Are Agglomeration Externalities Antithetical to Conservative Ideology?

A fascinating blog post by Felix Salmon yesterday got me thinking about the intersection of economics and politics in a new way.

[T]wo of the biggest and most daunting long-term problems facing the US economy are (1) the fact that Americans aren’t as well educated as their counterparts elsewhere in the world; and (2) the fast-growing obesity epidemic.

Both of these problems are caused, in large part, by America’s very high levels of child poverty.

So if you fix the child-poverty problem, you’ve made a serious dent in both the education problem and the obesity problem.

What’s more, the child-poverty problem really is one of those problems which can be fixed quite easily just by throwing money at it. Give enough money to children in poverty, and they’re not poor any more. Problem solved — at least to a first approximation.

***

Is there a conservative way of addressing such issues? I don’t think there is — I think that conservatives will simply say that questions of education and nutrition are a matter of individual choice, and that the government should not concern itself with such things. But if we continue down that road, I fear that the unemployable underclass will only continue to grow. And that anger at the powers that be — whether it comes from the Tea Party or from Occupy Wall Street — will only continue to grow along with it.

What Felix is talking about here is essentially an externality problem, what we in economics call agglomeration externalities. The easiest way to explain is to talk about what is perhaps the most common one: education. The idea is that the more educated people there are in a society the more society benefits as a whole. So an individual's education not only benefits themselves but also has an added benefit to the society as a whole.  We call this social returns to education.

Health problems are another potential example: sick people create a drag on both productivity as well as on the health care system which, in turn, creates a drag on the economy.

Assuming these types of externalities exist, they create a role for government to help promote the activities with positive externalities and restrict the activities with negative externalities.  And indeed governments do: we build bike lanes and tax cigarettes for example.

Whether such agglomeration externalities exist is a question that becomes political for these very reasons.  I recently had a editor of a top journal reject a paper for assuming social returns to education.  I had defended the assumption with what I thought was a pretty convincing set of empirical studies, but his conclusion was that the verdict had been decided the other way. I am not sure his judgement was a political one, but perhaps - he was from a certain school that has become synonymous with conservative economics.

But what Salmon makes me wonder is if the very idea of agglomeration externalities is somehow incongruous with the whole individualist philosophy of some conservatives.

I don't really have an answer here, just a question: suppose that for some activity agglomeration externalities are real, positive and large for economic growth - does this mean necessarily that government should be involved with actively promoting the activity?  And if so, does this undermine the idea of individual self-determination in favor of collective action?

Economist's Notebook: Agglomeration Externalities and Portland

Ed Glaeser has another in his continuing series of excellent posts on agglomeration externalities in The New York Times' Economix Blog.  Here is an excerpt:

Jed Kolko’s essay in the “Agglomeration Economics” volume reminds us that the delivery of services often involves face-to-face contact, whether in a barber shop or a law office. While the costs of moving goods has declined steadily over time, the cost of moving people has not because time has gotten more valuable and you need time to travel. This explains why business services, which generally rely on face-to-face contact, are disproportionately prone to locate in cities. Mr. Kolko finds that the most skilled service industries are most likely to locate near each other and to locate in dense urban areas.

Mr. Kolko’s paper also shows that customers and suppliers that rely most on information technology are also most likely to locate near one another.

Perhaps unexpectedly, information technology seems to be associated with a greater desire for geographic proximity. After all, Silicon Valley — the home of information technology — is the most famous example of a geographic cluster in the world today. One reason new technologies make face-to-face contact and proximity more valuable is that technological change increases the returns to being smart, and human beings get smart by hanging around other smart people.

Cities, like New York, thrive as places of pleasure as well as work. Urban proximity enables people to connect with one another by creating, among other things, urban marriage markets. Large urban scale can also cover the fixed costs of urban amenities like museums and live music venues and clubs.

The Joel Waldfogel essay in “Agglomeration Economics” emphasizes the benefit that comes from being around people who share your own tastes.

If you live in a county of vegetarians, then your local supermarket is likely to be short on beef. If you live surrounded by voracious (and prosperous) carnivores, then you are much more likely to find U.S.D.A. prime at the butcher counter. The shopping options in Manhattan are tremendous by world standards, because of the vast number of customers on the island, and that is part of New York’s appeal.

Cities are a natural topic for economists because urban density can magnify markets, like the markets for workers and balsamic vinegar and spouses. That urban strength also explains why cities have proved to be so robust, and so fun.

This has direct application to Portland. Portland has gained a reputation for a number of things: progressive politics, urban planning, local food, beer, art and design and outdoor amenities. The fact that prople decide to move here because they are attracted by these things makes these things more prevalent and cheaper. I wrote about a version of this in a beeronomics post on economies of scale: more and more local breweries make it cheaper to brew for all of the breweries in town. [A more specific version of this is the organic beer movement - it used to be very hard to find organic ingredients, especially malts, and choices were limited and often available only from Germany.  But now with Roots, HUB and lots of other breweries brewing at least one organic beer, local suppliers are beginning to provide organic malts in increasing varieties]

Portland also has a tech sector, but it is mostly the lower end production stuff that is cheaper here and doesn't need to be so close to the 'brains.'  This is, in my mind, the key to the continued success of Portland: having enough of the really smart and creative tech types locate here and create new businesses.  So far it hasn't happened in sufficient scale.  It is a hard thing to engineer and trying to specifically target the types, the specific industry, etc. is a mug's game.  Which is why I am generally both optimistic about the future of the city and a little less skeptical than some of otherwise hard-to-defend public investments.  They are hard to defend because the data that establish a causal link between, say, an MLS team or a streetcar and general economic development are simply unobtainable.  But absence of evidence in support is not evidence for the negative and it is hard to believe that these things about Portland that people like so much don't have a long-term economic impact.  It is slow and amorphous, but it is real.  I think Portland is, and will continue to be, a jewel in the American urban landscape and this will, through time, yield real and substantial economic results.

Agglomeration externalities also provide an economic explanation for the observed sorting of people by ideologies as argued in the book "The Big Sort."  Perhaps it is to find more and cheaper things that we value that really determines our location choices and the fact that these are correlated with ideology is accidental.  So we are not seeking the comfort of the people we agree with, but seeking the material things we like and prefer them to be accessible, plentiful and cheap.  The fact that they are things like local organic food, green energy solutions, mass transit, etc., correlate to our political and policy preferences but these non-material political preferences are not what made us self-sort.  Perhaps.

Economic Gardening?

An Op-Ed in today's Oregonian (not yet on-line so I have to go by memory here) by Brent Barton, Bob Jenson and Jefferson Smith promotes the idea of "economic gardening." This is a wonderfully evocative term, but the piece offers few specifics. Personally, this kind of public-policy mumbo-jumbo drives me nuts because it sounds so right yet means so little. In fact this is precisely what caused be to choose economics as a career: other disciplines frustrated me with the lack of specifics while economics excited me with its precision.

As much as I can figure out, the idea is that we make it as easy as possible for businesses to start up in our communities, and this is anything but a new idea. But if it were that easy we would not need to keep coming up with more and more wonderful terms to extoll its virtues. The problem in Oregon is not the bureaucratic environment - in most measures of business friendliness Oregon scores near the top - the problem is in the fundamentals, like education, where Oregon scores near the bottom.

Littleton, Colorado is an interesting example that was chosen to illustrate the 'economic gardening' idea.  Littleton exists in a sea of highly educated people, other high tech companies and institutions of higher education.  The Denver-Boulder metro area has a pretty good human capital infrastructure, in other words, which makes it easier for a new company to succeed.  I talked about this in my recent posts on agglomeration externalities.

The problem with agglomeration externalities is that they are not easy for policy makers to create by 'choosing' industries or businesses to promote.  So what I would like to see are policy makers, like the Op-Ed's authors, focus on the fundamental things that matter and stop trying to engineer economic prosperity.

As an aside, my attraction to economics is echoed by Paul Krugman in an interesting biographic piece in The New Yorker:

...Krugman went to Yale, in 1970, intending to study history, but he felt that history was too much about what and not enough about why, so he ended up in economics. Economics, he found, examined the same infinitely complicated social reality that history did but, instead of elucidating its complexity, looked for patterns and rules that made the complexity seem simple. Why did some societies have serfs or slaves and others not? You could talk about culture and national character and climate and changing mores and heroes and revolts and the history of agriculture and the Romans and the Christians and the Middle Ages and all the rest of it; or, like Krugman’s economics teacher Evsey Domar, you could argue that if peasants are barely surviving there’s no point in enslaving them, because they have nothing to give you, but if good new land becomes available it makes sense to enslave them, because you can skim off the difference between their output and what it takes to keep them alive. Suddenly, a simple story made sense of a huge and baffling swath of reality, and Krugman found that enormously satisfying.

The article also touched on his contribution to economic geography:

Later on, Krugman became interested in economic geography, in the related question of why there were regional specialties—why, in the United States, for instance, were cars produced in Detroit, carpets in Dalton, Georgia, jewelry in Providence, and chips in Silicon Valley? Again, the answer turned out to be history and accident. Once an industry started up in one place, for whatever reason (the carpet industry in Dalton appears to have its origin in a local teen-ager who in 1895 made a tufted bedspread as a wedding present), local workers became trained in its methods, skilled workers from elsewhere moved there, and related businesses sprang up close by. Then, as more skilled labor became available, the industry could grow and benefit from economies of scale. Soon, as long as it didn’t cost too much to transport the industry’s products, the advantages of the place would be such that it would be impractical for someone to open up a similar business anywhere else. Many economists found the idea that economic geography could be so arbitrary “deeply disturbing and troubling,” Krugman wrote, but he found it exciting.

This is essentially my point - it is impossible to predict which industries will thrive, so let's not try and create a local economic focal point, let's get the fundamentals right so one will spring up.

Agglomeration Externalities Redux

Enrico Moretti of the University of California at Berkeley has yet another fascinating empirical paper on agglomeration externalities which relates directly to my earlier piece. Here is the abstract:

We quantify agglomeration spillovers by estimating the impact of the opening of a large manufacturing plant on the total factor productivity (TFP) of incumbent plants in the same county. We use the location rankings of profit-maximizing firms to compare incumbent plants in the county where the new plant ultimately chose to locate (the “winning county”), with incumbent plants in the runner-up county (the “losing county”). Incumbent plants in winning and losing counties have economically and statistically similar trends in TFP in the 7 years before the new plant opening. Five years after the new plant opening, TFP of incumbent plants in winning counties is 12% higher than TFP of incumbent plants in losing counties. Consistent with some theories of agglomeration economies, this effect is larger for incumbent plants that share similar labor and technology pools with the new plant. We also find evidence of a relative increase in skill-adjusted labor costs in winning counties, indicating that the ultimate effect on profits is smaller than the direct increase in productivity.

This has to do with manufacturing specifically, but it is not hard to imagine the same is true for high-tech industries.

Economist's Notebook: Agglomeration Externalities

Agglomeration Externalities is the technical term of art economists use to describe the benefits non-participants get from other economic activities taking place nearby.  I have talked about them a lot in this blog, especially in conjunction with my advocacy for increased support for public education.  In the education example, an educated population not only increases individual productivity but also gives an extra boost to group productivity.  The empirical evidence has shown that, for example, cities with higher proportions of college graduates have higher productivity firms and pay higher real wages to those without a college education.  In other words, public investments in things like higher education benefit more than just the students themselves, but benefit the community as a whole.

Given the sorry state of Oregon's public investment in education then, it is interesting to read Mike Rogoway's article in The Oregonian on the flight of some high-tech start-ups from Portland.  Though not specifically tied to education, the theme is the lack of a critical mass of human capital - an example of agglomeration externalities.  Firms benefit from having that human capital surrounding them, not just from those they hire.  This contradicts those that think we don't need to invest in education because we can just import high productivity people.  But firms, especially small start-ups, find this difficult to do.  Young, talented and productive people in this fluid job market look for a community not just a single firm, as the likelihood that they will stay in one form for an extended period is small.  Add in things like the proximity of venture capital and the problem becomes even more acute.

The question then is what do you do?  Well, trying to manufacture a high-tech or a bio-tech sector is a fool's errand when you don't have the fundamentals in place, and one could argue (and I do) that the current paucity of high-tech human capital in Portland is a result of decades of underinvestment in education.  Unfortunately, those in government rarely think beyond dots that they can connect in four years or less, and so the emphasis is on new bio-tech districts, business energy tax credits and the like.  And voters, who gleefully pass measures like 5, 50 and the kicker need to accept that with limited government size can also come limited economic vitality. Sure government can be too big and can cause inefficiency, but we are no where near size that in Oregon.  Economic research has shown that government investments in human capital are vital for future economic prosperity and in Oregon we are no where near that size of investment either.

One final note, I think it is really time to stop talking about K-12 education and higher education in separate discussions.  In today's world the expectation should be that everyone should have access and be expected to complete a college degree.  Perhaps it is time to start talking about K-16.