficiency. If everybody thought this was the right thing to do, then we would actually all be better off in some objective sense.

So those issues weigh heavily in my thinking about policies. Economists have been a little slow to take those up, but frankly, I guess that is a part of the problem with economic analysis. Because many of the arguments about, say, allowing the market to run and giving people more freedom make more sense if, when you do that, you don’t change norms. But if, when you do that, you encourage norms that are destructive, that kind of laissez-faire approach can be harmful.

Let me take a trivial example. Suppose the promulgation of laissez-faire makes everybody feel like it’s okay to litter. We used to have a norm that we shouldn’t litter because it was inconsiderate and it was just wrong. Laissez-faire convinces us I can litter if I want to. It’s somebody else’s problem to deal with the litter. That kind of laissez-faire would be bad because we’d live in a world that was full of trash all over the place. And I think in more important areas, economists have been inattentive to the effects that their policies have had on norms.

COWEN: But if you take, say, litter, why wouldn’t the economic approach be either create a private property right, which we do sometimes? Other times that’s not possible, so we want something like a Pigouvian tax or a cap-and-trade. And your view, then, is not really far from the standard economics view, if at all.

ROMER: No, but I think, actually, there’s enormous value in norms that are kind of self-enforcing. Suppose people think litter’s bad. Suppose they think it’s bad when other people litter. So they’ll kind of scold or criticize when they see somebody litter. Then, without police powers, without courts, without taxes, you actually get the outcome we want, which is we live in a world with no litter. And if we lost those norms, we’ve got to overlay these more heavy, expensive governmental solutions.

COWEN: Let’s look at macroeconomics. If I look at the current crisis, which is turning into a depression, it seems to me we were on the verge of a financial implosion in March. The Fed acted to limit that. The macroeconomic response, to me, from the Fed seems to be quite good. So, isn’t all well in macroeconomics?

ROMER: Yeah. When I was talking about this cost-benefit exercise, one of the positives we mentioned was in the sphere of development and catch-up. I think another is in stabilization policy. The kind of practical macro policy as practiced at the Fed is much better now than it was during the 1930s, and we get real benefits from that. I think it’s good that the Fed is trying to make sure that we don’t have this cascade of bankruptcies. There’s no question that economists have learned something and contributed to society.

Just as a side note, I’ve been critical of the more theoretical, rational-expectations macro, but set that aside because that really hasn’t had that much impact on policy. So macro policy as practiced at the Fed or as practiced by the Congress right now is, I think, a reflection of things we’ve learned relative to, say, the 1930s, and that’s good.

But let me come back to what are the minus sides of this balance sheet. I talked about antitrust and the failure of competition policy. The other one is in regulation. If you ask me who’s my representative of somebody, an economist who overstepped, overreached, and did real harm, it’s Alan Greenspan. Greenspan was this tireless advocate for cutting regulation. He was quoted at one point, saying he’s never met a regulation that he thought was valuable. And he played a very important role in deregulating the financial system for decades running up to the financial crisis.

That financial crisis cost us an enormous amount worldwide, and it’s because we unwound systems of regulation that kept our financial system from being as fragile as it’s become. You look across the board at other types of regulation — we’ve failed to support the kinds of regulations that we need alongside of Pigouvian taxes. There’s some bad things that people do, bad in a sense they’re inefficient. We can try and tax them. Other times you just use regulation. But one way or the other, we collectively want to stop people from doing things that are harmful.

COWEN: But if you look at the profession as a whole, wouldn’t most economists agree that, say, tax preferences for owner-occupied housing are a bad idea? And various other subsidies built into the system for housing are a bad idea? If they had been listened to on that, we still might’ve had a crisis of some kind, but it would have been far smaller.

Scott Sumner has argued if we had targeted nominal GDP, the crisis would have been milder. You’re picking a bit on the one thing that Greenspan got wrong, but there’s many other things economists have said that would’ve made it much better.

ROMER: Yeah, but we still should have been saying, “Given the choices that voters are making, which reflect your preferences as voters, like supporting owner-occupied housing, the regulatory choices that we’re recommending as economists are actually exposing us to just massive, massive harm.”

I don’t remember the number off the top of my head. Haldane did some calculation where the worldwide cost of the financial crisis was, I think, in the hundreds of trillions of dollars. This is a really huge, huge mistake. And we’re still, I think, exposed to a financial system which could just blow up on us at any point in time. It’s part of why the Fed has to be so active right now with providing funds.

So deregulation, especially of financial markets, I think was harmful, and competition policy was a failure. And then the bottom line is, you just look at one of the most basic ways to measure progress: how long do people live? People in the United States are not living as long as they used to. Life expectancy is declining, and life expectancy hasn’t been keeping up with other nations around the world.

COWEN: Sure. Is that a failure of economists?

ROMER: Well, I think partly. When the pharmaceutical firms that were trying to make money off of Oxycontin and these opioid-based painkillers — when they went to Congress to try and stop the DEA from shutting them down, what they used was the language of economics.

“You have to have innovation. You’ve got to let the market proceed. There’ll be some creative destruction, but you have to let us do our thing. You can’t interfere — it will be bad for growth.” So, to the extent that we lent cover, indirectly, for those kinds of arguments against regulation so firms could make money killing people, we really did something bad.

COWEN: But again, there are people out there who have misused your ideas or misrepresented them.

ROMER: Oh, yeah.

COWEN: I think they’ve done the same with mine. I don’t blame you at all for that, right? If something bad happens with a charter city, I don’t say, “Oh, Paul Romer gave them cover.” I say, “No, it’s the fault of the people who did it.” So I would say economists were pretty much not to blame for the opioid crisis.

ROMER: There’s a speech Greenspan gave, where he doesn’t cite me, but he could. It’s all about “We can’t have regulation because we’ve got to have growth. Growth comes from innovation. Regulation slows innovation. It’ll stop creative destruction. We just have to live with greater destruction.”

So, I feel like, yeah, some of my ideas could have been used to support bad policy, but instead of asking whether I’m personally to blame or personally a bad person, what I’m stepping back and asking is, did we create a system that let someone like Greenspan make recommendations under the cover of science? Like, “I’m a scientist, I’m telling you how it should go.”

But those recommendations were really based on a worldview he got from a novel by Ayn Rand. There was no technical scientific basis for them. And they turned out to be really incredibly harmful, so we need to make sure that the system that we’re building isn’t misused in that way.

COWEN: Do you think there should be an obsession with math GRE scores when admitting people into graduate programs in economics? We know there is, right?

ROMER: Yeah, yeah. Well, it’s not the only — put it this way . . .

COWEN: Is that part of the problem, or is that how we need to do it?

ROMER: It’s not the only thing I think we should be looking at. And I’m not sure what are the other predictors, but I don’t think just practice in math is going to lead to a successful career in economics.

COWEN: You’ve been interacting a lot with epidemiologists due, in part, to your arguments for testing. What’s your opinion of that field?

ROMER: There’s actually an interesting parallel in epidemiology with a technical kind of issue in economics. In macro, we shifted towards model-based reasoning about macroeconomics. So representative agent — the whole rational-expectations movement was a shift towards “Let’s see what the models say,” rather than “Let’s see what the data say.”

In epidemiology, there’s a very well-established model — this SIR model that is behind a lot of these predictions, but there’s an alternative — the Institute for Health Metrics and Evaluation, which has this model that’s been very influential, widely watched these days. The IHME is using a much more data-driven approach, kind of a curve-fitting approach. It’s almost like old-style Keynesian macro, where you just say, “Well, let’s just fit something to the numbers and see what comes out of that without imposing a lot of theory onto the estimation process.”

And I just found it interesting to see that tension in another field, from the outside. The way it looks to me is, it’s good to have both of those wings active in a discipline. And it’s good to have them in contention with each other.

If I have a criticism of macro in economics, it’s like the criticism in epidemiology — we may have biased things a little bit too much towards the models, and we’re not giving enough weight to just the facts themselves. And I think that’s because it’s actually easier to do models than to look at data, so we need to have a little bit of collective pressure to, “Yeah, yeah, that’s what your theory says, but let’s look at the numbers.”

On Romer’s COVID-19 testing plan

COWEN: Many people have supported mass testing plans. Of course, you’ve been in the lead here. Why do you think they’re not getting more support? Because the benefit-cost ratio, if you can pull it off, seems to be quite high.

ROMER: Yeah. I’ve actually been working — oddly, mind you, I’m the theorist criticizing the use of models, so, go figure — but what have I been doing recently? I’ve been using a model to try and figure out, what is actually the value of an additional test relative to its cost? Models definitely have their role, but you’ve got to stick to the idea that a fact beats a theory every time.

Now, the thing is, just very clear, that a test is worth a lot more than it would cost us to provide. Why aren’t we delivering more? I think there’s a genuine confusion and puzzlement about how to increase the supply of tests. And because people don’t know how to go about increasing it, they say it’s not possible. They treat it as if it’s just something beyond our control.

I think we have to look carefully at what changes should we make in policy to increase the supply of tests. One part of that, as I’ve been saying, is we just have to pay for them. If we put up enough money, we can get tests. It’s like I’ve been saying: if we spent about twice as much on tests as we spend on soda, we could have all the tests we need, like 23 million tests a day.

If we spent about twice as much on tests as we spend on soda, we could have all the tests we need, like 23 million tests a day.

So first, you’ve got to provide money. Because tests are a public good, this has got to be money that comes from the government. It’s hard to get there with having consumers pay. But the Congress has allocated $25 billion. There’s a proposal now for another $75 billion from the Democratic plan. So, we’re getting there on the money.

The other side, which frankly, I’m in this position — think about what I was saying before — I’m generally in the position of defending regulation. The more I’ve looked at the role of the FDA in holding back progress in testing, the more I’ve concluded this is a case where we have to say, as economists often say, “This regulation is just getting the cost-benefit tradeoff wrong.”

It’s way too restrictive. There’s little harm from tests. Tests don’t hurt people. It’s not like a vaccine or a pharmaceutical agent, so that the FDA is just needlessly slowing down innovation that could otherwise flourish. So, pay some money, and then get the FDA out of the way. And then all of these very clever researchers and university labs all across this country — they could give us all the tests we need.

COWEN: Germany has done a great deal with testing, as you know. But at least now, as we’re speaking, as of, I think, May 12th, their R is still over one. Does that worry you? Does testing really get you into the promised land if you’re not a small island?

ROMER: It does worry me. And I don’t think that, as well as they’ve done . . .

Let’s just pause for a second because this is a little tricky. If R is equal to one, that means that the number of deaths, the number of infections will stay constant over time. So you can have R0 equal to one at a low level of infection and low rate of deaths, which is where Germany is. We have R0 about equal to one at a higher rate of death.

But in any case, it is worrisome because what we want for suppression is R0 significantly less than one. And Germany is not testing at the scale that I would propose, and I’m afraid that the way to get there is that even Germany is going to have to do more testing, including more testing where you’re just kind of fishing for people who were infected. You just test people who are asymptomatic. You’ve got no indication that they were in contact, but you test them anyway because that’s the only way to find some of the people who are currently infected.

COWEN: Do you worry that some of the countries that have done the best with testing have combined it with forced quarantine and that maybe you need forced quarantine for testing to work?

ROMER: Well, I was critical. It’s been very funny to go from that real kind of angst, almost like crisis of the review I wrote of what economists have done, but then to shift into economist mode where I think we can actually provide some real benefit and some clarity in these conversations.

The way I frame this on testing is, first ask, what would be the value of a particular piece of information? How valuable would it be if we know who’s infected and who’s not? Then, given that information, separately let’s think about, what’s a good way to use that information? And I think there’s some open questions about how best to use this.

I have some colleagues who I’ve written a paper on because they were also promoting this idea of test everyone. Their view is that one of the ways we might do this is, at home, get devices that can test you at home, so everybody finds out if they’re infected or not. Their attitude is, that may be all you need to do because once people know they’re infected, they’ll take decisions, take actions to protect their colleagues, the people they know.

Most people are responsible. Most people don’t want to inflict harm on others. They may well just self-isolate. So maybe with enough testing, we just let everybody know, are you infectious or not? And that’s all we have to do. You could go to the other extreme and have some government system where the test system has to report every positive, and the government forces quarantine on people.

I don’t think you need that. I don’t think you’d get that much benefit out of that. And it’s got a lot of potential costs. So let’s get that information, and then let’s use it very gently. First, just let people know and let them adjust. Second, maybe give someone like . . .

I keep talking about, recovery means I can actually go back to the dentist. Maybe my dentist will say, “Paul, I don’t want to be working in your mouth — and you can’t be wearing a mask when you’re in the dental chair — unless you have a recent test that shows that you’re not positive. Then fine, you can come on in, and I can work on your teeth.”

We might give other people the right and the ability to say, “There’s certain things you can’t do, certain services you can’t have access to unless you can show that you got a negative.” Restaurants might offer sit-down meals but say, “You can only get a reservation if you can show us you’ve got a negative test result in the last couple of days.”

We can use this information in ways that I think aren’t very oppressive, aren’t very risky. They could let us go back to going to the dentist and having restaurant meals and do it without big risks to our freedoms.

COWEN: Take the people who test positive. It seems that at some point they’re likely to be immune, and in a sense, they’re more valuable as workers.

ROMER: Yeah.

COWEN: When do we give them the clear? I read papers, “Oh, you can be infectious for up to five weeks, maybe more.” We’re in a very risk-averse society. Don’t you run the risk by getting a test at all that, in essence, you end up locked out of polite society?

ROMER: Well, again, this is where I’m defending science and economics as science. Here’s really the science of medicine. We need to help everybody know, “Here’s what the facts are. Based on these kinds of signals or this elapsed time, you can be confident that a person is not infectious any longer.”

And then people may still have some emotional aversive reactions. But I think if we can just credibly provide the facts, then that will start to change practice, and practice will start to change some of those deeper emotions.

COWEN: Should there be a liability waiver for businesses that test their employees? We all know there are false negatives and positives, in fact.

ROMER: Yeah.

COWEN: Say your business tests you. They tell you you don’t have it. It turns out you do have it. You infect your spouse. Should there be a liability waiver to encourage this testing?

ROMER: You know, for vaccines, we created a special compensation mechanism so that, instead of litigating, somebody who’s harmed by taking a vaccine — because there’s a small fraction of the population that has a negative side effect — there’s a separate compensation mechanism. I think there are many reasons to think that our judicial system is an ineffective way to address a harm or to provide insurance. And it slows down many important things that we need to do.

But I’d be more in favor of a broader look at ways to improve the functioning of the judicial system, rather than just do a . . . Actually, I don’t have a strong view on this. It may be that to move quickly, we want to have a special patch related to what firms do with test information. But I don’t think we should stop there. We really should be asking, how can we tune the judicial system to make it work better?

COWEN: Could it be that litigation is the ultimate reason why America is so slow in testing? That any big push for anything — someone can raise their hand and object. Someone could sue. “Well, this violates The Health Insurance Privacy Act (HIPAA).” I’m not even sure it does, but you would need a ruling. Someone sues on disabilities regulation. “Oh, I need to have this app. I can’t read it.” Someone sues about masks. “Well, I can’t do lip reading.” The actual solution — something we’re far from — and that’s to clear away all this emphasis on litigation in American policy. And economists have been mostly right about that, too. Or not?

ROMER: Yeah, yeah. Well, my dive into testing has persuaded me that the FDA is far more important as a force that’s slowing down progress there. There’s been speculation about lawsuits, but there’s really little indication that those will materialize. And the people I talk to who can’t do things they want to do in testing are failing to do it because of the concern about the FDA. So I don’t think the facts support that litigation is the big threat here.

Also, in terms of moving quickly, I think one of the things we could leverage — because this is a public good — is the sovereign immunity of the states. I think the states can actually just purchase the test, say, with money they get from the feds and then even give instructions about “here are ways to use these tests.”

Those could even be regulations. “Here is what you have to do. If you’re a restaurant, if your employees test negative, you can open. You have legal permission to open. And you have to require that people test negative. But if you do, that’s fine.”

If somebody who tests negative goes to restaurants and other people get infected because of that, the restaurant could actually have the protection of the mandate from the state that this is what you should do to protect public health. So I think the states could actually provide cover for firms to do — and individuals to do — what’s best in terms of how to use this test information.

COWEN: Let’s say we make you testing czar, and the Romer regime is put into place. Over the first month, what percentage of Americans do you think would show up to be tested?

ROMER: Well, I would try and do a calculation about where might tests be most valuable. And if the states are the ones who are buying tests and providing them, encourage states to use them for those high-value purposes. I think frequent testing in nursing homes might be all it would take to cut the death rate in half right now. The estimates are that as many as half of the deaths are actually taking place in nursing homes. And it seems to me that there’s no hope for contact tracing there.

There’s talk about rebuilding all of the nursing homes. That’s not going to happen anytime soon. But if you tested everybody, initially every day so you know exactly who’s infectious inside a nursing home — test all of the staff; test all of the visitors — then we should be able to isolate the few who are infectious and really bring down the deaths in nursing homes. So I’d use those, there, first.

Second, I think it would be great to get Major League Baseball started again. I think we should use the relatively small number of tests it would take to test all the baseball players every day, and let them start playing games in empty stadiums, because you need a lot more tests to test the fans, to have them come in. But we could be playing baseball in empty stadiums without any risks that we’re increasing this R factor, and people enjoy baseball.

It would be an important signal of how we go back to work in this regime. I think that could be an important complement to nursing homes. To keep going —

COWEN: There’s a study out. I think it came out May 10th or May 11th, and they did test everyone in Major League Baseball — a lot of the staff, not just the players. And hardly any of them are COVID positive. We tested so many of the NBA players. But given those sports are still not reopening, doesn’t that mean testing isn’t enough?

ROMER: Well, nobody made a plan, which says . . . Look, there was an initial plan, which is put all the baseball players in a big dome or something and isolate them. But obviously, they don’t want that. They’re going to be going home to their families. Some of them are going to get infected in their families, so you need a plan for testing and retesting the baseball players if you want to make sure that one player doesn’t infect another.

You need to do some calculations about, how frequently do we need to test? Also, you alluded to this point before, which is, how do we have to respond when somebody tests positive? How long should they be isolated, both from other players but also from the general public? But if we just put together the plan, I think we could safely restart baseball and do it with confidence, knowing that we’re not going to increase the number of infections.

COWEN: Not under the Romer regime, but in the world we live in, can we reopen our colleges and universities for this coming fall?

ROMER: Well that’s one of my list of plans to actually work out. So there’s Major League Baseball, but then universities and then K–12 education, I think, are the next two. Part of the reality is that people are afraid of opening universities and K–12 education right now.

If we had the tests, we can show everyone, if you test people frequently, isolate the few who are infectious as soon as you find out that they’re infectious, you can actually let people start to interact again without raising this R number. So I think it’s totally possible to reopen universities and reopen schools.

The universities — you may make some adjustments beyond just test and isolate. It may be that a 300-person lecture hall, unless it’s well ventilated, is just too risky because even just one person who’s infectious could infect many more, and we’d have to see if that’s true. So you might have to have better ventilation or not have those big lecture halls, but we could surely restart university education, restart K–12. And these would be very important things to do because we know how valuable human capital is. We know how high the returns are to those kinds of investments.

I said before, I was doing some calculations the last couple of days. The calculation I’m looking at is for each unit of testing capacity. And if we could test one more person each day, how many more jobs or how many more people could reenter, return to their previous activities?

The model suggests that it’s about nine. Testing one person per day throughout the year would free up about nine people who could go back to doing what they were doing before, get out of the shelter-in-place rules and have no net effect on the reproduction number R because the tests depress it. More people in circulation raise it. You just set those numbers so that they balance each other out. Nine economically active people is worth a hell of a lot more than it costs to provide one test a day for a year.

On Romer’s plan vs. Weyl’s

COWEN: How does your testing idea differ from Glen Weyl’s testing idea?

ROMER: I think Glen and I are in agreement that tests are very valuable. Glen thinks that we can target the tests. I’m saying just test everybody on a regular basis. Glen is saying you don’t have to test everybody. What you can do is target your tests to people who are more likely to be infectious. I agree that if you can target tests effectively, then you don’t have to test as many people because really all you have to do is find enough positives and get them into isolation.

But I think Glen is assuming that the way we’re going to predict safely and reliably who’s infectious is through apps that do digital contact tracing, and I’m skeptical that that’s going to be ready in time and ready in the sense that everybody will be comfortable using it.

I’m saying if we want to have a plan that we know we can execute now, where we know we’re not going to have a divisive fight and get stuck because we can’t make a decision, the way to do that is just don’t make the digital contact tracing part of the critical path. Just create a path where we get there, whether or not apps could work. If it works, great. I’m not opposed to targeting the tests if you’ve got a good way to do it, but don’t make that a requirement for the path that protects us all.

COWEN: Would you ever get involved in another charter city project?

ROMER: Actually, before I leave the testing — one of the things, as economists and scientists, I think we really can usefully bring to these debates is just quantification, just talk about the numbers. This morning, I was trying to think the best estimate, say, from New York State is that this infection fatality rate is about a half a percent. So if a half a percent of the people who catch an infection die, and if you look at where we have got about 2,000 people a day in the United States who are dying, that means there’re about 400,000 people a day who are newly infected.

Now, each of those 400,000 people has, say, 10 contacts, which I think is modest — it could be more. That means that there’s 4 million people a day that you’ve got to go out and find with your contact tracers. I’m not sure we’ve got the capacity to do that. But the real point here is that whether you follow Glen’s model or my model, you’re already up to 4 million tests a day, which is 10 times the capacity we’ve got.

So let’s not even argue about whether Glen’s right or I’m right. Let’s just get a lot more tests because both of us think we need way more than we have.

On trying again with charter cities

COWEN: Okay. Charter cities — would you try it again someday?

ROMER: Sure, absolutely.

COWEN: Under what conditions?

ROMER: I might rename it. I don’t know that, in communicating the idea, that charter city is the best name, but I think the idea is still a compelling alternative. And unfortunately — maybe this is now my shtick — it’s like a $100 billion a year on testing. It’s an unpleasant, bad idea that nobody likes, but it’s just better than the alternatives. The same thing is true, I think, on migration flows, as on dealing with the pandemic, which is that the alternatives are so terrible, the best option may be something that’s kind of bad, it’s kind of expensive — we just do it anyway.

I’m not sure we should call it charter city, but the thing I think we could do is create new cities that would solve the current impasse, where you’ve got 750 million people who say they want to leave the countries where they currently live, and the existing countries that say, “We don’t want to take that many people.”

So I’m saying, “Okay, what’s the middle ground here? What’s the deal we could do? Let’s create some new places that are still places that people want to go to, but where nobody in existing countries feels threatened by the creation of those new places. And let’s try and offer that as a solution to what seems like this impasse.”

COWEN: How do you frame what happened in Honduras, conceptually?

ROMER: I thought in selling this idea, to do this, you’d need both some country that is willing to volunteer or supply the location for a new jurisdiction, and then some countries — a country, or more than one country — that can help establish the new jurisdiction, like its legal systems and so forth, and administrative — all the systems you’d need. I thought the biggest risk constraint on this idea was that it would be hard to find countries that would be willing to say, “You could use our land to start something new.”

So I spent time in Madagascar. I spent time in Honduras. They were actually willing to try this, but what I think, in retrospect, I should have done, and what I’ll do now, is go first to the countries which are willing to help set something up because a country like Honduras was not — the reason it was willing to do something radical like a new charter city was that it did not have the internal capacity to do something like a charter city.

What went wrong was that we couldn’t get sufficient participation from outside of Honduras in setting this up. And then, frankly, in Honduras there was a little bit of lack of transparency. They didn’t really want outsiders either because it was kind of a small group that actually wanted to set these things up and control it internally.

So I think the scarce player, the short side of this market is going to be countries that are willing to say, “We will help set up a new place that people can go to.” It’s the citizens of those potential countries that we need to persuade. This would be worth trying. And if they’re willing to do it, then I think we can find locations where it could be done.

COWEN: Do you worry about a negative selection effect in the volunteers? In a lot of your work, you’re concerned with corruption, quite appropriately, I would say. And could it be, the countries that want to do charter cities — well, it’s one branch of the government wants to do something a little funny without the other branches of the government seeing, and, in essence, cut its own deal, and that there’s something intrinsically worrisome about a country volunteering to do it.

ROMER: I think this is one of these places where we have to be willing to just select from the feasible alternatives and not hold out for some ideal that we can’t achieve. It’s worth being specific here. My hunch is that China will eventually realize that the way to pay for the infrastructure that it’s building as part of this Belt and Road program is to do urban real estate development. The transport never pays for itself. It’s always the real estate that goes up in value that you use to pay for all of this.

So I think that the Belt and Road project is inevitably going to turn to a version of new city, city-scale real estate development to finance what they’re trying to do. In parallel, the United States could be offering its own version of cities around the world that are new. There’s gains in the value of the land that pay for the stuff you want to do.

Then to answer your question, would I be worried if that’s the way that China and the US compete with each other? Actually, no, I think that would be pretty good. The Chinese wouldn’t set up those cities and run them exactly the way that somebody from the United States might prefer. But if people who want to migrate could choose between a Chinese location and a US location, that would put some pressure on both the US and China to organize these new opportunities in ways that really benefit the people who will go there.

COWEN: How important is religion for explaining economic development? You said before, norms are important, and charter cities, in a way, are identifying laws, rules, norms as a public-good legal structure. So why isn’t religion also a key?

ROMER: Well, I think it’s important for us to think about what are the mechanisms that we use to try and shape norms over time. Some of them are just an invisible-hand process where nobody’s in charge and norms often, I think, go in directions that are beneficial and appropriate.

There’s a great book you may know of, called The Civilizing Process, that looked from the Middle Ages up to the present, looked at norms about just what it means to be polite or civilized, even just table manners. And it’s really a fascinating account.

Some of this happens automatically, but some of it happens because of activists and organizations and structures like churches, and we should be at least mindful of, what are the ways in which those different bodies can push norms? What are the ways that are beneficial to everyone, like that increase efficiency? What are the ones that might harm efficiency? How do we get more of the ones that increase efficiency?

COWEN: Say I’m a Christian missionary. I’m working in Nigeria, and say I’m fairly persuasive and effective. Is it possible I’m doing more for economic development than any economist?

ROMER: It’s possible, but you’d really want to look in detail and see which parts of the norms that are being conveyed there are beneficial and which parts are not. One also has to be thoughtful about the fact that you should ask, are the people who are being socialized into some new norms aware of what the transaction is? And are they agreeing in some sense? Do they actually have some agency and some ability to choose “Yes, I’m okay with this” or “No, maybe not”?

This is why I like the migration decision, because it involves a more affirmative choice. If some missionaries set up a city and said, “Here’s how this city will work. You’re welcome to come,” and people could choose to go to that city or not, and can choose to leave if they don’t like it when they get there — I’d be a lot more comfortable with it.

COWEN: How optimistic are you more generally about the developmental trajectory for sub-Saharan Africa?

ROMER: There’s a saying I picked up from Gordon Brown, that in establishing the rule of law, the first five centuries are always the hardest. I think some parts of this development process are just very slow. If you look around the world, all the efforts since World War II that’s gone into trying to build strong, effective states, to establish the rule of law in a functioning state, I think the external investments in building states have yielded very little.

So we need to think about ways to transfer the functioning of existing states rather than just build them from scratch in existing places. That’s a lot of the impetus behind this charter cities idea. It’s both — you select people coming in who have a particular set of norms that then become the dominant norms in this new place, but you also protect those norms by certain kinds of administrative structures, state functions that reinforce them.

We need to think about ways to transfer the functioning of existing states rather than just build them from scratch in existing places.

If we don’t pay attention to that and just keep doing what we’ve been doing in development assistance, I’m still fairly pessimistic about how many will make the radical transformation that China made.

On reforming the World Bank

COWEN: If you could reform the World Bank, what would you do?

ROMER: Oh, that’s an interesting question. I think the Bank is trying to serve two missions, and it can’t do both. One is a diplomatic function, which I think is very important. The World Bank is a place where somebody who represents the government of China and somebody who represents the government of the United States sit in a conference room and argue, “Should we do A or B?” Not just argue, but discuss, negotiate. On a regular basis, they make decisions.

And it isn’t just China and the US. It’s a bunch of countries. I think it’s very good for personal relationships, for the careers of people who will go on to have other positions in these governments, to have that kind of experience of, basically, diplomatic negotiation over a bunch of relatively small items because it’s a confidence-building measure that makes it possible for countries to make bigger diplomatic decisions when they have to.

That, I think, is the value of the World Bank right now. The problem is that that diplomatic function is inconsistent with the function of being a provider of scientific insight. The scientific endeavor has to be committed to truth, no matter whose feathers get ruffled. There’s certain convenient fictions that are required for diplomacy to work. You start accepting convenient fictions in science, and science is just dead.

So the Bank’s got to decide: is it engaged in diplomacy or science? I think the diplomacy is its unique comparative advantage. Therefore, I think it’s got to get out of the scientific business. It should just outsource its research. It shouldn’t try and be a research organization, and it should just be transparent about what it can be good at and is good at.

COWEN: Do you regret the time you spent there? Or what would you have done differently?

ROMER: Well, I was brought in to reform the research group. People in the Bank could tell that research was dysfunctional there, but shortly after I arrived, the number two, who I think had been behind this initiative, left to go take a position back in finance minister in Indonesia, and a different number two came into the Bank.

In retrospect, what happened was that number two decided we’re not going to reform research. We don’t want any noise. Because you reform things, you’re going to get noise. You’re going to get complaints. All other parts of the Bank had been reformed. Research hadn’t. So I wasted 16 months talking to the number two and the number one and saying, “You understand if I’m really going to reform the research group, there’s going to be noise, and it’s going to be a little contentious. You really want to do this, right?”

And they, “Yeah, no, no, absolutely, full speed ahead. We’re totally 100 percent behind you. We totally agree with each other.” And they were just lying to me. So I would go out and try and do something, and they would undercut every simple thing I tried to do. What I regret is the dishonesty of the leadership and failing to just say what was true, which is, “We changed our minds. We don’t want to reform research anymore.”

So I spent months and months doing really simple things like trying to move two direct reports, who reported to me, who didn’t have the integrity to have the kind of responsibility that they had. But I was facing a bureaucratic system that opposed moving these positions — and I’m not even talking about firing them, just moving them out of the critical positions so other people could fill those roles and do them correctly.

I faced not only internal bureaucratic delays, but my bosses were undercutting me and stopping me from doing this. I finally figured it out and said I was going to resign. They told me, “Oh no, it’d do enormous damage to the Bank if you resigned.” And I still took what they said seriously. So then I went out and just got myself fired. I gave an interview in the Wall Street Journal, which I knew would make them mad.

Then they said, “Okay, well, you broke the rules, so we have to have an investigation.” I said, “No, you don’t have to have an investigation. I broke the rules.” They said, “Okay, well then we have to put you on administrative leave, and you have to sign this agreement where you won’t say anything without our approval.” “I’m not going to do that.” And then they said, “Okay, well then you have to resign.” And I said, “Well, that was what I tried to do on Thursday. I resign.” And that was the end of it.

COWEN: Why are you interested in the American philosopher, Charles Sanders Peirce?

ROMER: Charles —

COWEN: Charles Sanders Peirce.

ROMER: Oh, is that how you pronounce his last name?

COWEN: People say it “Pierce” sometimes, but Peirce is —

ROMER: I was thinking “Pierce,” so funny. [This is better on the podcast -Ed.]

COWEN: The pragmaticist, yes.

ROMER: Because I’m really interested in science, and I think he was a very deep thinker about science from this pragmatic perspective of, how does it work? What does it accomplish? How can we get more of that? I think it was Tim Besley, actually, another economist, pointed me to him. I have to say, it’s heavy going to read his stuff, but I’m still quite interested.

If you go back to what we were saying before, about what could an existing successful society bring when it sets up a new one, I used to think a lot about — and as economists, we talked a lot about — the rule of law. Law is, in some sense, the basis for things like honesty and trust. I’m starting to think that science may have actually been more important for the West in developing a culture where a reputation for integrity and telling the truth became something that was valued. Science may have actually been more important than we realize for that.

COWEN: I very much agree with that — and engineering, right? And engineering also is a broader branch of science. And if you look today at software engineers who have to make things work, they tend to be blunt people who will frequently speak the truth.

ROMER: Yep. When you think about this level of norms — a commitment that it’s a good thing to be honest; it’s a good thing to be disapproving of people who were found not to be honest — that’s very helpful because it helps build trust, and trust is an important part of social interaction. I think we may have underestimated the value of science, so it’s all the more important to support it.

It isn’t just that it gives us some facts that feed into a discussion. It conveys norms about integrity. Also, there’s a harsh side to this, that when you are found to have misled people intentionally, those norms say you’re no longer taken seriously. You’re excluded. You’re not respected or listened to anymore. Those kinds of things are critical for supporting trust. I think that we should learn how to protect science and get it to do its job better in building those norms and encouraging trust.

COWEN: What do you find most interesting in French fiction?

ROMER: Well, actually, let me just bore on for one minute about this.

COWEN: Sure.

ROMER: One of my predecessors at the World Bank as chief economist, Justin Lin, has a very interesting paper on this puzzle of why didn’t China develop the industrial revolution. His argument is basically that China — because there were so many people looking and discovering; they discovered a lot of things, like gunpowder, steel, printing, and so forth — but what China didn’t do was invent the social system we call science. They had some knowledge and some technology. They didn’t invent science. And what was different in Europe was the invention of science.

I found that argument really compelling, and I’ve taken it one step further and think that part of what the West benefited from were notions about integrity and individual responsibility for what we say that fostered trust, and that science indirectly gave us those things.

For any country around the world, it’s worth thinking about — if you’re short on that, if there’s a tendency for a lot of people to cheat on their taxes, to lie about what’s true, if there’s norms that hold a society back in those ways, I think it would be good to think about, how do we rebuild a system where we respect and admire people who consistently tell the truth, and where we look down on, disapprove of people who are found to have intentionally misled us?

COWEN: Do you think the evolution of science in the West has much to do with Christianity and Christian norms, which do emphasize some of those values? And science evolved in the West, right? And out of the church.

ROMER: That’s a very good question. I speculated in one group meeting about a difference between the Old Testament version of Christianity and the New Testament version. And my conjecture was that some of the Old Testament norms were closer to the ones that matter for science.

Christianity really succeeded by competing with other religions, partly because it brought in redemption, forgiveness. The New Testament version of Christianity was a softer, kinder form of Christianity. It may be the older form of Christianity, which is a tradition shared with Judaism, where there was a little bit more strictness about truth and integrity and more harmful consequences from violations of that. It may actually be that earlier tradition that was the one that was most beneficial.

I tried to say this about Old Testament values, and somebody accused me of being anti-Semitic. I was talking about Christianity, and I was actually saying it was good, so I don’t really quite understand. But one has to be a little careful when you talk about these issues.

COWEN: French fiction — what do you find most interesting in that area?

ROMER: Oh, we have a division of labor in my house. My wife is the one who you should ask about French fiction. Right now, her goal is to get me to read any fiction at all. I’m heavily biased towards nonfiction, and she’s trying to broaden my horizons a little bit.

COWEN: But fiction is arguably one of the best ways to understand the norms of a society, right?

ROMER: Yeah, that’s true. So what am I going to cite to support that? A piece of nonfiction. A colleague of mine at NYU who had served as dean for many years — he looked at a large sample of promotion cases, and he then tried to generalize, what are the differences between the humanities and the sciences? What makes these things tick? Where are they similar? Where are they different?

He wrote a really nice book called The Geography of Insight that talks about what’s distinctive about humanities, as opposed to sciences, and how they both contribute to a better understanding of the world that we live in.

COWEN: Last question thread, what did you learn at Burning Man?

ROMER: Sometimes physical presence is necessary to appreciate something like scale. The scale of everything at Burning Man was just totally unexpected, a total surprise for me, even having looked at all of these pictures and so forth. That was one.

Another thing that really stood out, which is not exactly a surprise, but maybe it was the surprise in that group — if you ask, what do people do if you put them in a setting where there’s supposed to be no compensation, no quid pro quo, and you just give them a chance to be there for a week. What do they do?

They work. What people do at Burning Man is they go there and they work. They’ll do a different job, like they’ll work as part of the volunteer police force, or they’ll help maintain sanitation. They’ll work to set up something which offers a service to other people. But there’s enormous satisfaction that we draw from accomplishment and the provision of the output that we produce, making it available to others.

If somebody asked me, “What’s a post-scarcity society going to look like?” Somebody actually said this to me there. He was like, “What does post-scarcity society look like?” People work hard because they like it. They work on things that they care about and they think others will care about, and that’s an encouraging insight, I think, about people.

COWEN: We can leave it at that. Paul Romer, thank you very much. Hope we can do this again someday.

ROMER: Good. My pleasure.