What really drives regional economic development?

Every so often the question of what really drives economic development comes up. Sometimes it’s because someone raises the question about causality in my own 3Ts theory of economic development. Other times, it’s because they want to know what kind of factors matter in practice—tax cuts, company attraction, talent attraction, amenities or tolerance?

I have written on this before. Here is a longish essay published several years ago in Next American City and here is my response to Ed Glaeser's thoughtful review of Rise of the Creative Class. 

But in the intervening period, some great new research has surfaced, and my team and I are always examining new data and conducting additional analysis. So here's a brief summary of where I think the field is at, at the moment.

First off, it’s important to say that there is now wide consensus among economists and other students of economic development about the primary factors that drive economic development. Long ago, Robert Solow found that technology is critically important. Today, drawing primarily upon the work of Robert Lucas, who in turn drew upon Jane Jacobs, the primary factor is seen to be human capital or what I simply refer to as talent. "As Lucas writes, “much of life is ‘creative’ in much the same way that is ‘art’ and ‘science.’ …To an outsider it even looks the same. A collection of people doing pretty much the same thing, each emphasizing his own originality and uniqueness.”

Drawing on Jacobs’ insights, Lucas, declared the multiplier effects that stem from talent clustering to be the primary determinant of growth and he dubbed this multiplier effect “human capital externalities.”  He also said Jane Jacobs deserved a Nobel Prize for identifying it. Places that bring together diverse talent accelerate the local rate of economic development. When large numbers of entrepreneurs, financiers, engineers, designers, and other smart, creative people are constantly bumping into one another inside and outside of work, business ideas are more quickly formed, sharpened, executed, and—if successful—expanded. Lucas summed it up simply: “If we postulate only the usual list of economic forces, cities should fly apart.” This is because land, as Lucas reminds us, “is always far cheaper outside cities than inside.” With a penchant for common sense that seems to distinguish the greatest thinkers he sums it up with the question: “What can people be paying Manhattan or downtown Chicago rents for, if not to be around other people?” Ed Glaeser has a masterful review of the relevant literature, including the work of Paul Romer, here.

So what about writers like Joel Kotkin and Steven Malanga who argue that there is little correlation between regional development and places with high scores on my creativity index or high levels of human capital? Well, for one thing, they do not seem to be aware of this entire line of thinking on  economic development. More to the point, they fall victim to simple but utterly fatal mistake. They confuse short term growth with long-run economic development. This is no small error, mind you.

The standard measure of economic development is simply a measure of the level of overall development, for example GDP, income, or wages, controlling for the size of the population.  But when you use short-run changes in population or employment, not surprisingly you get very different results.  One can easily see how neither measure necessarily matches up to the level of wealth, output or income. Employment growth can and does fluctuate like wild across regions. Las Vegas is up, Silicon Valley down, and what not. But it does not necessarily impact the overall level of output or income in the medium to long run. Incomes and housing values, not to mention levels of innovation, remain much, much higher in Silicon Valley.  And, it's pretty clear that different places grow different kinds of jobs—some grow lower wage service jobs, while others grow higher-wage jobs.  Not all jobs are equal. And short-run job growth does not equal long run economic development.

My own work is squarely in line with the human capital theory of economic development. I agree that human capital is the driving force in economic development, but I seek to amend or supplement the theory in two ways.

First and foremost, I offer an alternative measure of human capital or talent, which I believe has advantages both conceptually and practically. Most studies of human capital measure it as educational attainment, that is typically by the percentage of people in a region or nation with a bachelor's degree or more.  I came to believe and still do, that while this is a robust measure, it is limited for two reasons.

• One, the educational attainment measure leaves out people who have been incredibly important to the economy, but who for one reason or another did not go to or finish college. Names that come quickly to mind are Bill Gates. Steve Jobs and Michael Dell, among countless others. My measure of creative occupations counts them all.
• Two, the educational attainment measure is quite broad and thus does not allow for nations or regions to identify, quantify or build strategy around specific types of human capital or talent. We all recognize for example that  Nashville is the center for country music talent, Hollywood for film, Silicon Valley for technology. And it is clear that nations and regions are coming more and more to specialize in particular kinds of economic activity, so my occupation based measure allows us to get at that.

The short of it is, for these two primary reasons, I turned to an occupation-based measure for looking at the economy broadly and the capabilities of its regions.  It has the special advantage of allowing for regions to understand their underlying capabilities and plan accordingly for development and to address other needs. Other leading urban and regional economists, such as Ann Markusen, have begun to develop their own measures of occupations and occupational based development.  And there is an emerging consensus about their importance. They are more and more viewed as the next wave of economic development analysis beyond the well-established industry cluster models used by Michael Porter and others.  Some have even interacted these two approaches with some success. With Kevin Stolarick and Steven Pedigo, we are developing tools for regional occupational cluster analysis that can be effectively used by regions to determine their strengths and weaknesses and to shape development strategy.

There have been several tests of the occupational versus educational approaches to human capital and growth. By far, the most thorough was by the Dutch economists, Gerard Marlet and Clemens van Woerkens in their studies of the Netherlands. They write: "we find that Florida's creative class is a better predictor of city growth than traditional education standards.Therefore, we conclude that Florida's major contribution is his successful attempt to create a population category that is a  better indicator for levels of human capital than average education levels or amounts of highly educated people. The point is, as Florida stated, not which or how much education people can boast of, but what they do in working life."

The second issue is somewhat larger and revolves around the question of what attracts talent or human capital in the first place. Since we know that talent drives economic growth, and we also know that talent is spread unevenly, it is important to understand the factors that account for this divergence. 

While most economists agree that the levels are different, most continue to conceptualize human capital as a "stock" or "endowment" of a given place--either you have it or you don't.  But the reality is that human capital is a flow.  The key question thus becomes: What factors shape that flow and determine the divergent levels of human capital across regions? That is the second question that lies at the heart of my work.

As far as I can determine, there are now three different answers to that question, though as we will see in a minute I have come to believe that they are actually complimentary. I refer to these three factors as regional institutional and cultural factors. As I see it, they shape economic development directly and indirectly by effecting the level of talent or human capital.

The first is the theory that amenities drive the attraction of human capital. This theory, which is also partly reflected in my own work, has now been proven in independent studies by Glaeser, Jesse Shapiro an economist and Terry Clark, a sociologist at the University of Chicago, among others. Shapiro's detailed research finds that "roughly 60 percent of the employment growth effect of college graduates is due to enhanced productivity growth, the rest being caused by growth in quality of life. This finding contrasts with the common argument that human capital generates employment growth in urban areas solely through changes in productivity."

The second, clearly reflected in my own work and that of many others, is that the university plays a key role in producing and concentrating human capital.

Still a core question remains: let's call it for our purposes the Pittsburgh paradox: That is, a region like Pittsburgh can have great universities and substantial amenities (symphony, ballet, opera, museums, professional sports, golf courses and the like) and still experience a significant outflow of talent. In addition to the two above, amenities and university, I offer a third.

Reflecting “Occum's razor” and basic economic logic, I posit that there can be an even simpler and more basic explanation. If firms and markets benefit from low barriers to entry, why wouldn't labor markets. Thus I contend key a driving factor in the divergence or flow of human capital is simply the openness of a given place to that human capital. The more open a place is, the more it will be able to capture the talents of its own people and to attract those from elsewhere.

Let's call this the San Francisco advantage--reflecting the fact that the San Francisco Bay area, and in particular it's Silicon Valley is the leading center of technological innovation in the world, but has long been a center for openness to different ethnic and racial groups, alternative lifestyles and artistic and cultural innovation. My argument is that a place that is open across the board, will also likely be more open to technology-oriented entrepreneurs. There are two reasons for that. One, such places possess some underlying characteristics that allow individuals, including entrepreneurs, to readily mobilize resources. Two, they are oriented toward personal self-expression and openness to experience which psychological studies show is a key characteristic of entrepreneurial behavior.  (Let me also note that the point skeptics bring up that Silicon Valley is not really San Francisco is a non-starter. According to the US Census these two areas, along with Oakland, comprise an integrated "consolidated metropolitan area" with a shared labor market and commuting shed).

My research with Gary Gates, reported in Rise of the Creative Class, suggests that openness works differently across regions of different sizes. That is, for smaller and medium sized regions immigration seems to play the most significant role. It is only for larger regions that the gay index and bohemian index really kick in to full effect. 

Ultimately, my theory and my work say that all three regional institutional and cultural factors are at play--openness, university and amenity.  They key is that each plays a somewhat different role in effecting the level and flow of human capital and effecting economic development directly as well as indirectly.

To date there are no published studies that have truly parsed this issue. Right now I am working with Charlotta Mellander to develop a general model which allows us to do so. Our preliminary results for Sweden are quite encouraging, and we should have a paper shortly. The next step is to replicate this analysis for the United States Our model does three things.

• First, it explicitly tests for the differential effects of education versus occupation measures of human capital on development.
• Second, it includes technology, the Solow factor, in the model alongside human capital or talent enabling us to parse the differential effects of each on economic development.
• Third, it tests for the effects of  regional cultural and institutional factors--amenity, university, and openness on human capital levels and in turn on economic development.

Briefly put, the findings suggest that the occupation-based measure is important and outperforms the education based measure in many ways.  We find that talent and technology also interact in important ways that effect economic development. And we find that all three regional factors—amenity, university, and openness—play a role in human capital and in development. In effect, each plays a different role in attracting talent and on economic development more generally.

So in sum, what can we say about the drivers of regional economic development? Despite hemming and hawing mainly among non-academics, it is well-established that human capital or talent drives economic development. My own work is squarely in line with that view. I supplement that view by suggesting an alternative measure of human capital based on occupations, and detailed studies confirm that it is equally or more efficacious. This occupation-based measure has the additional advantage of informing useful practical tools which regions can use to analyze their economies. Occupational cluster analysis is a tool that both compliments industrial cluster analysis and goes beyond it in important ways. In addition, my theory sees talent as a flow, rather than as a stock, and seeks to identify the factors that effect its divergence and concentration. This remains an open question, but there is emerging sense across the literature that three factors matter: university, amenity and openness.  My most recent work suggests that all three are important, but that each plays a different role in the level of talent and in economic development.

I cannot tell you how much I am delighted to be working in this field and how much I look forward to reporting new developments as they appear.

Comments

[Barkley Rosser]

Not too long ago there was a debate over your arguments on marginal revolution. It was argued by Robin Hanson that you overemphasize the role of bohemians and gays in the development of high tech industries. Thus, while San Francisco is the gay capital of the US, Silicon Valley and the software firms there are much more likely to be full of pretty straight and boring engineering types who are not bohemian at all and not more likely to be gay than anybody else. In the Washington area, the high tech sector is out near Dulles Airport or along Route 270 and not off Dupont Circle, and so forth.

Any response?

[[email protected]]

Barkely--Thank you for your comment. Yes, in fact, I have answered that many times--in the Cato Unbound discussion that you mention and actually in my post above. Both cases you mention are actually large integrated regions. As I mentioned in my post, the San Francisco Bay area is defined by the Census as a Consolidated Metropolitan Statistical area and actually the entire greater Washington area is a Primary Metropolitan Statistical Area. That means they are integrated population corridors, labor markets and commuting sheds. For instance I know LOT's and LOT's of people who commute from NoVa to downtown DC and also lot's, including gay individuals who commute from inner DC neighborhoods like DuPont or Logan Circle to NoVa.

It is also clear from historical research by John Markoff and others that the counter-culture/ alternative lifestyle character of the Bay Area was not just limited to San Francisco but was already a key element of Palo Alto and the Silicon Valley by the early 1960s. The Grateful Dead, for example, were founded in Palo Alto and Ken Kesey and other writers were based there.

Finally, a detailed look at the data shows that while perhaps the level of the bohemian or gay index is slightly higher in San Franciso proper than in Silicon Valley or in inner-DC than in NoVA, in all of those places there are much higher than average concentrations of all the above.

So the attempt to separate out these communities from their broader regions simply doesn't hold water.

[Barkley Rosser]

Thanks. In that debate I gave something like what you just said as an answer to Robin Hanson, to which he declared that he had worked in a software firm in Silicon Valley for nine years, and everybody there was a non-Bohemian, married, engineering nerd.

[[email protected]]

Barkley-- double thank you! Rich

[Glen Magder]

I would also argue that a lot of these "straight and boring engineering types" are within the creative class itself. Perhaps not a visible as the "bohemian" type but certainly a part of the creative class. A number of those whom I know that fit that stereotype, found commingling with the regular class difficult during youth. As such, acceptance (or not being ostracized) was easier outside of the non creative class, which was more accepting.

[Trevor David]

Richard I was wondering what your thoughts are on high concentrations of Black people as part of the propellant that gives rise to the Creative Class, if I am not mistaken Black people/youth are the ones who are the main creators of American culture. Look at Jazz, Rock & Roll, Disco, Hip Hop culture which is the most powerful youth driven culture on the planet. It all started in Black Urban centres. I note you talk about Gay and Boheminian Culture as main drivers of "cultural creativity" and the rise of the Creative Class, but nothing is said of Black Urban Culture probably the single most important influencer of Western Culture there is. Black youth are the ones who set the trends in music, fashion, dance and are usually the "first adopters" of whats "hip" in terms of technology gadgets.
I was wondering what your thoughts are about the above.

As a Black social entrepreneur up here in the Great White North culture is usually seen and defined thru the eyes and or prism of White Canada, all of the cultural institutions that are promoted or recieve government and Foundation grants are Euro-centric institutions while Black cultural institutions get the back of their hand. Maybe you can help us in educating the white elites up here that even the Rock Roll that they dance to came from Black America.

By the way welcome to Toronto, we can use some of your ideas about "openess" up here in the "Great White North".

[Scott Davis]

Richard,

Perhaps Glaeser's work on the mean average January temperature explains your Pittsburgh Paradox.

http://www.hks.harvard.edu/taubmancenter/pdfs/sunbelt.pdf

I have seen your comments on some of his other work but not this.

Bought Who is Your City at the airport today -- excellent thus far.

Thanks