In the acknowledgements of his book he wrote: “Serious academic economists are not supposed to write books—they are supposed to write technical papers…but after fifteen years doing research on questions at the intersection of labor and urban economics, I developed an increasing desire to reach a larger audience than the one that reads my technical papers...Above all, I am grateful to Illaria. With unreasonable optimism, she has always supported the idea of trying something different, even when my unreasonable pessimism led me to postpone and delay. She has been invariably right.”
I have Illaria to thank too because if Enrico had not written the book, I would probably have never known about his work! After reading The New Geography of Jobs I realized that our areas of research and writing had some overlap. I had a number of questions, so I gave him a call. What follows is a transcript of our conversation.
JON: In The New Geography of Jobs, you write: “The sorting of highly educated Americans into some communities and less educated into others tends to magnify and exacerbate all other socioeconomic differences.” As education serves as a strong signal for intelligence, IQ, or ability, does this mean that there are cities that are literally smarter than others in terms of the concentration of brainpower? And do you think this trend will continue to increase?
ENRICO: I would say yes on both questions. There are certainly cities where the schooling achievement and the intellectual abilities of residents are higher than the average. And the trend that we have observed in the last few decades of high human capital people congregating in a handful of selected group of cities is going to continue. So if we look across cities and metropolitan areas in the next ten or twenty years, I predict the differences are going to be even larger.
Do you have any ideas for what kind of impact this will have?
I think is going to mean an increased divergence in the wage level and productivity level for those who have a high level of education but also those who don’t have a high level of education. But it will also have a broader impact on other socioeconomic outcomes—everything from health, to crime, to marital stability, and political attitudes. There will likely be an increased divergence between regions in the U.S. on these socioeconomic outcomes.
You note that “For the first time in history, the factor that is scarce is not physical capital but creativity.” How would you distinguish between the terms human capital, IQ, intelligence, and creativity? Do you think there might be overlap between the constructs these words are intended to represent?
I think there is a lot of overlap, at least in terms of the measures economists use. Human capital is usually intended to measure education and other productive skills that will make a person more or less productive in the labor force, meaning he or she is going to be producing more output and will be paid more. Certainly creativity is associated with it, and in the book I tried to use a broad set of indicators of creativity. The one measure of output of the creative process that is most easily measured at the company level is patents. For example, one could measure creativity of an area in terms of where new jobs and new occupations appear. There are studies that do that. When people invent new ways of doing things, you can actually see it in large scale datasets, where and when they appear. Or you can measure the appearance of new technologies or products. Of course, not all good ideas are patented, and not all patents are good ideas. The number of patents is an imperfect proxy that economists have been using to measure creativity.
In some of my research I’ve also used patents, because that is probably the most objective measure of creativity we have. What are your thoughts about how we might be limited by what we are able to measure in terms of creativity?
Absolutely. I think social scientists are limited by what we are able to measure regarding creativity. But there are proxies that are available. I think the difference between a good or bad social scientist is whether they can fully understand the limits of their data. Take patents. Some are pure noise. Companies might patent for legal reasons or protective reasons. For example, tenure decisions in foreign countries are often determined by the number of patents achieved. But on average I think the noise cancels out and a central tendency can be captured that measures creativity. The danger is when social scientists forget that we are using proxies and take the data too literally.
Today people don’t think about creativity in terms of patents, but in terms of musical creativity or writing, so the term is very broad.
One could measure the differences in artistic tendencies across metropolitan areas by looking at how many people self-report being a writer, or artist, or painter in the Census. I have the data and I’ve been doing research on this issue. Even this measure is not perfect. It is self-reported, and there are people who spend 95% of their working time as waiters and do something artistic at night, yet when they are asked by the Census what is their main occupation they answer “I am an artist”. Although they are imperfect, on average I think these measures do capture differences in artistic creativity across areas. For example in the Bay Area there are a lot of self-reported “artists” because there a lot of people actually engaged in those types of creative behaviors.
Are you working on some papers related to that?
Yes, I am working on a project on how the presence of a university affects the local economy. In particular what types of jobs and what types of residents live there. For example, how many artists, writers, and creative people tend to agglomerate around large universities.
Speaking of creative artists around cities, how does you work intersect with Richard Florida’s, in any way?
It does. Some of his conclusions are consistent with mine. But we fundamentally disagree on some key points, particularly on economic policy. I think part of the reason Florida was so successful was that he offered a very simple and cheap solution for localities to turn around their economies. He was providing an easy-to-achieve blueprint for economic redevelopment, centered around the idea that all a city with a stagnant economy needs to do to jumpstart its economy is to provide cultural amenities to appeal to the “creative class”. But the data tell a very different story. If you look at the history of American creative cities, what I would call the innovation hubs today, typically they first became rich and prosperous and then they became creative and cool. The evidence point to a blueprint for economic development where jobs come first, and cultural amenities follow. This is exactly the opposite of Florida’s blueprint.
You argue a company’s innovation depends not only the on the quality of individual workers, but also the surrounding ecosystem. For example, you mention that “Being around smart people makes us smarter and more innovative.” I envision a smart person being surrounded by other smart people in a broader ecosystem that facilitates innovation. What elements other than a high concentration of smart people are essential in this ecosystem?
What will surely help is an urban form that facilities the interaction of those smart people. I think one reason you hear over and over again when you talk with engineers and entrepreneurs in Silicon Valley, and ask them “Why are you here? Why did you move your business from Israel or a state like Ohio?” they want to be where the action is. When you are in the business of creating new products, ideas, or technologies, you need to be close to other people who are in your field. And I’m sure you as an academic will appreciate that. The reason academics are so obsessed with who their colleagues are is not just prestige: it is productivity. The person that we hire and sits in the office next door influences our creativity and our thinking. This extends to private sector research. For specific fields, the presence of a university is crucial. For example, in life science research, being physically very close to a university is important. Being able to talk to the academics involved in basic science, attending their seminars, sharing ideas is crucial. Thus it is not an accident that biomedical researchers tend to cluster around universities. Imagine trying to be a biotech company in the middle of a state where there is no strong research university, you would feel completely cut out of the creative process. Even if you can go online and have access to the same publications as everyone else, and see the same patents, you will still miss out on the element of the human conversation and exchange of ideas.
Even in academics, just as in business, the social interaction does appear important. For example, the trust built between collaborators. Do you think that’s true?
I think so. You and I are exchanging ideas now, but it would be very hard for us to come up with a new research idea for a top notch paper over the phone. But if we were in the same locality and met for lunch and we did this a few times and I told you what I was working on and you told me what you were working on, the chances of coming up with a joint project would be much higher. Once the idea is in place, we can execute it through Skype, phone, or email. But really, 95% of an intellectual contribution is the question you pose and the methodology. And it typically occurs face to face.
You argue that “[Silicon] Valley keeps its position as the world’s number-one innovation hub not because those who are born there are smarter than everyone else but because of its unparalleled power to attract great ideas and great talent from elsewhere.” I would argue that the key factors at play for the Valley are its ability to attract new generations of talent from outside America, within America, and even within the Valley itself. For example, Stuart Anderson discusses a reproductive type of multiplier effect: when smart people increasingly marry smart people and end up having smarter kids who then grow up in innovation hubs like the Valley. What are your thoughts?
I think your claim is consistent with my observation. The Valley does attract talent within the U.S. It’s a good point. Once two PhDs come here and marry and have kids this does strengthen the regional advantage. But overall, these people tend to be a pretty mobile bunch. If the son of the two smart PhDs who is also smart and chooses to stay in the Valley it is probably because the Valley is attractive to him. There are plenty of sons and daughters of PhDs in Ohio, for example, who end up moving to the two coasts. Most workers with a PhD who reside in Silicon Valley and San Francisco were not educated in the Bay Area. They came from somewhere else. I completely agree there is a multiplier effect, but I think for the population of highly mobile people, it ultimately is where the jobs and where the center of new ideas are rather than where they were born.
So you’re saying geography trumps a lot of things.
Yes. I think your argument about the multiplier is true. But if you look at Detroit today, the children of parents with PhDs who are still in Detroit are much lower than couples with PhDs because they have left.
You write: “In the coming decades, global competition will be about attracting innovative human capital and innovative companies…The number and strength of a country’s brain hubs will determine whether it will prosper or decline.” I would argue that it won’t necessarily be about the number and quality of brain hubs but the number and quality of the brains within those hubs or communities. Or perhaps what will be most important is identifying and developing the talent of brains that come from all kinds of communities so that they can reach those brain hubs?
I actually think we completely agree. It really does depend on how many individual smart or creative or entrepreneurial people are congregating in these cities.
You note that “America does not create enough human capital” and that there are two ways to increase it. “One way is to dramatically improve the quality of education—particularly high school math and science—in order to increase the number of Americans with college degrees. The other is to import human capital from abroad by allowing skilled immigrants to move here.” You also note that “not all innovators deserve the same level of subsidization” because some generate greater social return than others. Skilled immigrants represent the best and the brightest from other countries. What about skilled Americans or the best and brightest born here in America? Shouldn’t the focus be on developing the talent of skilled Americans rather than increasing the number of college degrees? After all, I would argue that not all college degrees are created equal and simply increasing the number who graduate from college isn’t going to develop the critical fraction of human capital that we need. For example, I would argue that the rate of return to investment in human capital is much greater for the top 5% or 1% in brains than for the middle part of the ability distribution.
I agree on where to invest and think that one college degree is not equal to another. You focus on the intelligence but I would also focus on the content of the college degree. For example, STEM college degrees have a higher return in the labor market than other majors. When I say there are two ways to increase human capital in the U.S., they are not necessarily identical. Increasing the number of highly skilled immigrants, it has the advantage in that you can add human capital into the U.S. without having to pay for it.
Other countries have developed that capital right? For example, you were born in Italy and were educated there.
I came after college to the U.S. In that case, somebody else had paid for my college degree. The same is true for Indian or Chinese scientists and engineers. When the U.S. allows an Indian engineer to come here, someone else has paid for that human capital. Please note that I am not in any way implying that the two solutions have the same effect on U.S. natives. From the point of view of innovative companies, as long as you can attract the best and brightest, they don’t really care too much. But from the point of view of the U.S. natives, the two solutions are profoundly different in consequences. I think it’s up to us as a nation to choose the solution that we prefer. There are different costs and benefits.
Your section titled “Math Races” discusses how kids who are born here in America have little interest in STEM. But the majority of scientists and engineers in many tech firms are foreign born. It is true there is a different cultural emphasis but also the selectivity of foreigners is high such that American born kids have a hard time competing with foreign born students for these jobs. I would argue that of those who have the potential be able to compete with the top students from other countries, they are simply not being developed to the extent that they should. For example, funding for gifted education at the federal level is currently zero! So to me it shouldn’t be a surprise if the talented kids in America don’t want to compete in the STEM arena because 1. there are many other interesting career options to pursue and 2. it might be intimidating to compete with the best and brightest in STEM from other countries. Your thoughts?
Let’s say you are a gifted American, maybe in the top 5%. I would say if a top 5% kid had to decide between a humanities or STEM major, if he chose STEM there would be greater labor market returns. There is likely a higher return to intellectual skills in STEM rather than the humanities, in part because a lot of the humanities or public sector jobs are not rewarding the top 5% kids.
The question is what would be the economic return to choosing STEM if the U.S. were to allow in zero foreign STEM workers. On the one hand, the supply of smart workers would diminish in the U.S if we did not allow any foreign STEM workers. US natives in the STEM field would have less competition and this would raise their salaries. At the same time, the demand for smart workers would diminish because the U.S. would lose some of its preeminence as an innovation center. There would probably be less innovation developed in the U.S., and this would lower the salaries of U.S. natives in the STEM fields.
On net, I believe that the economic return for the native STEM workers would be lower. The experience of Japan is useful in this respect. Japan has struggled in attracting foreign workers in innovation. One reason the Japanese high tech companies that were dominating in the 80’s are now struggling is likely because they can’t attract the best and brightest globally. Culturally, legally, and institutionally, it is much harder to immigrate to Japan than it is to the U.S. What would be the salary of a top 5% native U.S. kid if there were no foreign STEM workers to compete with him? I would guess it would be lower.
I guess I just worry that we are just not developing a lot of kids at the high end.
I agree with that. A lot of educational policies tend to be compensatory rather than developing those at the top.
You aptly point out: “While short-term issues can be pressing, their importance pales relative to that of long-term ones, because the latter are the ones that really affect our standard of living in profound and permanent ways. The magic of compound growth means that even tiny differences in growth rates can have enormous consequences for our future jobs and incomes. Thus, policies that can increase growth even marginally are vastly more important than any short-term fix to the economy.” I believe that we should invest in kids at all levels of ability. But the fact that federal funding for gifted education is zero sends a very strong signal that investing in our brightest minds here in America just is not a priority. Given that the rate of return to investment in human capital is significantly higher for the very smartest people, in addition to attracting the best and brightest from other countries, don’t you think we should be making it a priority to develop the best and brightest of our own country for our long term survival as a nation?
I certainly think so. The question is what are the impediments to do that? So you raised the notion that a lot of public resources are not going to developing the best and brightest. However, the economic return to human capital or IQ has been increasing in the labor market. Yet it’s almost like the U.S. has given up in investing in its own future. In some sense this is costly in the short run, and it will be much more costly in the long run.
A lot of the people in the Valley (like Mark Zuckerberg, Vivek Wadhwa, Steve Case, etc.) are arguing that we have to let these smart people in. And like you said, it’s a quick fix: we don’t have to pay for that human capital. But how long can we keep doing that? I guess for quite a while, right?
In my view, bringing more educated immigrants is probably good even for the less educated Americans. This is a zero cost reform that should have happened 10 or 20 years ago. But at the same time we need to invest in the human capital of U.S. children. This is by far the most important investment that we can make for the future of our country.
© 2013 by Jonathan Wai
You can follow me on Twitter, Facebook, or G+. For more of Finding the Next Einstein: Why Smart is Relative go here.