EconTalk |
Paul Romer on Growth, Cities, and the State of Economics
Apr 22 2019

economic-growth.jpg Nobel Laureate Paul Romer of New York University talks with EconTalk host Russ Roberts about the nature of growth, the role of cities in the economy, and the state of economics. Romer also reflects on his time at the World Bank and why he left his position there as Chief Economist.

Alex Tabarrok on Private Cities
Alex Tabarrok of George Mason University talks to EconTalk host Russ Roberts about a recent paper Tabarrok co-authored with Shruti Rajagopalan on Gurgaon, a city in India that until recently had little or no municipal government. The two discuss the...
Simeon Djankov and Matt Warner on the Doing Business Report and Development Aid
Simeon Djankov, creator of the World Bank's Doing Business Report, and Matt Warner, Chief Operating Officer of Atlas Network talk with EconTalk host Russ Roberts about the role regulation plays in economic development and the challenges of measuring regulatory barriers...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Apr 24 2019 at 8:36am

I particularly enjoyed the honest discussions about the World Bank. We need more transparency from these large institutions that are supposed to be helping the world to be a better place. Applying the charter cities idea, we need to build charter institutions by drastically changing the organizational culture by placing people with the right mindset and who would like to do good and be held accountable.

Apr 24 2019 at 9:29am

I found the discussion on housing very interesting. I hope Russ has more guests in this vein. Perhaps Kevin Erdmann ?

Apr 24 2019 at 12:46pm

Thanks for the discussion.

Romer says that in cities wages go up faster than outside cities.  Is this data adjusted for cost of living or cost of housing?  I moved from a rural area and my pay went up but my housing costs went up dramatically – I paid 3 times as much for a house that was worse than my prior home which wiped out all the gains.  My paycheck was bigger but my disposable income was smaller.

Edward Dodson
May 28 2019 at 8:39pm

Your comment is extremely important, with one correction. The value of a housing unit is the cost of replacement, less depreciation. Within some small differences in labor costs and regulatory fees, housing units can be constructed for nearly the same amount in most locations. What causes the difference in total cost if the price of the land parcel on which the housing unit sits.

What I came to see during years as a market analyst in the banking sector was that when the land-to-total value ratios climbed above 35-40%, residential (and also commercial) property markets became less and less affordable, adding increasing stress on local economies. Thus, for example, any increase in mortgage interest rates could trigger a market contraction.


Luke J
Apr 24 2019 at 3:48pm

I keep lobbying Jeff Bezos to read Romer, Glaeser and instead of asking city mayors for handouts build an Amazon-ville in a state with low property taxes. If he builds it, they will come…

Carl A B Pearson
Apr 25 2019 at 4:03am

Thoroughly enjoyed this episode, particularly the focus on what I would call philosophy and practice of science issues.  One point that really struck me, just after 46m: Romer’s worry re the lack of data-oriented career rewards.

I work in public health (infectious disease outbreaks), and we have a similar challenge: too little data, too much theory. I know my discipline would be so much better if there were more publications that only tried to provide thoroughly document data. But there’s basically no professional reward for “just” data – you have to have some model or regression or whathaveyous.

No idea how to fix that yet, but good to hear other people are struggling with the problem. Hopefully someone will figure it out.

Todd Kreider
Apr 25 2019 at 12:13pm

Paul Romer said:

“I think we’ve acquired more data that the financial sector is very dangerous and the costs to the world are extraordinarily high. This sounds like a joke, but I actually mean it literally – I think if you quantify the losses from the financial sector and set them along side the losses of nuclear power, the financial sector’s done much more harm with its accidents than nuclear power’s done. And why is it that we have such radically different presumptions about the role of regulation for these two industries?”

I thought it was a little funny that Romer felt he had to emphasize his comparison was no joke when what he said was so obviously correct but may not have been aware of it. Cleaning up after Three Mile Island cost over a billion dollars while the Great Recession cost over a trillion dollars – a difference of a factor of a thousand.

Cleaning up after the nuclear accident at Fukushima will cost $180 billion over 30 years, so around $6 billion a year, but Romer’s comparison still easily holds. Much of the $180 billion clean up bill will be due to the Japanese government wrongly decontaminating the area until radiation levels are back to background radiation, one hundred times lower than what is considered safe and ten times lower than Denver. And had there been sounder nuclear regulation, far fewer would have been evacuated and far fewer would have died due to forced evacuation when nobody was at risk from cancer outside the plant.


May 12 2019 at 3:54pm

I worry much more about the costs of collectivism, socialist illfare, Keynesianism, neo-Keynesianism.

I see the World Bank & IMF as more in the way of sleight of hand, con artists trying to obscure, raise the dust or the silt from the bottom of the swamp, shift costs & prosperity from these people to those people.

I see many cities since at least the mid-1800s as over-crowded, shifting wealth from the hinterlands to the power-mad political bosses…in part by declaring the rural value production as being less than it really is, but also via various fraud-based confidence schemes. “Oh, you value not being crowded; that is no value.  See here our wonderful high-rises with 10K people/square mile.  But those few square miles of forest, that ranch, that farm in Texas, Ohio… producing thousands of bushels of nourishment, oxygen & serenity-restoring hiking territory is ‘undeveloped’ ‘waste’ of little value.”

Mort Dubois
Apr 25 2019 at 12:23pm

Re housing prices in cities:  the same old point, that more high rises would help, keeps coming up.  Several thoughts:

 Housing prices are zooming in cities with both good job growth and geographic limits.  Remove one or the other and it’s not such a problem.
You can build buildings as tall as you want, but it doesn’t increase the capacity of the streets.  I suggest you try to deliver a truckload of supplies to any business on Manhattan during the day before you assume that more floors = a real solution.
One would expect a gradual readjustment of jobs to places with lower housing prices, as this would allow employers to reduce costs and attract a better grade of worker for a given dollar.  It looks as though this is starting to happen to the Bay Area and Seattle.
It’s not clear what mechanism other than letting time and markets do their work would lower prices without great imposition on the ability of individual property owners to manage their assets to their best advantage, by political or other means.  It always surprises me to hear Russ sigh about high housing prices, as the alternatives would all involve some kind of top-down imposition on a large number of people.
As to why more people don’t move from poor places to where the jobs are, consider the value of long-established social networks.  Economists never assign any dollar value to these but they are clearly valuable, and difficult to transfer to a new location.


Edward Dodson
May 28 2019 at 8:45pm

The policy solution that addresses the problem is simple economics: cities need to move to a land-only property tax base. Zoning, height restrictions, set-back requirements, off-street parking requirements, etc. will then determine the price one is willing to pay for site acquisition. The annual tax on the land should come close to the potential annual rental value of the location, in this way discouraging holding locations off the market purely for speculation. Land owners would have a financial incentive to bring the land held to its highest, best use — or sell to someone who would.

Apr 25 2019 at 8:53pm

Zero-sum is complicated. But from an ecological view the apex predator takes resources orders of magnitude more than she produces. It is not zero-sum it is 1/100th zero-sum. The effects are just not limited to NYC or even the human species. Order here, ten times that chaos there. It’s physics.

An ecologist with wings should sit on the right shoulder of every economist. I see him looking a little like Reverent Malthus.

Apr 26 2019 at 6:13am

I liked this conversation, and I will put it on my list of potential candidates for 2019 favourites. I would have liked to hear more about the World Bank as I am interested how international organizations operate and what incentives they face.

I also appreciate Romer’s views on the troubles of macroeconomics. When I was a student of economics, I felt that something is wrong with many macroeconomic models. They seemed to be kind of fiction and were not helpful in understanding the real world. I was also troubled by the idea that productivity shocks explain the business cycles. I remember that it was difficult to reconcile the arrival of internet (a clear example of a positive productivity shock) with the real business cycle model because it was not evident in the data. People were torturing data and theory to make the model consistent with the arrival of internet. However, I was too young and unexperienced to challenge my professors on this; I tried but was not successful. I also did not have much knowledge of philosophy of science, which would be helpful in this kind of discussion. So, it is really pleasant to hear that a Nobel prize winner also feels that there is a problem with macro models.  I would have appreciated to hear more arguments on what exactly is wrong with macroeconomics. The discussion was very short on this and Russ quickly moved to the next topic.

I like the idea of charter cities. Of course, there are many barriers, which probably make the idea unfeasible (for the moment). There are legal, political, financial, ethical (how do you keep out people with the wrong values?) and logistical obstacles. However, ideas might also have division of labour. Paul Romer comes up with a general vision and maybe one day there will be somebody who comes up with innovations on how to implement them. However, I do not share Romer’s view that the charter city idea is more promising in countries like Canada, Germany or the UK. People in those countries generally enjoy nice cities. Institutions are good, neighbours are polite and pick up dog poop, people are generally honest, there are many job opportunities and possibilities for entertainment and culture. I do not see any reason why someone living there would like to move to some cleared forest. However, the situation is completely different in many poor countries. If you want to do a business there, you have to deal with excessive and arbitrary bureaucracy, bribes, gangs who force to pay for their “protection”, high level of crime, unenforceable property rights, corrupt courts and police. Someone who has to deal with all this might very much wish to move to a better place. I believe that institutions and norms are key to economic development, and I see a big potential of charter cities or similar innovations/experimentations in poor countries.

Michael Pettengill
Apr 27 2019 at 7:46am

God created a fixed number of objects so more workers must produce fewer objects?

Economist’s understanding of “zero sum” is contrary to that of scientists who coined it.

Zero sum does not mean that when my race car goes faster, yours must go slower, but for my car to beat you I need to figure out how to use more energy to accelerate my car with more force, to which you will respond by using more energy to match or beat me.

And if we race electric cars, we will not deplete the land, even if the US population were to increase to a billion people. The sun shines so much on “wasteland” that existing solar tech caan easily generate plenty of power for a billion electric cars.

It just requires paying consumers lots of money to spend buying stuff they produce in exchange for getting paid to consume. Zero sum is the wages spent to buy stuff workers produce equals the labor cost of the business getting money from customers buying stuff.

Wage income minus living costs equals zero.

Business income minus labor costs equals zero.

Saying entrepreneurs are not workers is insulting to them.

Gertrude Mugizi
Apr 27 2019 at 3:15pm

Very interesting podcast. Thank you very much. I am particularly curious about how you relate learning with wage increases. in my experience this relationship is not as straightforward as you describe. I would be interested to understand the underlying logic of this assumption better.

Apr 29 2019 at 1:00pm

In terms of reduced mobility – high housing prices is mentioned as a reason that people don’t move to the city. It’s also increases your risk of leaving, if you live in a high cost area, and have a manageable situation. I had a very interesting experience when we tried to move our warehouse from Emeryville to Nashville. Our warehouse workers would have gone from always scraping buy to being able to buy a home, yet 100% declined. The reasons were:  (1) reliance on informal networks, esp. for child care, that could not be replicated elsewhere and (2) if they left the SF Bay Area, and it did not work out, there was no way they could get back in. The increased child care costs would have wiped out much of the housing savings for many of them, and there was no way we could ensure they’d be able to get housing again in the Bay Area. At the end of the day, it was just too risky.

Max möllers
Apr 30 2019 at 1:18pm

1:15 mins in they talk about higher productivity in cities than in rural areas. One piece of the Puzzle I did  not hear about is that more jobs and more people means a significantly higher chance of a great match between the two. Better match of the individual’s Skills and what is necessary for the Job means higher productivity for the company means bigger budget for the wage.

Talking from personal experience: when searching for a job to apply my    Niche Phd Computer Science Knowledge was super hard and Would have meant to go to Silicon Valley or be lucky in Finding the Few openings in Munich or Berlin. So I ended up „just“ doing CS stuff without using the additional Knowledge and thus getting an average income for that skillset



Mike Duffy
May 6 2019 at 2:57pm

A bit after 35:53, Paul observes:

“one of my frustrations has been this failure to engage when someone like me raises a question about the logic of an argument. Or, the telling thing for me about the pseudoscience here was the inattention to data. They just stopped letting evidence be the, you know, the decider of what’s going on in the world. The theory somehow became, you know, the definitive way to establish what’s going on.”

Sounds like the climate-change debate, doesn’t it?

Edward Dodson
May 28 2019 at 9:05pm

It was the decentralist reformer Ralph Borsodi who stimulated the creation of self-sustaining communities adopting the land trust structure. Today, the CLT movement is expanding as a way to provide a structure that removes land as a commodity to be bought and sold. One of the most successful such communities in the United States is Fairhope, on Mobile Bay in Alabama. Ground rent charges pay for public goods and services and provide for other public amenities.


Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

This week's guest:

This week's focus:

Additional ideas and people mentioned in this podcast episode:

A few more readings and background resources:

A few more EconTalk podcast episodes:

TimePodcast Episode Highlights

Intro. [Recording date: March 12, 2019.]

Russ Roberts: My guest is economist and 2018 Nobel Laureate Paul Romer of New York University. This is his fourth appearance on EconTalk, the last being in March of 2015. Paul, welcome back.

Paul Romer: And, how many more times do I have to be on before I become the leader? You know?

Russ Roberts: I think only about 32.

Paul Romer: Okay. Well, let's get busy.

Russ Roberts: You're on your way. You're on your way. I want to start with your thoughts on growth, which we've talked about at length here. But I want to talk a recent formulation you gave which I really liked. You contrasted the economics of objects and the economics of ideas. What do you mean by that distinction?

Paul Romer: We have these fundamental units in our models, our economic models, that there's people--and we make some assumptions about people--but there's an external world which is usually characterized by physical objects. I think of that as kind of the world of Malthus. I think it was also the world that we evolved in as a species in, you know, the Pleistocene Era. So, there's scarce objects; we are rivals for the use of those objects. There's some food resources that one group of people can have or another group of people can have; and we're going to be fighting over it: there's a zero-sum game. And, I think the most important implication of that world of objects is if there's more people, it's worse for everybody on average, because with a fixed set of objects, there's fewer objects per person if you have more people. It's just like this iron law, inescapable. Now, the world we live in, especially since the Neolithic Revolution when we settled down and started discovering things is a world with both objects and ideas. Ideas are insights about how to rearrange the objects--to transform them to turn them from things that are less valuable to us into things that are more valuable to us. So, ideas let us get more value out of a fixed set of objects. And the really exciting implication about this to me is that, if there's more people--even more people that are remote from me: I don't know all of them, I don't like 'em--it may still be good to have them around because even though there's less total objects per person, they might discover something valuable. And this is where the difference between objects and ideas surely matters. If they discover something valuable, then I can use it; they can use it at the same time. So, for a smaller set of objects per person on earth, we might actually get more value per person because we're using a lot more ideas to extract value from the fixed set of objects. So, that's the fundamental difference between objects and ideas: this notion of everybody can share it or it's only one person who can use it. It does have this implication that if you keep discovering more ideas, you can keep getting more value, so we'd get growth. But I've recently gotten more excited about the, you know, the broader implications about this--about how we treat other people. So that we start to see other people as allies, at least, not as foes. And I think, in some ways, what's happening is we're learning to live in a world of ideas with a mentality that's carried over, you know, from the Malthusian Pleistocene. And that tension between us versus them and 'they're coming to get us,' versus, 'Wow--when people work together we can do these amazing things'--that tension I think shows up in all different parts of modern life.

Russ Roberts: Well, it's interesting: Jonathan Sacks, former Chief Rabbi of the United Kingdom talks about the dignity of difference and the importance of difference in trade--going back to Ricardo, which is a very narrow example of what you are talking about. The idea that if you are not quite the same as I am, there's opportunities for specialization and then for trade. But I really love the point you are making more generally, because I think so many people's common-sense view of the world is zero-sum; and one of the things that economics training does for you is it allows you to see the possibility of non-zero-sum. Which is radical. And you're pointing out that if another country does better, that doesn't make us poorer. And that's so hard for people to understand and accept. And it goes against how we're hardwired, probably, from those pre-Neolithic times.

Paul Romer: I think so. And what's really exciting me about ideas right now is that the difference in our endowments of physical objects--like, the Portuguese had land with more sun; the British had land with more grass--that could lead to differences in the agricultural products. And then we could show people this notion of gains from trade, a non-zero-sum world. But that was all derived from differences in the kind of endowments. And, the other thing that we were kind of embarrassed about so we didn't highlight, was that if the British could have killed the Portuguese and had the Portuguese land, they would have been even better off. So, it wasn't--they didn't want the Portuguese. They wanted the Portuguese land. They [?]. So, that Pleistocene mentality was very much, 'Let's go pillage and take their stuff. If their stuff is useful we could have it all.' This new insight about ideas says: You know, it's actually good to have the Portuguese around, because even if we all start out being similar, even if we could take away all of their land or they took our land from us, it would still be good to have other people out there discovering things because we're more likely to stumble onto something which is going to be better for everybody.

Russ Roberts: Yeah, that's very deep. And of course--

Paul Romer: Let me push this one point, because some people kind of push me not to use--like they've told me in my Nobel speech, 'Don't use this story about the Portuguese,' or something. But the thing that's really interesting is we can value the Portuguese, not just Portugal. And the usual story about gains from trade says it's good that Portugal exists. The new insight is that 'It's good that the Portuguese exist.' And I think that somehow makes us better people, if we start to think of other people in that kind of light.

Russ Roberts: I couldn't agree more. I think that's fabulous. In fact, you argue that ideas are also about who we are. And I think that's partly getting at this point you just made.


Russ Roberts: I want to push you in a different direction and get your reaction to something I've been thinking about lately, which is that: I don't think economics--what you just said is profound; it does reduce, potentially the dichotomy between us and them--whoever 'them' are. But if we just think about daily life and the things that make our hearts sing, that bring joy, surely some of those things are interactions with other people. Our friendships, love, our families; the connections we make, the communities we belong to. It strikes me that economics has virtually nothing to say about that; and yet we purport to do what we call 'welfare economics,' which is: How do people feel about different policies? And all that stuff, that enormous corpus of economic theory just ignores that social aspect that I mentioned earlier. Do you think about that? Does that bother you?

Paul Romer: Yeah. Well, I always adopt the, kind of the glass half full interpretation of these things if I can. So, I think of it as, 'Oh. This is a way we could expand economics and make it much more useful.' And the way I would do it is by saying this thing we call the utility function, which leads to, you know, some kind of notion of some kind of wellbeing, or, you know, kind of quality of life--there are some things that go into that utility function that we ignore. Some inputs: consumption inputs. And one of the consumption inputs might be consuming the physical presence of another person, and consuming the interaction of the conversation with that person. So, we could include. as part of consumption in utility, social interaction. And I think if we keep pushing along those lines towards what--at one point I was calling 'extended preferences,' then I think we've got just an extension of our usual framework of analysis. We can still do things like welfare economics, talk about efficiency. But, I think we would have much better insight into all the different things that lead to an experience of wellbeing for people.

Russ Roberts: Yes. I don't think that works. I'll tell you why, and then you can see if you agree.

Paul Romer: Yeah. Okay.

Russ Roberts: We know somebody who did that. Gary Becker was really good at it, for an economist.

Paul Romer: Yep.

Russ Roberts: And in his hands it was a somewhat insightful framework, certainly more than it would have been in my hands. I try--

Paul Romer: But, like, which papers are you thinking of by Gary?

Russ Roberts: "A Theory of Social Interactions," which is directly out of an attempt, I would say--I didn't appreciate it at the time, because I didn't know much about Adam Smith, but it was an attempt to bring Smith's The Theory of Moral Sentiments into the framework of utility maximization. And in some ways I think that's trying to put a round peg into a square hole. I don't think it's made for that. Just, for example, an example I use on here sometimes: Smith is about propriety. It's about doing what's proper. Some of Smith--some of Smith. And, of course, that's part of human interaction. That's me not talking too much on this program. It's me not interrupting you too much--

Paul Romer: What are you talking about, Russ? I mean, what's the problem with interrupting?

Russ Roberts: Heh, heh. But it is--it is also, by the way, it's also me trying to make you look good. So, I might edit that part out where you made that interruption, in case people don't know it's a joke. Maybe people don't know that's a joke. And I'd say, 'Even though that made me look better than Paul, the right thing'--and you could say, 'Yeah, but that's in the long run; that's in your wellbeing also,' blah, blah, blah. But my point is, I think we fundamentally miss what's important, many things that are important about life. And why does that matter? It matters for a lot of reasons, but one of which is our obsession with efficiency and growth, both of which you and I care a lot about, but both of which we understand are not the only things that matter.

Paul Romer: Yeah. Yup. So, I think we're agreed about what life is like for us, as we live it. And then, of course, it's just how do we respond to the fact as economists that we are not capturing important parts of this. The reason I asked about which paper by Gary is that in this thinking about extended preferences, I was very influenced by his work on addiction. But, that addresses a different part of our preferences: which is not just that there's other things that influence whether we feel good or bad, but the history of our interactions can influence how we feel about something today. So, if I've, first time I taste beer, I might think beer tastes awful. Or coffee--pick coffee. First time you taste coffee, it's bitter, it's aversive[?], 'Why the hell would anybody drink this?' But then, over time, you can really develop a taste for it. So, I think that's another part of this aspect that we're trying to get at. And I agree: I'm adding, if I can mix my metaphors, I'm adding epicycles to this square peg that I'm trying to force through the hole. But, the way I'm thinking about it is that I have feelings, like utility, about something like--I'll give you an example. When I showed up in New York City from California, I'm standing on the sidewalk; the sign says 'Don't Walk,' and all these people start to walk. So, I have this feeling, which is, I have this itch, this urge to scold those people and tell them, 'You shouldn't do that. The sign says Don't Walk.' So, we can actually feel, like, moralistic about somebody else's behavior, and even feel this desire to scold them, even with some risk: You scold a New Yorker, you might get yelled at or punched. Now, fundamentally I was smart enough to hold my tongue and not say anything about the jaywalking. But, if you're willing to buy into the idea that we not only get utility out of interacting with people, but we can even have a kind of utility which is like a small desire to punish somebody when we think they are violating the rules, then I think even I start to understand why we do things, like scold each other for like not picking up after your dog, or jaywalking. And, if you'll bear with me, now let me bring it back to this history[?] dependence that Gary was talking about. When I came from California, I had developed a set of moral preferences that put crossing against the light in the 'that's bad behavior'-kind of category. So, I saw people doing bad things; I'd have this built-in tendency to want to scold them. After I'd lived in New York--and I paid attention because it was so different--after about 6 weeks, that desire to scold people had completely gone away. I looked at people crossing the street--I crossed the street--when it said 'Don't Walk.' And it was like, 'Yeah, of course. That's what everybody does.' So, my desire to punish people for crossing the street went away, because my preferences shifted--you know, walking against the light--from the bucket of 'That's a bad thing,' to 'Oh, that's just the normal, accepted thing.' So, you could say that when you make it so complicated, you've got preferences which are so flexible they don't really mean anything. But, my view is that they're actually kind of rich enough to describe this complicated behavior that explains why people sometimes want to scold each other, and why we can have like an equilibrium like in my neighborhood everybody picks up after their dog even though the police don't ever give anybody a ticket for doing it.


Russ Roberts: Well, I think you're getting at something else, I think maybe even more important, which is our human urge--also not in our models--to belong. To be part of something. A feeling of community. You happen to have picked something that's actually somewhat dangerous; and it's a dangerous habit that has benefits, because you don't sit and waste time standing on the corner if there's no traffic. So, we understand that. But your general impulse to go along with what's happening around you, whether it's wearing a certain type of clothing, whether it's your hairstyle, whether it's what you do on Sunday morning versus on Sunday night--those are really important tribal parts of our makeup that we neglect at our peril, and have failed to think about as economists. And we shouldn't fail. It's really important. And we assume that stuff compensates for that. So, if we have more stuff but less belonging, we sort of ignore the belonging because we can't measure it. But we can measure the stuff, so we count that. And I've become obsessed in the last few weeks that by looking where the light is, where the stuff we can measure is, we are missing a really important part of the human experience.

Paul Romer: Yeah. Oh, I think that's a very important insight. That, we can have a very rich analytical framework for doing economics, but if it uses as inputs measurements or data on certain kinds of things, we're almost inevitably going to over-attend to the measureable things and not pay attention to the things that are tougher to measure. But, I think what that suggests is that we need to work harder to try and measure those important but hard-to-measure things, rather than just say, 'Well, they're out of bounds. We're just not going to talk about them.' And I think we're actually pretty similar in our view of the world; then the kind of question is how best to get economics to relate to this. It could be that we kind of have a division of labor where other people think about something like norms and moral beliefs in a social context--

Russ Roberts: Community--

Paul Romer: and economists just, we could use, you know, kind of insights from that group. But, maybe it's forced, but I think if we can feed it into our existing apparatus, like a preference function, we could make much more effective use of that stuff. Those insights. And I think those insights are really important.


Russ Roberts: You said, uh, you quoted Faulkner, one of my favorite writers--you said

to be relevant, to offer practical policy advice, economists must embrace the full range of motivation that William Faulkner alluded to in his Nobel banquet speech: "love and honor, pity and pride, and compassion and sacrifice."
And I'm going to--we don't have to go much more on this, but I think--I disagree with you. I don't think we should be looking for better ways to measure those things--

Paul Romer: Okay?

Russ Roberts: I think that's a fool's game. I think that's scientism. I think those are things that don't lend themselves to measurement, don't lend themselves to cost-benefit analysis. They do lend themselves to tradeoffs--there's where I think the apparatus of economics has something to say. And scarcity. That matters. But, I don't think the human heart works that well in a maximization framework all the time. And I think our apparatus encourages us to think of anything we do in that area as scientific when often it's not.

Paul Romer: Yeah. Yep. Well, I hear you. But I think--this is such an important area. If you'll bear with me, I'd like to keep going on it for a second.

Russ Roberts: Sure.

Paul Romer: I spoke with Martin Wolf recently, who was trained as an economist, worked in development, now writes a column. Martin said that years ago he was thinking about writing a paper that he was going to call 'The Value of Values.' And what he was thinking of was that a society can develop certain values. Like, ones you were referring to before: like, courtesy, or propriety. Those might be very important values when you live in a dense, urban context. And, if we can accumulate more of those values, we might all be better off. So, however economists go about their work and what they measure and what statistics they consider and so forth, I think it would be good if we could at least allow for the possibility that makes sense to say, 'As a society, it would be good if we could invest in more values of this type.' Like, more courtesy. Or more value on a reputation for honesty. Now, it still leaves you with some practical questions, like: Can you actually change values in a social context? It's hard, but, you know, if one could--and historically I think societies have sometimes, it's important to be able to talk about the real benefits that could come from different sets of prevailing moral beliefs.

Russ Roberts: Well, I think they matter a ton. And I think all that's correct except for maybe the last part, which is the implication that maybe we can do something about it. I think it's--I think most people would agree that it's better to live in a world with trust than a world without trust, if people are generally honest about trusting each other. It's a little tricky, right--

Paul Romer: Yeah--

Russ Roberts: because cheaters can exploit that; and a world of trust is great for cheaters; and if there's too many cheaters then trust evaporates. But, I think most of us agree that those intrinsic motivations are often much more effective than extrinsic ones, whether, you know, it's prices or punishment or various types of costs and benefits that we might impose on each other. I think the challenge is, like you said: We don't really know how to get there from here. And we might disagree about what they are. And so, civilization, it seems to me, is just the attempt of what we would normally call market forces but they're not typical ones--they are norms and social pressures of various kinds--ideologies, movements, fiction. All kinds of stuff that affects how people look at the world. And then things emerge out of that. Some work better than others. The ones that work better tend to create better societies. But, it's a pretty blunt system. So, you could have a pretty dysfunctional society that works for a long time, where there's little trust; and people struggle and suffer--relative to a different equilibrium, right?--within that culture. But there's not much to do about it. It's pretty hard to change.

Paul Romer: Yeah--

Russ Roberts: A better way to say it is: We don't understand that process very well. If we did, maybe we could do something about it.

Paul Romer: Yeah. Yup. Well, if I can draw a link to charter cities, for a minute, this was part of what was on my mind when I was thinking about charter cities. And let's--let me give you a specific example--

Russ Roberts: Tell people what they are--explain what they are for people who don't remember.

Paul Romer: Okay. But first let me set it up. So, let's imagine--take people in Greece after the midst of the Crisis [the Financial Crisis of 2008--Econlib Ed.]. It's quite possible that most people in Greece would say to themselves, 'I would rather live in a society where everybody feels like it's right to pay taxes.' And, they feel shame if they don't. And, if it is uncovered that somebody didn't pay taxes, there's social--you know--disregard: loss of status for that person. Now, a person living in Greece might say, 'But, I don't live in that society. If I just try and pay my taxes and people find out, they'll think I'm an idiot. And I don't know how to change them.' And so, they might say, 'Boy, there's just nothing we can do about the fact that everybody thinks you should pay your taxes and in Greece, they don't.' But, what we have to think about now is not so much competition within one of these equilibria, but competition between. People can move from Greece to Switzerland. We know that when you take somebody who had like the Greek values that pay--it's a kind of a game to cheat and everybody thinks you are cool if you cheat--you move that person to Switzerland, and pretty soon they are going to think it's bad to cheat on your taxes. And they're going to think you are supposed to wait for the light to say, 'Walk' before you cross the street. So, I think you need to start thinking about competition between different social groups, to think of these dynamics of norms. Now, so, what was the idea of charter cities? It was to basically create a sort of a blank sheet of paper on which one could build a new social group. So, imagine you had a bunch of Greeks, some of whom are in the tail, who really feel like it's important to pay taxes: They wish their fellow citizens agreed with them. Suppose a group of the moralistic, 'we should really contribute to the society,' 'we should bear taxes'--suppose a group of those Greeks moved to a brand new place. Instead of being in the minority, they are the majority population in this new city. And then, as other people move in and join them, they are going to assimilate to their values--like, everybody pays their taxes--just as they would assimilate to Swiss values. So, a new, startup society gives you an option for changing social norms by a breakaway group. And then, slow rate of immigration and assimilation to the new sets of norms. So, working back, what I was trying to think about was: What kinds of practical, legal, and social arrangements would it take to create opportunities for starting from nothing? Just empty land. The possibility of a new city growing up. Because I thought a new city was the smallest--like a city that could get to 10 million people. That was the smallest viable unit you could imagine trying to create in the modern world. But, I think it would be, give us an opportunity to change norms in a way that people living in societies they are unhappy with would really like to change them. So, this was really the motivation for that project.

Russ Roberts: Along with, I assume, the opportunity to have a legal and other extra-norm system of enforcement that might lead to good behavior, as well. Right?

Paul Romer: Well, yeah; but this is--and I wasn't totally explicit about this, because this was evolving in my mind. And I also thought it was going to be a hard, you know, slog to persuade people. But, I'll tell you where I came out. I started thinking about this at first the way economists do, saying, 'Okay. If we had a brand new place, we could start a new city, we could "impose law." A rule of law. And then people would come in, and we'd have the benefits of the rule of law. I think that analysis is just deeply flawed. Because, I think--if you look at places where somebody comes in and they pass laws, they set up courts--if those laws are not consistent with the norms in society--

Russ Roberts: yep--

Paul Romer: nobody follows the laws. The system just collapses. Yeah. So, I think--I decided we had to start first with the norms; and then figure out a way to change the norms. Then on top of those norms, you could build, you know, some supporting laws; and you get a kind of a positive feedback loop between the norms and the laws. But I don't thinking moving first with the laws is likely to succeed. There may be rare cases when it does. But, generally, I think, you really have to figure out how to get motion, how to get things going on the norms. So, a lot of the attention to the Charter Cities turned to practical details. And it also got caught up in two very distinct applications. One could be within a country where they are trying to do an internal reform project by creating a new zone. I kind of refer to this, these days, now as kind of the Shenzhen approach under Deng Xiaoping in China. The other would be an external jurisdiction where somebody outside says, 'Here's a place people could move to if they want to. And if you want to leave the country, you are welcome to come.' And that's more the Hong Kong model. And especially that second model has gotten caught up in the very kind of fraught debate about how to respond to pressure from migration right now. But it was all grounded in a belief that we needed to think about how to give people a chance to actually help change the norms of the society, or choose the norms of the society in which they want to live.


Russ Roberts: And, is there any prospect of them actually happening? It's such an interesting and potentially transformational idea. And I know you've gotten close. But, where do we stand now?

Paul Romer: Yeah. So, I had a few conversations in Madagascar, and then Honduras, where the approach was a little bit like the Deng-Xiaoping tries internal reform with a special zone within a country. This is--even if this was likely to succeed in a few cases, you'd have to expect a fairly high failure rate. But having looked at these things up close, my sense is that a country which is at a low level of development and with low government capacity is going to have real trouble trying to set up a zone that is run differently. You know, in Honduras in particular I was pushing very hard to try and bring in an outside influence; and, you know, that was not in the interest of the actors who were pushing it domestically. So, I've kind of given up on, for most of the developing world, this internal Shenzhen model. I still think there's room for developed countries--like Britain, Germany, Canada--to look for places where they say, 'We're going to replicate the Hong Kong experience. We're going to acquire some land voluntarily'--you know, this, like, there's nothing about the historical experience taking land by force that you want to replicate; but there's ways to replicate, just to get a deal where you acquire some land. And then one of these countries could say, 'We're going to set this up as a jurisdiction that people could, you know, voluntarily move into.' And I think it would be good to think about that as a way to, you know, change norms about things like paying taxes. You know, also change broader norms about, you know, kind of the propriety and civility that we talked about before. Change norms on a bunch of dimensions, that might create societies that people would rather live in. And, I don't think--everybody else seems to think that this is a ridiculously infeasible prospect. I think it's just outside of our realm of comfort and familiarity. But I just don't see any reason why Canada couldn't create, you know, a 20x20, you know, square mile zone some place in the world and say, 'Let's try and create a place people could move to, live in, work in,' and see how it goes.

Russ Roberts: Yeah, it's just such a thought-provoking idea. I'm thinking about areas of the United States that are struggling economically. West Virginia, say. Or rural Ohio. Parts of Kentucky. Places that have lost, say, industries of coal-mining and other activities that used to be productive. The people who used to do those things, for reasons that we might talk about, stay, instead of leaving. And their opportunities are quite diminished as a result. They don't migrate to where there's more opportunity. So, it's not--but I don't think the problem is, is that they've got bad norms there. It's just that the people who are there[?] have low levels of skills and there's nothing there to enhance those skills. Now, Amazon has changed that a little bit. Amazon has put warehouses in low-density places because land is cheap and they've put some stuff in the middle of nowhere. Or, you could put a server farm in the middle of nowhere. But it's not Hong Kong.

Paul Romer: Yep. Well, I think one temptation that I have to resist and I think everybody should resist is this 'I have a hammer, I want to go find a nail'-kind of phenomenon. I don't think charter cities is the right way to think about any kind of policy response to West Virginia or other places where there is some distress. But, on the other hand, it is the case that there are hundreds of millions of people who say they want to leave this year, this society they currently live in. So, I think that is this demand worldwide we should think about trying to accommodate somehow. And, you know, there's obvious concerns about this charter city proposal. Reasonable people, I think, should have qualms, I think, about setting something like this up. But, I think we have to ask ourselves: What's the alternatives that would let us offer, in some humanitarian way, an option to people who want to leave the area? You know, very dysfunctional, dangerous environments? And I think it's very hard to come up with some feasible alternatives to this need.

Russ Roberts: Yeah: no. I'm with you. And of course when the United States had large swaths of undeveloped territory, it was the appropriate place to let people come and just push that western frontier out. And, you know, I like the idea of taking a forest that's owned by the government and not being very productive, either as a forest or as a beautiful place or as a place to grow trees of commercial value, but just let it--let some entrepreneur do something crazy and creative there, and let people come there. So, I love the idea of it. It's great science fiction. And it might be even feasible. Who knows?

Paul Romer: You know, I think--sometimes I try and put myself in the mindset of people in, like, 1730 or something. And, if, you know, if we'd been in a seminar room in 1730 and somebody came in--Thomas Jefferson came in--and said, 'I'm going to design a new country. We're going to have a constitution; we're going to have separation of powers. What do you guys think?' we'd laugh him out of the room. 'That'll never happen.' But, you know, sometimes these things happen. So, I think it's worth keeping that conversation alive, but recognizing that the dynamics are slow and the odds are not very high.

Russ Roberts: Before we leave this topic I just want to ask you: Now that you live more in New York than in California, besides jaywalking, do you walk more quickly from place to place?

Paul Romer: Oh, yeah; I think I do. I walk a lot more. And I think--I haven't quantified this precisely--but, my impression is that even distances that are the same, there was a kind of a habit about, I'd get in my car and drive to the coffee shop. Where here it's like, I walk out the door and I walk to the coffee shop. Even comparable distances I think I'm more likely to walk here than there. That's one of the things I really love about New York, is really freeing myself from my car. If you think about it, it's a kind of strange thing to say, because [?] the car didn't force me to drive it. But somehow an environment where I don't drive it so much makes me feel differently.

Russ Roberts: Yeah: no. I hear you.


Russ Roberts: Let's shift gears. I want to talk about a paper you wrote that got a lot of attention called "The Trouble with Macroeconomics." Without getting too deeply into the weeds of real business cycle theory and other arcana, what did you identify in that paper, in a philosophical sense, or a philosophy of science sense, that you thought was wrong with the state of academic macroeconomics?

Paul Romer: Yeah. Well, let me tell the story of how it emerged. I had written this paper about "Mathiness," which was specific to growth theory and how we used math to, you know, advocate for different growth models. I was critical of a paper that Ed Prescott had written. Somebody sent me an email describing an interaction with Ed in Minnesota where Ed was referring to these productivity shocks that were at the heart of these fluctuations in the real business cycle models. And, as the person described, someone asked, 'What are these productivity shocks? What hits the economy and then makes a lot of people unemployed, or makes output go down?' And the way he described it, Ed was kind of like sputtering for a minute, like couldn't come up with this, say; and he looked out at the traffic and said, 'The productivity shocks are like that congestion out there,' he said, 'the traffic congestion.' This had a very funny effect on me, because it sort of like hit me like a jolt of electricity. Which was: 'Yeah. What are these productivity shocks?' These things are just kind of like made up in the models. And so I started calling them in the paper, like, phlogiston. You know, phlogiston: it was like an imaginary thing that physicists referred to at some point. And, it made me realize that there was something really deeply wrong, I think, with the style of explanation there. And I felt responsible in some sense, as a macroeconomist. I didn't do business cycle theory, but I'm a macroeconomist: I have some responsibility for the profession. I think others of us should have been critical of the style of explanation long before the time when this issue was raised, for me. So, I shifted from my kind-of critique that was specific to growth theory in math to a critique of how macroeconomists--kind of the default leading group of macroeconomists in the profession--were doing their work, and, um, made what is, I think, a very harsh critique of their approach: that, it was not just that they were coming to the wrong answers, but in a more fundamental sense, they were not following the practice of science. They had really departed from science. And I called it pseudoscience. And, I think, this is really a version of an ad hominem attack. It's a kind of criticism you don't want to make lightly; you don't want to encourage easily. But, on the other hand, if you think you've gotten to the point where this is what's going on, I think other members of the profession have a responsibility to, you know, raise their hands and say, 'We need to stop and look at this.' So--and, I should be clear--that paper, I still haven't published, because it was so critical. It was in a circulated draft form, but it still isn't out. The journal would very much like to publish it; and I think I need to go ahead and publish it. But, I not only said that: I think the problem in macro is not just that I think the answers are wrong. I think they've stopped doing science. And I give some illustrations of what I thought about that. I also did something I've never done before in the paper, and I generally disapprove of, which is to use sarcasm as a way to, you know, criticize the results of this group.

Russ Roberts: It's a biting paper. It's a little bit acerbic. Not--I've seen worse. But it's got a little bit.

Paul Romer: Yeah. But I, generally, I don't like the easy recourse that [?]. Just like I don't like ad hominem argument. But, one of the frustrations over my career--and, you know, I kind of know these guys that were friends, classmates, you know; I like 'em as people--one of my frustrations has been this failure to engage when someone like me raises a question about the logic of an argument. Or, the telling thing for me about the pseudoscience here was the inattention to data. They just stopped letting evidence be the, you know, the decider of what's going on in the world. The theory somehow became, you know, the definitive way to establish what's going on. What I felt like--I felt like there was an unwillingness to even engage with well-intended critique. And so, the assertion that this whole pattern of behavior was unacceptable, and the use of the sarcasm, was my attempt to really call attention to this, and say, 'If I'm right, this is a very serious problem, and we need to do something about it.' Because, it's kind of like when somebody pulls the fire alarm. If you do it and it's a false alarm, a person ought to pay. And so, I think if there's a consensus that I was just flat-out wrong--and not just wrong that it's good macroeconomics, but I was wrong to accuse them of not engaging in science--I should pay a big reputational price for that. I really think that. Because you just don't want this to be an everyday thing. But, on the other hand, I do feel like--I think I'm right. And I think it's something that deserves some attention. And, unfortunately, I think it deserves attention even if it hurts the feelings or upsets people that I like and care about. I think what we're doing is more important than just our feelings. And, whether we get the right answers really matters.


Russ Roberts: I have a similar issue, and I don't know if you want to weigh in on it. I'm just going to mention it and make a little bit of a confession to my listeners, who have heard me bring this up many times: I think it's come to be commonly believed by the public, and by very thoughtful journalists, and at least on paper many brilliant economists, that the average American has made no economic progress for the last 40 years--

Paul Romer: Mmhmm.

Russ Roberts: And there's data to support that, of course. And I argue that data is flawed. And, you know, that's an interesting question: you can debate how flawed it is and whether the CPI [Consumer Price Index] is correctly measured; the rate of inflation. The problem I have is that there are other data that show progress--

Paul Romer: Mmhmm.

Russ Roberts: And that side just doesn't engage it. Doesn't put it in their paper. It might be in a footnote--

Paul Romer: Yeah. Yeah.

Russ Roberts: They might say in a footnote, 'So-and-so has done some, made some different assumptions'--say, I'm thinking about the growth in inequality. Which "everyone knows" has grown dramatically. And yet, if you make different assumptions about family size, and how you measure non-monetary benefits--fringe benefits and other things--you get a dramatically--it's not like, 'Oh, you know, it changes by 10%.' You get, like, a 2/3rds reduction. So, it really matters. And if you are a scientist, you shouldn't just say, 'Oh, there are other opinions.' Or not even mention them. You should be grappling with them and trying to understand why you get different results in one group of assumptions and another. And I don't want to accuse those people of having an agenda. And I hate that.--

Paul Romer: Yeah. Right.

Russ Roberts: That's the ad hominem. So, I don't even want to accuse them of being wedded to their own theories--which is part of what you were writing about--that after a while, you become religious about the beliefs you have about how the world works, in what should be, otherwise be, a science. But, it's scary. Because it matters.

Paul Romer: Yeah. Well, you know, I think the kind of things one could do in any area is first, try and do a kind of a census, and see: Do we think it's possible to have a consensus about the facts? We might, okay[?], 'We don't have a consensus yet.' But: Is it possible that there exist facts and we could reach a consensus about those? And I think, if people don't agree on that, then you've really got trouble. I think, in this area, I think people would agree. I'm betting that in some cases what you have is people saying, 'I don't really want to be engaged right now in sorting through all the facts and coming to the consensus about those. I want to start work right away on how to deal with inequality, because I'm pretty convinced it's a really important issue, and I'm guessing that the way it will come out is that inequality has gone up.' You know, I think--we could probably tolerate a certain division of labor along those lines. And then you've got to watch for attendancy[?] for someone doing that to overstate, and say, 'Well, we know that inequality has gone up, so here's a way we could respond to it.' And you kind of call 'em out and say, 'Well, you know, there's actually not a consensus about that yet. We need to look at the measurement issues.' You know, if that's the way that debate is playing out--a little bit of overstatement, the baseline and the motivation for what they want to go do--I think that's all, you know, manageable within science. But there's another problem that I think one should be worried about, which is that there may not be enough professional reward for the hard work of sorting through the facts. And some rising them in a way that people can use them. But, yet, really being careful about the different, you know, ways to summarize them, and the possibility that you are building in some assumptions that are, you know, influencing how you summarize them. So, I, you know, I wish--over time, I've become more and more convinced of the importance of that pure hard work of collecting data and summarizing them in a manageable way. And, frankly, I think we give too many rewards to people who do theory, like I do. And not enough to the data that is actually the basis for judgments about, you know, like, growth, and growth and GDP. And so, as a tweak in the profession, I'd like to see us pay more attention to, you know, the analysis of data and the process of summarizing the data that we then use to make models and make policy recommendations.


Russ Roberts: So, there's something in between, I want your opinion on, actually. Which you've taken us to. Which is, a lot of people are saying now that theory is on the decline. What sells--what is honored and rewarded in our profession--are various econometric techniques for identifying causal connections and actual experiments, and the like. So, it's not the hard work that you are talking about of getting the data right. So, I'm just going to mention something in passing as an example. Average hourly earnings would seem to be the most basic kind of data on this question of how people are doing. And yet, when you look a little more closely at it, you find out it has compositional effects that are quite complicated--

Paul Romer: Mmhmm--

Russ Roberts: and that are ignored, endlessly. In any public presentation of that data: People don't just say, 'It doesn't matter. I'll just put it up there.'

Paul Romer: Yeah.

Russ Roberts: But, that doesn't get any reward. Certainly very little reward. What gets reward is manipulating whatever data is out there in new and creative ways.

Paul Romer: Mmhmm--

Russ Roberts: And, the younger economists that I talk to--we are not the younger economists--

Paul Romer: Mmhmm, mmmhmm--

Russ Roberts: The younger economists like to, they say things like, 'Well, I just let the data speak. I don't have a theory. I don't want to be constrained by theory. I just want to see what the data say.' And I think that is actually intellectually bankrupt. But I'm curious what you think, Paul?

Paul Romer: Well, yeah. Bankrupt--I might attenuate[?], distance myself from like the judgmental or moralistic kind of annotations[?] there.

Russ Roberts: Yup, go ahead.

Paul Romer: But, the equivalent thing we could say is, it's unrealistic, or it's inaccurate to say that the summaries that most people provide of data are free of, you know, beliefs about policy, or free of beliefs about the underlying causal mechanisms. So, I think we need to be skeptical of people who say, 'Oh, my data summary'--you know, 'activities, were, kind of theory-free, model-free, policy-free. And you can just take my results and work with them.' I think we need to look carefully at all the steps and the construction of this data, and just, basically, look empirically to see: What if a bunch of different people went through the same process of summarizing the data? Would they all come to the same--and they were, supposed in the best case, they were working independently--would they all come to the same results? And if not, why not? And then, let's try and figure out what we need to do to reflect this, um, lack of consensus about how to summarize them.

Russ Roberts: That's well said. I probably mentioned this before on the program--I don't go to many seminars any more, but I was at one recently where someone made some rather bold claims about the magnitude of some relationship between, say, the financial sector and, say, GDP [Gross Domestic Product] or growth rates. And I found them unlikely and implausible; but that doesn't matter. I just--but I asked an innocent question. I said, 'How many regressions did you run to get that result?' I don't want to hear why these are the right variables--yet. I mean, I'd like to. I'd like to hear that, too. But before we get there, could you just tell me--like, how many did you run? Did you run 12? Or a thousand? Three hundred? And, eventually I'd like to see--and I asked--how many of them found this result? Because, it's okay if you 300 and 295 show the same thing: it's just a question of magnitude and what else is in the soup. But, otherwise, I kind of want to see what was in the kitchen.

Paul Romer: Yeah. How did he take that question? He or she?

Russ Roberts: It was a he, and he was shocked. He was really taken aback by it.

Paul Romer: Yeah.

Russ Roberts: But, to all young economists out there listening and who go to a lot of more empirical seminars than I do: Please ask that question. Let's turn to your experience--

Paul Romer: Let me just say two things. One, I'm going to make a plug for a book by a colleague, this colleague at Berkeley, Ted McGill. Ted and his co-authors have a new book about, really, kind of bring more transparency and reproducibility to social science. And, um, it's just coming out from Berkeley. And it's great, because it's very practical and pragmatic, and just says, 'Here are some things we could do: keep track of, for example, how much pre-testing has been done on the data. How can we re-set our standards of significant in light of pretesting?' And I think with technology there are actually a lot of ways to do more on this topic. So, I can't remember the name of the book, but it's coming out--

Russ Roberts: We'll link to it--

Paul Romer: and if you are worried, you know, people should be worried about these issues. I think at some point it could become a vulnerability for all of social science if, you know, if we keep, if it keeps coming out that there are weaknesses in our empirical methods or, you know, things that aren't reproducible, then, like, the whole integrity of our, you know, our discipline gets called into question. Ach, here's the title: Transparent and Reproducible Social Science Research. Who could be against that?

Russ Roberts: Well, a long time ago, Ed Leamer wrote a paper called "Taking the Con Out of Econometrics." And, it should have transformed empirical economics. And it did not. You know. He's still working at it. Can't hurt.


Russ Roberts: I want to turn to the World Bank. You were there for about, a little over a year, and year and a half or so.

Paul Romer: Well, yeah. Like, a year and a quarter.

Russ Roberts: And then you left. And some might consider that a short tenure.

Paul Romer: Oh, yeah.

Russ Roberts: I felt--I'm surprised you lasted that long, Paul. So, what happened there? It was a little bit rocky. You were the Chief Economist. But you left. Why? What happened?

Paul Romer: Well, um, if you look my career, I've done a bunch of different things. And, I think, one of the general patterns I've tried to stick to is: Be willing to take a chance. And, like, the Charter Cities, I was willing to take a chance on, say, going to Madagascar, talking to people in Honduras, even knowing that there's maybe a 1 in 10 chance that it's going to succeed. If it was a high-enough value if it succeeded, it's worth doing things that might not succeed. So, take some risks. But, if you learn that it's not succeeding, just go do something else. And, don't--you know, don't just dig in and entrench yourself and keep doing it if it's not going to succeed. So, I thought it was a risky endeavor for me to go the Bank. And, I concluded that I was wasting my time there. I wasn't doing anything of value, I don't think, for the Bank or the world. And so I just decided I was going to leave. And people there were worried about the fact that I wanted to leave, and so there was a little bit--it was kind of messy on the way out; I can fill in the details if you want. But the fundamental dynamic is like, every job is a match. There's a job and there's a person. It wasn't clear to me that this was a good match, to have me go to the Bank. And I think ex post it was not a good match. And, so, once you figure that out, then it's time to just, to go on and move on and do something else. Now, you want me to fill in more of those details?

Russ Roberts: Not yet. First what I want you to talk about is: Most people have no idea what the World Bank does. And, of course, what it actually does is not the same thing as what it is romantically described as doing. I have many friends who work there, by the way. They are the salt of the earth. I'm big fans of 'em. I like 'em a lot. They are well-intentioned good economists. But I don't think it's fulfilled its promise. So, talk about what the Bank is supposed to do; what you think it actually does; and why you hoped to do something when you were there: what that might have been.

Paul Romer: Well, in doing that I want to illustrate part of why it was the awkward match between me and the Bank. I'm increasingly insistent about clarity and communication. And, I'm going to describe the Bank in what I think is a very simple and clear way. They are not comfortable with this level of clarity. But, the Bank is a bank. It's got assets of about $200 billion. It's got liabilities of the same size--I mean, there's equity and then liabilities. The World Bank can borrow at essentially the sovereign rate--the same rate that the United States could borrow at. So, it borrows at a low rate; it lends the countries at a higher rate. And on that $200-billion portfolio, it makes a kind of a net of about $2 billion a year. And it spends that $2 billion a year on its staff. So, that's what the World Bank does. There's a couple of things to note. First, $200 billion dollars is a drop in the bucket in the world of really very substantial financial flows. After World War II, $200 billion would have been a lot. These days, it's pretty small. So, it's appropriate to ask, 'Well, what function is this organization now serving?' Because it has--it's a bank, and the shareholders of the bank are countries. So, there's a Board, with, you know, shares of ownership, because it's the countries that are the shareholders and they back the bank. They can borrow at this sovereign rate. So, as a kind of government-like entity with this privileged position in the financial markets, it's a reasonable question to ask: What should we do with that ability to borrow at these low rates? I think--there used to be a view that the $200 billion in loans could let countries do things they wouldn't otherwise be able to do. But, now, these days, countries could issue debt. They can borrow in private markets. The loans themselves are probably generating a lot less value than they did, uh, you know, 50, 70 years ago. I think the question that one should ask is: With $2 billion a year, if this organization set about providing global public goods, and especially public goods that would be valuable, particularly valuable for people in the developing world, you know--$2 billion could make a big difference. And, I think the fundamental tension at the Bank is that, that $2 billion is really devoted to staff salaries. And, I really doubt the value that's created by the work that those people do--just to be totally blunt about it. And I think the organization should be willing to ask hard questions about, 'Let's quantify: What are the public goods that we think we are providing to the world?' And, 'Is there a way to produce more of them for the resources we could spend?'


Russ Roberts: It's nominally supposed to fight poverty with those loans that it makes. Or to encourage growth, by subsidizing or financing infrastructure of various kinds, or reforms. I think in most people's minds who are casual readers of the newspaper, or Twitter, or Facebook, they see the World Bank as an ideological arm of free market capitalism: That the World Bank lends money to poor countries and then extorts them to change their policies toward globalization, and free market policies in return for those loans, as sort of a form of a blackmail. What's your perception of that perception?

Paul Romer: Yeah. Well, I think, first, if that's your fear, the good news is the Bank doesn't have nearly as much power as it used to. Because, a lot of these countries can go get financial resources from other places. So, part of what's, I think, you know, making people to think the Bank question their new role is: They don't have as much leverage as they used to have. Now, we could talk about whether they used it well or poorly in the past. But they just don't have as much of it. But, let me give you--I tried, when I got there at a Bank desk: 'Well, what's our greatest hits album? What are the best things the Bank has ever done? If we've done a bunch of good things, we ought to be able to quantify and measure particular instances.' Here's, I think, the best thing it's ever done--and this alone might justify its existence. When Deng Xiaoping was trying to, you know, undertake reform in China, they had very serious deficits in terms of their understanding of market economy. They were very worried about removing price controls. They could see that argument for some efficiency benefits for price controls were very worried about removing price controls because they feared inflation, and thought that inflation could lead to, you know, rebellion and collapse of the government. So, the, um, the World Bank Representative in China, was a guy I've met, Edwin Lim: Edwin arranged some basic Macroeconomics 101. So, like, James Tobin gave some macro lectures along with some other macroeconomists on a boat trip on the Yangtze, where very high-level government officials were kind of confined to this boat: they couldn't be distracted to other things, and had to go through the basics of how, at the time, people thought about using aggregate demand policy as a way to control inflation. And that gave them the confidence--it was close fight, but it gave the reformers enough confidence to win the battle and start to liberalize prices. And this was the really critical step towards moving towards the market economy. So, being in that position--being a trusted adviser, the Bank and then the representative of the Bank--being in a position of, as the trusted adviser who could help when a question has a technocratic issue about what can we do in this circumstance? So, you tell them what you know. You bring in experts who fill in the details. That can be a very valuable function. And it may have been critical to the path that China took towards, you know, the rapid growth that it's enjoyed. I should mention that the other thing Deng Xiaoping did was create these Special Zones. Of which ShenZhen was, I think, the most successful. Really of the four, only ShenZhen was a success. But ShenZhen was a phenomenal success. And so, having people who told them, 'This is how you could manage inflation if you liberalize,' and this separate idea about, you know, we are going to do reforms on: Those two things were, I think, the keys, in China. But, going back to the Bank. The question is: How many times has it actually served that function. There are lots of reports at people at the Bank produce. There's a lot of talk about, 'We have to be engaged,' and ongoing loan projects. Sure, that somebody else could finance this bridge instead of us, but if we're engaged we know the context; we'll be ready to provide advice. But I was actually very disappointed at how hard it was for anybody to come up with a case where the Bank actually provided useful advice which changed the policy direction and in a way that was beneficial.


Russ Roberts: Now, we had two guests on--about a year and a half ago, Simeon Djankov and Matt Warner, talking about the Doing Business Report that the World Bank produces. And it's an attempt to quantify the business climate: the ability to cut through red tape, the ability to open a business. Things that are basic. And, you know, one can debate whether those kind of

Paul Romer: What were the two? It was Simeon--

Russ Roberts: Yeah. Simeon, thank you. Simeon Djankov, who is the head of the Atlas Institute--he's not a World Bank person.

Paul Romer: and he's the head of what?

Russ Roberts: The Atlas network, which spawns free market think tanks outside the United States and other parts of the world.

Paul Romer: Sure. Okay.

Russ Roberts: And they are fans of that report. And I am not, in general, kind of fans of that kind of aggregate. And I am not a fan of the freedom measures, particularly. Nice friends of mine work on those; and God bless 'em. But I think those are fraught with measurement challenges and easily manipulated. And you've got entangled with some of that. So, talk about some of that.

Paul Romer: Sure. Well, okay. To put this in a larger context: I remember I was ready to leave the Bank. I had a concern about a fact that I think needed to be out in the open and be transparent. And, part of how I--you know, when they said, 'You can't resign,' I said, 'Fine. I'm going to go to a newspaper and talk about my concerns.' And then they said, 'You have to resign.' And I said, 'Well that's what I wanted to do.' But, uh, here is the fact. I was worried about the possibility that some kind of ideological, political views had distorted the numbers that were reported from, in the Doing Business Report. And, in particular, I was very worried about a pattern in Chile, where, again, it could have been just random chance. But the Doing Business ranking just moved down, you know, every year, when the party of the Center Left was in control, and moved up when the party of the Center Right was in control.

Russ Roberts: Could they have had different policies. Right?

Paul Romer: Um, well, what turns out is if you hold the measures of what's going on in Chile constant, um, all of those changes, virtually all of those go away. So, it's like saying 'Okay, Chileans are weaker this year because they can only do 20 pushups.' And you say, 'Wait a minute. Last year, Chileans are [?] because it's chin-ups instead of pushups.' And so you're like, 'Oh, well, we're using chin-ups instead of pushups this year.' And then it's, you know, like lunges the year after. They kept changing the measures that went into this doing business ranking. And if you just hold constant these measures, most of these changes went away.

Russ Roberts: Not good.

Paul Romer: So, yeah. This ought to make you worry. And I also think that, in science, the burden of proof for integrity lies with the individual. Not with me. This is not a case of, um, you know a legalistic right to do a job until, you know, a definitive proof that they are not qualified. When, you know, a certain pattern of facts emerged that raised doubts in my mind about the integrity of some of the people involved, my attitude is that, you know, my job as a scientist is to say I have concerns about the integrity of this process. Now, um, let's pull back from the fact of what happened in Chile, how is the Doing Business Report constructed. Let me talk about the way the Bank responded. They commissioned an audit--and maybe Simeon described this, because he was--

Russ Roberts: Now we didn't talk about that--this was before the--November 2017--I think it was before. Maybe not.

Paul Romer: Yeah. Yeah. That was before. Yeah. So, they commissioned "an audit," which, the report was by--I'm going to forget--Randall Moore[?]. Has a report that, Randall wrote--which is pretty reasonable. It says, 'We don't think Romer's right.' But, you know, reasonable people could differ. But, the Bank did not release the underlying work that, you know, Randall had done looking at the different indices. Like, when I left the Bank, I actually taught myself Python and put the code on the Internet and said, 'Look, here's how you could construct the Doing Business for Chile without changing the components that go into the index.' 'Here's how they actually do it.' 'Notice the difference.' That, there's virtually no change in the index for Chile if you don't change the components in the index. And so, I didn't even take a phone call from Randall when they were looking at this, because I thought they should do the report on [?]. And everything I had done on this issue was publicly visible. That's the way it's supposed to work in science. The Bank wrote up this report, which was the basis for a summary by Randall, which was the basis for PR--a Press Release that said, 'Oh, Romer is wrong; there's no problem here.' They would not release the underlying analysis. And I just think that's, kind of the--that's indicative of all of the problems with the intellectual culture at the Bank. And I think the questionable value of the stuff they call research. That, I think, you know, you just can't trust. If they aren't going to comply with the most basic practices about transparency: there is nothing sensitive about these numbers. And it was really just a question of how you weight them when you can produce these indices. And basically their criticism of me about Chile was basically that, 'Look, when you change all the components in the index, all these countries are moving up and down all the time.' So, there was nothing unusual about Chile. I still was a little worried about the directionality. But, still, there is a kind of a problem here. But, the fact that they wouldn't release the underlying data is, I think, the really telling point.


Russ Roberts: Quick thing I wanted to ask you when we were talking before about macroeconomics: Did you learn anything from the Great Recession? Do you think the profession learned anything?

Paul Romer: Well, I think the first thing is I think we've learned some humility. I think we did not anticipate the severity of the Crisis. And so, the confidence we had before the Crisis was misplaced. We should have been more, you know, more humble about how much we really knew. I think we've also acquired more data which suggest that the financial sector is very dangerous, and the cost to the world are extraordinarily high. This sounds like a joke, but I actually mean it literally: I think if you quantify the losses from the financial sector and set them alongside the losses of nuclear power, the financial sector has done much more harm with its accidents than nuclear power has done. And, why is it that we have such radically different presumptions about the role of regulation for these two industries?

Russ Roberts: I think we know the answer to that--

Paul Romer: Yeah. Nnn-yeah.

Russ Roberts: I love quoting--I know you know the Paul Pfleiderer work, which, he's talked about on EconTalk where we actually start to believe that our models describe reality, and then you can make welfare judgments. That's one thing that comes to mind. But the other is interviewing Luigi Zingales where he points out that--and he's pointed this out in print, too--that economists are the only group that we pretend doesn't respond to incentives. So, I'll just leave that there. It's crazy.

Paul Romer: For what it's worth, you know, I wrote this paper with George Akerlof about the big damage that could come from deposit insurance. We called it "Looting," and it was about the Savings and Loan Crisis.

Russ Roberts: I learned a lot from that paper, Paul

Paul Romer: Yeah. Some people have said that that gave them--they took it more seriously after the Great Crisis. And [?] kind of realized that if things are not set exactly right with your regulations, a lot of bad things can happen. I didn't feel like the Financial Crisis gave me and George the right to pick a victory lap. I don't think deposit insurance, per se, was the causal factor in this Crisis. It may have contributed. But I do think it's worth asking, more seriously: What--in a simple, you know, takeaway expression summary--What happened here? And how are we going to make sure it doesn't happen again? I feel[?] that there's work to be done to boil this down to a simple takeaway about what happened.

Russ Roberts: Well, the thing I learned from your paper is that if you think you are not going to lose your money, you're going to be a lot less careful. Which is not profound. But for some reason, it's not easy to remember. And, I do think that Too Big to Fail had something--it's not the only thing--but it had something to do with the mess we got ourselves in.

Paul Romer: Sure. Yeah. Could I just suggest one takeaway that I do think is interesting? And, I'm not the only person to say this. But, there's an insight from how we regulate our safety, which is to have the sharp separation between the entity that's the finder of fact and the other entity that actually proposes regulation. So, it's the job of the NTSB [National Traffic and Safety Board] to establish what happened in each crash. It's the job of the FAA [Federal Aviation Administration] to specify what regulatory changes might be required. I think that separation between finding a fact and then setting regulations is a really valuable one, and it's something we should replicate in many other contexts. It's related to this problem I had with the Bank. It's trying to do a job and it's trying to be an objective finder of fact about its job. And it just can't do jobs. After Dodd-Frank--part of Dodd-Frank was to set up this entity that was supposed to be a better data-gatherer so that we could do a post-mortem when the next crash happens. I think it wasn't strong enough; it's been weakened since then. But, it is a little bit of a sign of progress in the sense that we've recognized that we should have an independent entity whose job is only to look at all the facts and the data and say, 'Here's what happened,' and to leave it to the regulatory agencies to deal with. You know, like, who is to be blamed? Who cares about that? What changes should we make? The changes we should make are just different from the facts. And I hope that insight gets taken up and gets replicated in many other parts of the government.

Russ Roberts: Yeah; we need a black box for Lehman and Bear Stearns. We kind of have it. But, of course, we have a record of their transactions. We have their emails--to some extent. We have an auditor who now increasingly hangs out there, in those kind of institutions. But, it's not quite the same. But I like the point.

Paul Romer: Part of me--matters like, what are the institutional incentives and what is the organizational structure of the people who go in and, you know, read what was in the black box?

Russ Roberts: Yeah.

Paul Romer: Because, just taking what the SEC [Securities and Exchange Commission] or the Congress says is not going to work.

Russ Roberts: Yeah. I agree. It's like--no comment. I have a lot of really tasteless and macabre, dark jokes to make about that; but I'm not going to make them.


Russ Roberts: I want to close, and I want to talk about cities, because I know that's something that you think about increasingly. And I've argued--well, let me ask you a couple of things. You can just opine on them because I'm sure you've thought about them a great deal. One is the fact that people in cities seem to be more productive. And it seems to be an example of the complementary and synergy of ideas that you have spent a lot of time working on. And I've always wondered as a skeptic whether the people who go to the cities happen to be the more productive ones. Do you have thoughts on that?

Paul Romer: Yeah: you know, that's a question we can ask empirically, address empirically. And again, I think this should always be with a spirit of humility--that we're often just seeing a piece of the puzzle. But, you can look at, for example, movers--people who move from a rural area to the city--and see. So, what was the rate of growth of their wages in the rural area and what was the rate of growth in the city? And sort of holding the individual effect constant, try to see what's different in these two environments. There's a little bit of evidence that holding individual effects constant, people's wages grow faster in cities; even, the bigger the city is, the faster the growth. So that suggests something like a potential for learning that's greater in cities. I think, um, this general approach of looking at wage changes over the life cycle is something that we should be doing a lot more of, because there's a lot of learning that goes on when people are on the job. So, it's really that there are better jobs--just like there are some schools that are good and some that are bad at teaching you valuable things, there are some jobs that are better, some that are worse. And, it may be that on average cities tend to have more of the better jobs. But, we really should be figuring out what are those better jobs and how can we get more of those; and then, it's really secondary to where they tend to locate. I do think--actually, one of the empirical projects we did at the Bank was look at--these are what they call Mincer Regressions: you look at wages as a function of how many years of education you've had, and then, how long have you been on the job? And this increase in wage with time on the job, which you can interpret as a sign of learning, people move; it isn't just a contract with one firm. Wages grow faster by about a percent per year in an urban area in the developing world than in a rural area. This suggests there's big advantages to moving people from rural to urban. But then, if you compare the developing countries as a whole with the richest countries, wages grow about a percent per year faster in rich countries than in developing countries. Say, like, urban to urban, you get another point. So urban environments in the developing world are somehow not offering the same opportunity to learn and experience wage growth that somebody could get by moving to another country. And, so this is actually a little bit disappointing for me, because I thought we might be able to get more by just encouraging urbanization in the developing countries. I think we really have to deal with something that's at least as big, which is: What are these differences between countries and how can we raise the chance to learn in the developing country.


Russ Roberts: And lastly, which is related to this question, I've speculated--I don't know where I got it; maybe I thought of it; maybe I got it, I probably got it somewhere else--but, it appears to be something of a fact that mobility in the United States is falling steadily over time. Which is shocking. Most people, when I tell them that, they go, 'Oh, it can't be true.' It's akin to--well, there's a bunch of facts like that, that I like. But--I don't like them, but I find them intriguing--

Paul Romer: Interesting, yeah.

Russ Roberts: Yeah. So, one of the things I worry about is, urban areas, a handful of urban areas--not all of them--have made it difficult to build new buildings. And they've made it difficult to build denser buildings. And so, as a result, rents have risen dramatically, in certain cities. And that's got to make it hard for people to move, it strikes me. How important do you think those things are?

Paul Romer: Well, so, I haven't looked at that data about falling mobility. But it does seem that people who look at this think there's something there. So that's worrisome. And a number of people have conjectured that this reflects something about the growing relative price of rental accommodations in cities compared to rural areas or smaller cities than the big cities. And so, you can see a mechanism that operates here. If you have a minimum floor space in an apartment, and if the price of floor space goes up, people at the bottom end of the skill distribution simply will not be able to live in a city with very high income. Now, and those cities might be like the urban areas in the developing world, where they offer the best chance to learn if you come in with low skills. So, it's a problem if we're denying people access to what might be--if you think of work as like school; and I like the saying that 'Work is school'--denying people a chance to go work might be like denying them a chance to go to the best schools. So, I think it should be a very high priority to think about ways to get the entry-level price of accommodations down in places that might be the most vibrant, offer the most opportunity. And that's got to be both increases in supply, and removing restrictions on minimum apartment sizes. Both of those could lower the entry price for a starting apartment. If I can kind of push this point more broadly: I think this is a much, much bigger question and more important than most people are recognizing. Suppose that people spend a constant percentage of their income on housing. You know, like a third or something--rent or mortgage payments and so forth. That means that, you know, a 10% increase in income means a 10% increase in spending on floor space. So, then, the demand for what people are willing to pay--the demand for floor space in urban areas is going to grow at the rate of growth of urban income. Urban income will grow partly because income per capita grows. Also, in the developing world because a lot of new people will move in to urban areas. So then, if you project forward: What's the rate of growth of urban income? And ask: Is there a mechanism for a supply that could keep up with that rate of growth? It's really scary. There just isn't. And so, with business as usual, we're going to face skyrocketing increases in the price per square meter of floor space in urban areas. And with minimum apartment sizes, that means people are going to--we're going to see segregation, where people with low levels of skill are not going to be in those places. So, I think the whole world needs to start paying attention to this simple market for floor space--urban floor space. What's the demand, what's the supply for urban floor space? And then, look at these questions on building--on building heights, on, you know, expansion in the city through that lens of: What would it take to have an increase in supply that could meet the increase in demand and keep the price of floor space constant? It's a very sobering set of calculations.

Russ Roberts: A related social phenomenon that just makes it worse is another one of my favorite, little-known facts, which is that the marriage age in the United States has risen quite a bit. And, the marriage rate, especially among low-skilled, low-education people who haven't finished high school or who have only finished high school, has fallen off the table. Which means, if you are not married--if you are married and you have children, you could imagine living in the suburbs; you could imagine wanting to commute. If you are single, you tend to want to be in the city, for a whole bunch of reasons. And there's a lot more people trying to do that than before.

Paul Romer: Yep. Yep. Well, I should say that--I have this colleague, Alain Bertaud, who by the way has a great book out that's going to revolutionize, I think, urban design--

Russ Roberts: On the schedule. On the schedule for EconTalk, Paul.

Paul Romer: But Alain tells me about this little apartment that he and his wife moved into in Paris, which was tiny when they were a young couple. But yet, they loved it, when they lived there. And we have to remember, you know--think about what dormitory housing is like for our own children. Oh, sure, you've got to share bathrooms, two people, and blah, blah, blah. You know, we think it's fine when kids go to college. And we think it would be like a crime if somebody had to live in that outside of college. But, we ought to be a little bit more open-minded about these things. But, back on this point about--you know, historically the only way to have a growth in the supply of urban floor space that keeps up with the demand is a very big expansion in built urban area [?]. This is my takeaway from that. You can get some extra floor space by building taller in the center. But the bulk of the additional floor space is going to come from expansions in built area. And Paris and London grew by a factor of about 200 in built area over about 200 years. So[?] we dealt with it easily[?]; it got a lot bigger; and they are both perfectly livable, exciting cities. But, for reasons that are really hard to understand, there's a very widely shared belief that cities shouldn't expand any more. And if you don't let cities expand, that's where we're headed for a crisis. If we just let cities expand--and it doesn't have to be all of them--if one city doesn't expand, we'll find the other ones expand. But, if all of your cities stop expanding, you are really going to, I think, face a crisis.


More EconTalk Episodes