The Economics of Urban Growth

March 2010

Moderator: Welcome to Research Insights, an occasional podcast from the Federal Reserve Bank of Atlanta. We're talking today with Ed Glaeser, professor of economics at Harvard University. He's in Atlanta to speak at a conference here at the Bank on the subject of economics of urban growth.

Good to see you, Ed. Thank you for coming here.

Ed GlaeserEd Glaeser: Thank you for having me.

Moderator: I'd like to begin with a simple question about your work on house prices in the Southeast. Can you tell me how changes in house prices in this region have compared with the rest of the country?

Glaeser: Since the peak of the housing market in April 2006, Atlanta's housing prices, according to the Case-Shiller index, are down about 19 percent. That doesn't sound good, but it's actually 10 percent better than the 30-percent drop for the Case-Shiller 20-city average. That 20-city average, of course, contains cities like Phoenix, which have dropped 50 percent from the peak. Certainly, relative to that sort of extreme, Atlanta looks pretty stable.

The real difference across metropolitan areas in the extent of the decline seems to be explained primarily by the extent of the boom during the previous five years. There's an enormously tight correlation between the growth in housing prices between 2001 and 2006 and the decline in housing prices between 2006 and 2009. Atlanta had less of a boom, and so it had less of a bust. I think part of the reason for that is that Atlanta's relatively permissive building environment [and] it's ready availability of land insured a fairly substantial construction response to rising prices, which mitigated the price rise on the upside.

Moderator: A big part of the story of the Southeast is population growth. People over the recent decades have moved to Atlanta and the Southeast at a rapid pace, and the region has grown more rapidly than the rest of the country in terms of population and the economy. The financial crisis and the recession seem to have changed that. How would you describe the impact of the recession and financial crisis in your latest data and research in this region?

Glaeser: The starting point, as you said, is that the Atlanta metropolitan area has grown by 1.12 million people between 2000 and 2008—that's the latest Census data that I was able to find. That, of course, means that Atlanta grew more than any other metropolitan area in the country, other than Dallas. It was an extraordinary eight years, and in fact it's been an extraordinary 30 years of growth for the city. The growth of Atlanta in the longer run is fueled by a combination of economic strength, some of which is just the resurgence of the South more generally—the fact that the South in 1950 was still a fairly poor part of the country, and poorer areas tended to do better than richer areas did because capital moved to places where labor was cheaper and because the South changed from having, shall we say, difficult institutions for economic growth to having very pro-growth economic institutions.

Coupled with those forces, which led to economic resurgence in the South, Atlanta also had—as did Dallas, as did Houston, as did Phoenix—this very permissive building policy, a pro-housing-growth environment, which meant that, unlike coastal California, rising housing demand was met with enormously elastic supply, and a vast amount of new homes were brought on market. Now, in the long run, even over five- to ten-year periods, which isn't even that long, the number of people in a metropolitan area is pretty closely tied to the number of homes in a metropolitan area. In fact, the correlation over ten-year frequencies is pretty perfect. And you can bet in places like Boston or coastal California, where it's difficult to build, either because of geography or just as often because of land-use regulations that make it difficult to build, rising demand cannot be sated by housing supply. So prices go up, but new construction doesn't. It's all very basic Economics 101.

In a place like Atlanta, any sort of rise in demand is met effortlessly, seemingly, by new homes being built. So relatively good economic environment is combined with this perfectly elastic housing supply, and you have massive housing growth and, since there are people living in those homes, massive population growth.

Now, of course, the issue is that the years prior to 2006, Atlanta built an enormously large number of homes, and Atlanta, more than any other metropolitan area in this country, has had an incredible cutback in the number of homes, at least as measured by the number of permits. Atlanta went from having 74,000 permits per year in 2005 to less than 7,000 in 2009, according to the preliminary figures. That's the largest drop of any metropolitan area in the country, and there are good reasons for that. Given that you built so much during the boom, you've got this enormous hangover, as the country does. On top of that, at the country-wide level, we haven't been forming new households at the normal rate at all. As opposed to a normal rate of 1.3 million, we're forming less than 600,000, accordingly to the March CPS [Current Population Survey] numbers.

I think it's going to be a tough number of years for the housing market and for the construction market in Atlanta. I think it would be odd to believe otherwise. But that shouldn't make us think that Atlanta doesn't have a solid longer-term future. Its combination of scale and skill is a powerful one. As a result, I bet that Atlanta is going to weather this, but it may well be a tough couple of years ahead.

Moderator: What do you think is going to replace the housing industry for employment and business growth here in the Southeast, especially Georgia, where carpet manufacturing and housing was so important to drive employment growth? What's going to replace that?

Glaeser: One of the great problems of cities is that the genius of cities is inherently unpredictable. The magical thing that happens when smart people connect in dense, large metropolitan areas like Atlanta is that they come up with new ideas, they come up with new industries, they come up with new occupations, they figure out new ways to employ people. As a result, it's very hard to figure out where the new jobs will come from. I would bet they're more likely to show up in information-intensive industries. I'd bet they're more likely to show up in industries that are connected with the larger world because of the ability of cities to be gateways across oceans and continents.

There is, however, a big problem, which is that all this talk about information-intensive industries doesn't suggest a lot of jobs for construction workers. This is, of course, a problem that's not just Atlanta's problem—it's America's problem. We have a problem that the building boom employed a lot of people with relatively modest levels of, shall we say, traditional human capital. They certainly had lots of skills in specific dimensions. The downturn has created a tremendous unemployment gap between people with college degrees, where the unemployment rate hovers around 5 percent, and people who are high school dropouts, where the unemployment rate hovers around 15 percent. There's an incredible gulf that's associated with those skill gaps. And I don't think it's very easy to imagine what, in the long run, are going to be the big new employers of less-skilled workers. I think it's likely to continue to be the service sector. But, certainly at the national level, it's hard not to think that the best long-run solution is to improve our investment in this nation's human capital.

Moderator: You've done some work on enterprising businesses. In your research, have you discovered any secret to creating a critical mass of enterprising businesses as we see in Silicon Valley? If so, what's the trick to making that happen?

Glaeser: You know, there's no pixie dust that you can just sprinkle over a location and make it work. It is certainly true—we just think about Silicon Valley, about what happened, and above all, it was Stanford. The roots of Silicon Valley go back to Federal Telegraph, which was a wireless company before World War I. Federal Telegraph was founded originally by a young genius who was working on these technologies in San Francisco and then got put in the hands of one of the smartest, young electrical engineering students at Stanford, and that's where it's comparative advantage came [from]. And I think, throughout, if you look at Silicon Valley, it's this combination of education, of being smart about science, with this thing that was coded into the university that Leland Stanford formed, which is this very practical side. Stanford was very clear that he didn't want a place that felt like Harvard. He wanted a place that felt like it was engaged with the real world.

The variable that best predicts urban growth over the twentieth century is January temperature, but after January temperature, the share of the population with college degrees is a pretty darned good predictor of which places are getting both more populated and wealthier. On average, as the share of the population with college degrees in an area increases by 10 percent, the wages of other workers in that area increase by 8 percent, holding their skills constant. If there's one sort of magic ingredient for human capital success, for entrepreneurial success in these high-tech areas, it's having robust universities that are engaged with the real world.

Beyond that, there are other advantages, which Silicon Valley had. It happens to be an enormously pleasant place to live, and, as a result, very successful, very wealthy people actually want to live there, which is an added plus. And, just as a sort of third thing, which I think has probably proved to be—there's still debate on this, but California has only a limited amount of enforcement of non-compete clauses, which makes it very easy for people to go and spin off and create new firms.

Moderator: One of the things I wanted to ask you before you go is about fiscal problems. Obviously, at the state and local level, there are an abundance of problems with budgets and fiscal issues. What does that suggest for urban growth?

Glaeser: Well, there's surprisingly little connection between most measures of government spending and urban success over the past 40 years. Somewhat surprisingly, if you go back to 1960, a higher level of debt-per-capita at the city level actually positively predicted subsequent growth. Now, the reason for that is actually not all that surprising, that the places that actually were growing for completely unrelated reasons had to borrow to build infrastructure to accommodate that growth. But it does suggest at the very least that debt isn't automatically a terrible thing at the local level if that debt is being used for something that's productive.

That being said, I think that there's no question that good government services are important. The evidence on tax levels is mixed because it's not that when money is being taxed and spent on things that are valued, like good education, it's not a particularly bad thing to have local taxes. If the money is being taxed and wasted, it clearly must be bad, but we have such poor measures of the quality of local government services along many dimensions that it's very hard to actually say for certain with any of these relationships. So anything that I say has to be taken with a fair degree of skepticism.

I think the states and localities right now face this awful quandary. They're sitting at this position where the demand for public services is greater than ever, and yet they often face very strong arrangements that stop them from borrowing to actually meet those services, and their revenues have plummeted. Neither option is good. Cutting services, particularly along areas like education or things that impact the quality of life, can be enormously detrimental. On the other hand, raising taxes can be enormously costly as well, and I don't think there are any easy answers.

I do think the federal government is playing a helpful role in some ways by easing the financial constraints facing states during this particular time period. But, if I were sitting in a mayor's office or in a governor's office, the long-run strength of the state depends upon its retaining its human capital, and the critical thing is to not cut the services that are going to push out the skilled people, and to not particularly view them as a golden goose that can be taxed with impunity. And, unfortunately, that really does limit what you can do.

The other piece of advice that I would give is this is a great time to rethink those regulations that may be imposing costs that don't need to be there. That actually, given that you can't raise taxes, you don't want to cut services, the one place where you can actually act right now is on the regulatory line, and I think that's the best chance to actually do something positive during this downturn at the state and local level.

Moderator: Very good. Well, that's an interesting analysis. Thanks, Ed. I appreciate your time.

Again, we've been speaking with Ed Glaeser of Harvard University. This concludes our Research Insights podcast on the economics of urban growth. Thanks for listening and please return for more podcasts.