Paul Romer: “I Figured Out How to Get Myself Fired From the World Bank.”

Steven LEVITT: When I was first starting grad school, if you would ask me which economist I wanted to be when I grew up, I would have said Paul Romer. He was the hotshot young economist at the time who was transforming the field of macroeconomics with his breakthrough ideas. 

Well, it only took one graduate-level macroeconomics course for me to figure out that I was not going to be the next Paul Romer. It turned out I had no talent or intuition at all for the issues covered by macroeconomics — topics like inflation and fiscal policy, central banks and economic growth. Instead, I turned my attention to microeconomics: the study of incentives and why people make the choices they make and how public policy can affect those choices. 

Welcome to People I (Mostly) Admire with Steve Levitt.

LEVITT: My life as an economist has turned out pretty great, but there’s still a little part of me that wishes I could have been Paul Romer. Not only has he won the Nobel Prize for his academic breakthroughs, but he also managed to launch a successful tech company, and serve as chief economist for the World Bank. So today, talking with Paul Romer, I’m talking to one of my heroes. 

What makes me a little nervous, though, heading into this interview, is the fact that many of his ideas are very abstract. And I think my biggest challenge will be: how can I get him to express those ideas in a simple enough way that regular people can understand? And to be honest, since I barely understand macroeconomics myself, if I don’t do a good job in this interview, I’m going to be sitting here scratching my head, just like the people listening. 

*      *      *

Steve LEVITT: I feel really lucky to have the chance to talk to you today. Thanks for being on the show.

Paul ROMER: Yeah, it’s a treat. And I feel remiss that we haven’t connected before. But I think for much of my life, I have to go by better late than never, so —

LEVITT: Now, you’re an economist who has taken the insights of economics way beyond the classroom, which we’ll get to. But first, I wanted to talk a little bit about your approach to economics, because myself, I’ve always been heavily driven by empirics and by data. But I think as I’ve gotten older, I have an increasing appreciation for the importance of theory and that, in fact, the data alone almost never give us the answer to the question that we care about.

But what’s so interesting to me is that it seems like watching your career, you’ve had the exact opposite trajectory. You started out being heavily influenced by theory. And more and more, it seems to me, you’ve come to be driven by data. Is that fair?

ROMER: Yeah, this is an interesting arc in my career. It’s true that I was very focused on theory. It’s good when it can help us by encouraging abstraction — take away the distractions and focus on the things that really matter. But the main value is when it shows you something unexpected. You know, the connection that you didn’t anticipate, like when Newton sees that the falling of the apple is linked conceptually to the orbit of the moon around the Earth.

And I still believe in the value of good theory. But over time, I’ve come to have a much greater appreciation for the importance of evidence as well, so my new mantra is that a fact beats a theory every time. And if we keep reminding ourselves about that, it can avoid the hubris that can set in when we get very attached to our own theories.

LEVITT: So I’m about 10 years younger than you are. And so when I got to M.I.T. to do graduate work, it was just at the time when your research on economic growth was having such a profound influence on the profession, and every single macroeconomist at M.I.T. they wanted to be you. Did your impact surprise you?

ROMER: When I first started working on growth, it was literally dead. I was at M.I.T. for two years — I went to Queens University to follow the woman who was going to become my wife, and then set my thesis topic when I had no adult supervision. And it went from being out of fashion, to then being kind of a fad.

LEVITT: It’s interesting that your success came out of not having adult supervision to make sure that you didn’t do anything too important or too radical. My success at M.I.T. actually came from a different source. And that source was that I was so completely and totally ill-equipped to be a graduate student that I didn’t really have any other choice. So while everyone else was out pursuing endogenous growth or trying to write fancy game theory models, I didn’t have any tools. So all I could do is go to data and muck around.

And it turned out to be an amazingly effective constraint because I just decided, “Look, I’m going to fail in this profession anyway. I’m just going to do what’s fun for a couple of years until I go and do some other thing that someone will appreciate.” And it was through a whole lot of luck that people ended up liking what I did, even though what I did was not, in a general sense, highly respected by the profession.

ROMER: So did you have a mentor who encouraged you at all at M.I.T. or were you basically just self-directed?

LEVITT: Jim Poterba was my advisor and he was the best possible advisor I could have had because he stayed out of my way. He was incredibly encouraging. And the only thing that he did wrong — although, of course, it’s crazy to say it’s wrong — is I had written a bunch of my silly little papers and Jim pulled me aside and said, “Look, you’ve written a bunch of silly papers. What you need to do is to write something seminal.”

And I went back and I sat on my couch — and I can remember the fabric on the couch and the smell of everything — I sat there for three months trying to think big, seminal thoughts. And three months later, I literally had nothing. And I realized, “Look, you can’t pretend to be something you’re not. I’m not a seminal guy. I’m a little mess-around-in-the-data guy.” But it was a good lesson to learn, because I’ve been stuck true to that ever since. I know what I’m good at. And I keep on doing that.

ROMER: I think you probably dealt with some of what I dealt with, which is after you do something which gets recognized as very good, everybody’s saying, “O.K., well, do it again. What have you done lately?” And a big important idea is something you stumble on. I think it’s not something you manufacture, and so I ended up going and doing other things because I just didn’t see anything on the horizon.

LEVITT: So you won the Nobel Prize. And I thought that the Nobel Prize committee did an amazing job of elegantly summarizing what you do. So here’s what they said about you and about economics.

They said, “At its heart, economics deals with the management of scarce resources. Nature dictates the main constraints on economic growth and our knowledge determines how well we deal with these constraints. Romer demonstrates how knowledge can function as a driver of long-term economic growth, demonstrating how economic forces govern the willingness of firms to produce new ideas and innovations.” So first, do you feel like that captures well the way you think about economics?

ROMER: Yeah, it does. In a sense, that conclusion was not really a surprise. I mean, people understood technological change was what was driving growth. A lot of people had some intuitive sense for this. The thing that has really sunk in for me after decades is this same possibility of sharing ideas that makes them a source for improvements in standards of living over time.

So I think that was the surprise connection for me, that it’s this non-rivalry or shareability of discoveries that not only makes them a driver of growth going forward, but explains why there’s so much benefit from connecting with so many other people.

LEVITT: Could you just define “non-rival”? Because I think there are some non-economists listening that won’t know what you’re talking about.

ROMER: Yeah. So imagine we had a bottle with some liquid in it that could save the life of your child. And if some other family had a sick child there’d be this intense rivalry over who gets to drink the liquid in the bottle. That would be a rival good. The formula for making that liquid is a non-rival good. That formula is something that everybody can use at the same time. It was difficult to discover. But once they discovered it, at essentially no cost, you can communicate it to any family in the world.

So accommodating those non-rival goods into economics required a departure from the model of perfect competition. Which, as a mathematical theory, perfect competition is the most beautiful theory I’ve ever seen. It’s elegant. It’s beautiful. I love it. But it couldn’t accommodate this notion of non-rival good, which was so important.

And my thesis model of growth was just exactly along these lines, but one of the things I’m actually most proud about in my career is that even by the time I had published my first paper and finished my thesis, I had concluded that that model was just wrong. It just missed something essential. And this actually has a moral and ethical side to it, because if you go back to the Pleistocene, other people were just a threat. The more people around, the less we have per person.

But you switch to this world where we can start discovering ideas. They might discover something which we can then take advantage of. They can all be people that we work with and we all benefit from each other. So I think it enables a kind of growth in the human spirit that could be as important as the growth in the material goods that we enjoy.

LEVITT: So I hear a wonderful optimism in you. But in the back of my mind, my pessimism is: couldn’t we get more gains just by conquering them than trading with them? I mean, the ability for exploitation hangs in the background.

ROMER: Yeah, but fortunately, history shows that conquest is not a very good way to get benefit out of human capital. In that old story about why Britain benefits from Portugal — they trade the wool for the port. The Portuguese have all of this very detailed knowledge and this ability to discover new things. And if you just took over and wiped them out, you’d just have a sunny location. If you make them be your slaves, they’re not going to be making very good port. They’re certainly not going to be innovating.

So I really think the way forward is to find ways to cooperate in a quid pro quo kind of basis. And it does hold out this possibility of a much better world. But then, to be honest, I’ve been struggling with the fact that instead of expanding what I was calling this notion of who is “us,” we’re living through a time where “us-versus-them” is becoming a lot more pronounced and intense. So I do have an empirical problem with my theory that, “Oh, we’re just going to become better people. It’s going to keep getting better.”

LEVITT: Whenever economists talk about insights, the things you’ve been saying, I’m guessing they seemed obvious to regular people who haven’t been trained out of thinking that way by economics. What makes your research hard? Why didn’t the smart economists who were thinking about these questions 20 years ago, 200 years ago — why couldn’t they come to the same insight that you came to?

ROMER: One of the things that struck me is that when you have a group of people trying to figure something out, it’s not a good model to treat them as if they’re a unified individual discovering things. When different people are discovering things and following their own interests, competing, discussing, the outcome is just inevitably going to be different than if you had an omniscient social planner who’s in charge.

One of the inevitable consequences of this structure, with many different people all trying to contribute to discovery, is that we’re likely to get excessive herding. There’s times in science where you can see everybody starts to say, “Oh, I’ve measured the, you know, the mass of the electron. It’s here.” Others try and measure it. They all come up with numbers that are close. A consensus emerges that the distribution of possible values is within this very narrow region. And then all of a sudden everybody shifts. Some new piece of evidence comes in and it’s like, “Oh, my goodness, it’s over here.” And then they all start clustering or herding around a different value.

I think in general, we listen to others who know a lot. We don’t have this unbounded faith in our own observations, our own ability to understand things. So we use, as an input, what others have found. And so, I think one of the less obvious things, but that’s important to explain to outsiders, is that to get the benefits we want from discovery and collective learning, it’s important to have incentives for a wider range of views that get expressed, so that when a new bit of information shows up, it at least gets entered into the conversation rather than having somebody self-censor it because they don’t want to be too far outside of the norm.

There is another side to this, though, that if the barriers of entry go down, you just get crazy views that get thrown in there, which we’ve seen with the Internet, and it takes more time to dispute a crazy view than it takes to articulate one. So there’s a certain level of integrity and a reputation that you’re worried about maintaining in science that has to be in place before you’re qualified to join the conversation.

LEVITT: I want to ask you about one of the craziest things I think that maybe an economist has ever tried to do, which was that you essentially tried to create a city in a developing country that would be separate from the country and independently run. Could you tell us about that idea?

ROMER: Oh, yeah. The origins of this was a presentation I gave, commissioned by U.S.A.I.D. The title was, “Building a Hong Kong for Africa.” So what I was thinking was: Hong Kong created a local example of a successful economy that China and other Asian economies could learn from. Why don’t we think about trying to do the same thing in Africa? Find a way to create by bringing in Western institutions like rule of law, create a small city-state that could be the seed from which a more modern economic system could develop in the rest of Africa.

As anybody, I guess, besides me, will notice, this raises all kinds of alarms about colonialism and the history of colonial exploitation in Africa. And even though the intent of what I was proposing didn’t imply anything like that, it had this guilt by association that meant that everybody was a little bit horrified by the suggestion.

But I felt like we just needed to talk about how we could try something else in development, because development assistance, frankly, has been a failure. I think economists in official aid have really done very little. So I thought it was time to broaden the set of things we were talking about.

LEVITT: Could you explain a bit more about one of these cities, charter cities, you call them, which are deliberately separated from everything else? So how would that change the surrounding cities and countries?

ROMER: The more I thought about this notion of a charter city, the more I realized it was actually a mechanism to help a new set of norms emerge in a situation where the traditional norms, which had been efficient in the past, were no longer appropriate in a new context. And because norms tend to be transmitted horizontally — we’re conformists, we tend to do what other people around us do — it’s very difficult to get a coordinated change in the norms in some location.

So then what you do to create new norms and to give them a chance to survive, create a new place where they can go, and then when they associate with each other, the new norms can be self-reinforcing. So the charter cities idea is really a story about selecting a founding population with a different set of norms than the prevailing one in the existing population and then letting those norms reinforce each other in this new location.

And if they’re genuinely better, other people will join to just get the advantages those offer. They’ll get socialized into the new norms. And it gives you a way to basically carry out a change in what would otherwise be a stable system of norms. So that was the intellectual idea behind all of this.

LEVITT: I think you make a really fundamental insight in what I’ve heard you talk about this before, which is: there’s something like 750 million people who, given the choice to leave where they are, would like to be somewhere else, a place with better-developed, rule-of-law norms.

And the kinds of discussions we have an immigration policy are not even beginning to address that fundamental need. Now, what’s interesting about you is you didn’t just give this presentation, but you actually went out and tried to make this a reality. And it sounds like you got pretty close.

ROMER: If somebody just says, “I don’t think it’s ever going to happen,” those conversations just don’t go anywhere. So I thought if we could make something happen, then people would have to engage with this idea. So I was fairly involved in conversations in Madagascar. I was engaged in conversations with Honduras where they actually passed some enabling legislation.

I think, to be honest, I’d made a miscalculation, which was: I thought the hardest thing would be to get a developing country that could say, “We could do this within our borders, invite somebody else in.” The harder thing was to get some kind of a Western sponsor to play the role that Britain played in Hong Kong. And so I should have been focusing first on the sponsor rather than the place, because the fundamental problem here is that if a country has not got strong enough institutions to develop successfully, they may not have strong enough institutions to really stick to the idea of this sort of charter city.

So it ended up not succeeding in either Madagascar or Honduras. But, after the refugee crisis in Europe, European governments got very interested in this idea. So I think it will come back to life at some point, but I just think some of these ideas have a very long gestation lag. And you have to wait for the right opportunity.

You’re listening to People I (Mostly) Admire with Steve Levitt and his conversation with Nobel-Prize-winning economist, Paul Romer. They’ll return after this short break.

*      *      *

LEVITT: So I have to say, as crazy as that charter city idea is, I love it. I think it would be so important, so valuable. And I hope that somehow, some way, Paul Romer will get that launched. Now, for the second part of this interview, I want to ask Paul about quitting. It’s something he has done a lot of and I suspect he’ll have deep insights into it. And also, I want to ask him about a bunch of different controversies he’s been right in the middle of. Let’s see if I can get him to open up on some touchy subjects.

LEVITT: So you served as chief economist at the World Bank, which is a dream job for many economists, but for you it seems like it was more like a nightmare. What happened there?

ROMER: With the World Bank, I was brought in to reform the research department and then once I got there, they decided they didn’t want any noise. You can’t reform an organization if people don’t want noise. And so I spent 14, 16 months fighting to try and do some reforms against the opposition from the top. I realized I was never going to win that fight.

I was persuaded that it would be very damaging to the bank if I quit, so then I gave an interview that I knew would be controversial. And I was appropriately contrite afterwards, because I didn’t want to do damage to the organization, but I did think it was a waste of time to continue there. So I figured out how to get myself fired and it worked.

LEVITT: So I’m going to call you a quitter, and for most people those would be fighting words, but I think you’ll actually be one of the few people who would consider that a compliment. Because as I look at your past, you have quit more things than just about anybody I’ve ever met. Do you have some insight into what makes you such a quitter?

ROMER: Yeah. Well, I’d never framed it the way you do, but I like that in-your-face descriptive of “quitter.” The way I’ve tried to offer advice to Ph.D. students is by saying, “There’s always something else you can do.” So if you reach a roadblock, and especially if you feel like your notion of right and wrong is going to require a compromise to continue to get past the roadblock, go do something else.

I think this was something that I’d learned in my family, because my father had been a Democratic politician. He ran for the U.S. Senate and lost in the middle of Johnson‘s term, ’66. But in ‘65, he went and visited Vietnam. And he was already having doubts about the Democratic Party’s commitment to the Vietnam War at that point. He started voicing those doubts and was basically kicked out of the Democratic Party.

So he’d been this rising star and was sort of out in the wilderness. And then some 10 years later, when another Democrat, Dick Lamm, became the governor, my dad was pulled in to be the secretary of agriculture, eventually became the elected state treasurer, then elected to become governor.

When he decided he was just leaving the party rather than conform about the war, he had no idea that it would play out that way. But he knew that there were other things in life he could do. So I think that gave me the confidence to just be unafraid about trying something different.

LEVITT: So tell us some of the things you’ve quit.

ROMER: I’m trying to think about clear cases where I’ve quit.

LEVITT: Well, you quit cosmology to be an economist. I don’t know if you consider that quitting, but that was a switch that you made, I know.

ROMER: Oh, that’s true. That was because physics had benefited quite a bit from all of the funding that went into the NASA program. Those funds were being cut, so it looked like it was going to be tough to make a living in physics. And to be honest, I didn’t think I was enough of a genius to survive in physics. I wasn’t going to be Einstein, but I just stuck at it in economics and it turned out well.

LEVITT: Although even then you’ve gone in and out of economics or in and out of academia, at least, at one point, you went to Silicon Valley.

ROMER: I just needed a change. So, I decided I’d go start this EdTech software company. This was — my lawyer told me we were the last deal that got funded in 2001 before the bubble completely burst.

LEVITT: So when I listen to you talk, you are clearly an academic to me. I can see the effect of academia in the way you think. I’m having a hard time imagining you running a successful Internet business. What was that like for you? That must have been such a culture clash with how you’ve been used to operating.

ROMER: Well, I knew I was not a manager, so I hired a C.E.O. She hired a C.T.O. I was the random idea generator. And that was the right choice for a while. We did get to this point, though, where they were betting on the Valley strategy, which is that you just have a good story to go do a second round and raise some more money.

And what I could see was, there was going to be no more money, so we had to get to cash positive where we could stay alive. And we had a real division over this. And then I had to fire the C.E.O., the C.T.O.. I managed to do a family and friends round, just to get a little bit more money. And we drove towards getting to just cash positive. And we saved the company.

LEVITT: So I’ve thought a lot about quitting. I did this big randomized experiment where I got people who were having trouble making decisions to come online to my site, and they flipped a bunch of coins. And the people who got heads, I said, “You should quit whatever you’re doing. Break up your relationship. Quit your job.” And people who got tails, I told them not to.

And it turned out that the people who were told to quit, they both were more likely to quit and they ended up being happier six months later. So I’ve got this very empirical view that people don’t quit enough.

ROMER: You know, it’s funny. I do remember reading about that work, but I hadn’t — there’s a part of that I carried away, which was I’m often in a public policy context or other context, telling people, “If it’s hard to make the decision, if it’s really hard to tell which one’s better, it doesn’t matter which one you choose. So just choose one and go.”

LEVITT: Yeah. So I’ve always believed that. And now I’ve gone even one step further, which is: when it’s hard to make a choice, you should always make the change. Whenever anyone comes to me with a problem, whether it’s a family member, student, or anyone, I just listen to what they’re saying.

And I confirm with them that they’re having a hard time making a choice, and then whatever is the biggest change from what they’re doing now is always what I suggest they should do. I’ve always thought of quitting really as just about efficiency maybe, but for you, there’s been this moral piece injected into your decision to quit, which I find really interesting.

ROMER: Yeah. It’s something I think I’ve become more attuned to. I think feeling good about the decisions we make, consistent with our notions of right and wrong, is a very important part of success in life. With my own kids, I used to tell them, “It’s very hard to turn money into happiness, so don’t give up a bunch of happiness to get some money, because you’re not going to be able to turn it back into happiness.” So sticking to your code of beliefs is I think a big part of feeling like one has led a successful life.

LEVITT: One case where you have really maybe sacrificed for the greater good along these exact lines have been with your strong criticisms of your fellow macroeconomists in recent years. What do you see as the big problem with modern macro? And am I right in that you took a real personal hit to try to challenge what’s going on?

ROMER: Oh, yeah. Well, I alluded to this a bit ago. I think this notion of a non-rival good — which is really so fundamental to economics — that notion ran into some real objection amongst the people I most admired in the Midwestern macroeconomic community. And I increasingly found they wouldn’t even use the word “non-rival.” They denied that non-rival goods existed and I got very discouraged about my failure to get them to see what seemed so clearly right to me.

And so, part of quitting and doing the software was that I didn’t see a way forward within economics to get this idea across that I was pushing. And I was afraid that I had actually become a force for encouraging the resistance. So my thought was, “I’ll just leave the scene. This idea is good enough. Other people will come along and push it forward and maybe we’ll make faster progress if I’m not the one who’s doing it.”

LEVITT: But you haven’t always walked away from fights within economics. I know you’ve clashed with some of those same economists over the years.

ROMER: So much later, there was something I saw when I looked back at a paper that my thesis advisor, Bob Lucas, published with a co-author where I thought the math was just, frankly, misleading. And this blended in with broader concerns I’d been having about: how do we protect the integrity of science?

And I felt like it was important to just say that, “Something has gone wrong here — that we’re not using math to clarify what we’re saying, we’re using it to obfuscate, and we’re using it as a tool to try and win an argument instead of trying to figure out what’s true.” And I felt like it was important to make that assertion.

But it was a very painful thing. I’ve never been able to talk to Bob since I made this point. But I think this is one of these cases where what you believe in has to take precedence over your affection for individual people. The other funny thing is people had been talking about me getting the Nobel Prize for like 20 years at that point, maybe more. And if I were the Nobel Prize committee, my calculation would be: we don’t want to take a side in that fight. And then, to my surprise, I got it anyway.

LEVITT: Maybe this is karma for your embracing right and wrong. You were rewarded by the committee.

ROMER: Maybe.

LEVITT: There was a lot of ambivalence towards behavioral economics, bringing in psychological defects and flaws in decision making. But I think microeconomics has really embraced, at the individual level, the idea that there are all sorts of mistakes going on and that’s just O.K. and it’s a better, richer description of the world.

But I get the feeling there’s not yet room in academic macroeconomics for the kind of macro which might be the equivalent of behavioral economics. What do you think?

ROMER: Yeah, I think you’re right, but there should be. There’s actually something very satisfying about a back of the envelope calculation. There’s something satisfying about a nice little proof. But there’s also this complexity threshold. And even if you knew every single detail about the microstructure of a complicated system, if it’s sufficiently complicated, we don’t have the mental capacity to understand the gross macro behavior from the micro foundations.

And so when I realized that no matter how far we go in improving our micro foundations, we’ll never be able to construct a macro model of the entire economy, it really was liberating and it made me think, “Yeah, we have to stop trying to force the strictures of the micro modeling onto the aggregate behavior. You’ve got to allow for things to emerge in the aggregate that are very hard to anticipate from the microstructures.”

LEVITT: It sounds like you aren’t that happy with the current state of economics.

ROMER: Let me just say one thing in favor of economics, because I’ve been critical sometimes about it. But there’s this odd pattern where I can be very moralistic and very critical about deviations from what I think is the right thing to do in terms of the practice of science. But I can also be very dispassionate and distant from what many other people would think of as important moral issues.

And I think that capacity to be aware of your moral judgments and to use them when appropriate, but not to be caught up in them when I think they might get in your way, I think that’s really one of the great strengths of economics, is that it teaches us to think about emotionally-charged issues with a certain level of dispassion. And that clarity of mind can be very important. And it’s not inconsistent with living by a moral code. It’s just part of thinking clearly.

LEVITT: You know what the most interesting thing was for me in that interview? When I asked Paul to name all the things he’s quitting his life, he literally couldn’t come up with one example. In reality, he has quit things over and over — more than just about anyone else I know, so I had expected him to have deep insights about quitting.

But honestly, I got the feeling it was the first time he had ever really thought about it, which is fascinating because this is a guy who has thought deeply about nearly everything — just not about the behavior that, to an outsider, seems to have been the defining feature of his life.

And on a separate note, if you are someone who likes to think deeply about problems, the center I run at the University of Chicago is launching a contest you might be interested in. We’re looking for great ideas for improving animal welfare. We spent months ourselves thinking about the issue and we haven’t had a single good idea. So we’ve put together a competition, and you can learn more about it by visiting our website: risc.uchicago.edu. That’s RISC — spelled R-I-S-C — .uchicago.edu. Check it out. 

*      *      *

People I (Mostly) Admire is part of the Freakonomics Radio Network, and is produced by Freakonomics Radio and Stitcher. Matt Hickey is the producer, and Dan Dzula was the engineer. Our staff also includes Alison Craiglow, Mark McClusky, and Greg Rippin; our intern is Emma Tyrrell. All of the music you heard on this show was composed by Luis Guerra. To listen ad-free, subscribe to Stitcher Premium. We can be reached at radio@freakonomics.com. Thanks for listening.