Sunday, September 23rd, 2012
The New Geography of Jobs
by Enrico Moretti
Houghton Mifflin Harcourt (2012)
If there’s one current book I’d recommend to leaders in American cities today, it’s Enrico Moretti’s The New Geography of Jobs. It’s not without flaws. However, this book lays out a readable and compelling vision of the innovation economy in the 21st century and what it means to America’s cities. And it should accomplish what is desperately needed in many places, namely lighting a fire under people’s asses to realize that they and their city are likely in a fight for their very economic survival.
Moretti lays out what has happened in America over the past 30-40 years. It’s a trend he labels the “Great Divergence” that is illustrated by the quote at the top of this piece. There has been an increasing division among American communities between the haves and the have-nots. As radical productivity enhancements and global competition reduced employment and wages in traditional sectors like manufacturing, new knowledge based industries took their place. However, these knowledge industries require, of course, highly educated workers with specialized skills. This leads to clustering of workers and jobs in select hubs, leaving many communities out the cold.
Moretti illustrates this with a few examples. His first was a comparison of Menlo Park with Visalia, California. Forty years ago these communities were similar enough that a middle class, educated professional might well choose to make the move from Menlo Park to Visalia. Fast forward and these communities could not be more different. Menlo Park is part of Silicon Valley, while Visalia has the second lowest percentage of college educated workers in the country, with one of the lowest average wages to boot, along with high crime, poor K-12 schools, terrible pollution, etc.
On a larger scale he contrasts Albuquerque to Seattle. Microsoft had been founded in Albuquerque, but moved to Seattle in 1979. This wasn’t because Seattle was a garden spot. Far from it. This was the era of the infamous billboard that said, “Will the last one leaving Seattle please turn out the lights?” The Economist had labeled it the “city of despair.” People forget that places like Seattle also suffered greatly in the Rust Belt era.
Forty years ago, Albuquerque and Seattle were fairly similar. In 1970 Seattle had a five percent edge in college educated workers (adjusted for population) and slightly higher salaries – both likely because of Boeing – but the gap was not huge. Today, Seattle has 45 percent more college educated workers (adjusted for population), and the income gap has widened considerably as well. Though it was not obvious at the time, Microsoft’s relocation – which was simply because Gates and Allen wanted to move back to their hometown – was the catalyst that transformed Seattle’s economy into a tech hub. Now companies like Amazon set up shop there even though the founder doesn’t have a personal connection.
Not only did Seattle, Menlo Park, and other places that successfully made the transition to the innovation economy win by getting the new high paying innovation jobs directly. Those jobs also caused huge spillover benefits into other areas of the community. There’s a multiplier effect of five on tech jobs, meaning every tech job supports five other non-tech jobs in the community. This is a much higher multiplier than export industries like manufacturing. Innovation jobs also raise incomes across the board, support higher quality public services through the taxes they generate, and a slew of other civic benefits. Places that did not make the transition not only did not see the benefits of the innovation economy, they also experienced severe dysfunction as their older industries crumbled.
This divergence of fortunes has happened in cities across America. Rather than a red vs. blue America, Moretti sees three Americas: the winners, the losers, and those on the bubble.
A large number of the cities that I’ve traditionally covered on this blog, especially smaller major Midwest metros, are places that I would classify as on the bubble in Moretti’s classification system. These are places that haven’t become Flint, Michigan yet, but they aren’t Austin, Texas either. They are cities that could go either way. They may in some cases be winning, but they have not yet won. It’s not inconceivable that any of them could flame out and implode.
Yet I see in places across America little sense of this existential struggle. Leaders are missing a framework like Moretti’s (or anyone else’s) to help them make sense of what has been happening in the world. They do not appear to have a sense of urgency around change. In particular, they are not serious about attracting talent or knowledge industries. They might pay it lip service, but everything they do says otherwise. It is actually those who are already in the best shape that are the most worried about securing their future. Parable of the talents indeed.
One of the key takeways for me from the book is that cities should be extremely keen not to go backwards. To the extent that you have a nascent talent cluster, you should take care of those young shoots until they mature. That’s because Moretti examined various strategies for trying to turn around one of the “loser” cities, and basically came up empty. They all failed. As he noted:
Regions without an innovation cluster will find it difficult to start one. It is a chicken-and-egg problem. Specialized high-tech workers will not move to a city that does not have a cluster because it will be hard to find an employer that values their unique skills. High-tech companies will not move there because finding specialized labor will be difficult. This presents a terrible challenge for communities that have fallen on hard times and are struggling to reinvent themselves.
Moretti’s only recipe is what he terms the “Big Push”:
A city stuck in a poverty trap faces the same challenges. It is trapped by its past. The only way to move a city from a bad equilibrium to a good one is with a big push: a coordinated policy that breaks the impasse and simultaneously brings skilled workers, employers, and specialized business services to a new location. Only the government can initiate these big push policies.
In short, Moretti advocates industrial policy on steroids. There’s just one problem: he never actually cites a single example of where it works for a city. There’s only one success story that’s at all relevant that he mentions, and that’s the Tennessee Valley Authority. And the TVA, however successful it might have been, was not dealing at all with knowledge or innovation industries.
We do have a few examples out there today of at least partial big push type efforts, both private and public. New York and Chicago both have put major efforts into trying to build tech hubs from scratch. Both of these places were already talent hubs, however. Billionaire Dan Gilbert is trying to do a sort of one man big push in downtown Detroit and Zappos founder Tony Hsieh is trying something similar in Las Vegas. These two both have money to spend and also have clout in their communities to bring others along with them. These will be interesting to watch.
Where Moretti does provide some hope is in a observations about the genesis of clusters that already exist. For many of them, the key factor seems to be the presence of “superstars” at a critical phase. In our Seattle example, Gates and Allen were the superstars. (Tony Hsieh may yet prove to be the superstar for Vegas).
The superstar theory is especially applicable to biotech. Tons of cities have great academic medical centers and such. Why did biotech then end up disproportionately concentrated in Boston, the Bay Area, and San Diego? (Other economists like Joe Cortright have a somewhat broader list that includes Los Angeles, Philadelphia, Raleigh-Durham, Washington-Baltimore, and New York). According to researchers, it was the presence of superstars that drove this. As Moretti notes:
What really explains the location and success of private biotech companies is the presence of academic stars – researchers who have published the most articles reporting specific gene-sequencing discoveries…The data suggest that the magnetic effect of academic stars is impressive. Zucker and Darby estimate that stars are more important than proximity to venture capital firms or the effects of government spending….Success in high technology, especially in its formative years, comes down to a small number of extraordinary scientists with vision and a mastery of breakthrough technology. Indeed, we can’t overestimate the impact that these unusual individuals have on the economic development of cities and regions.
Looking at stars would then appear to be the best bet. As the Seattle example showed, a couple of stars showing up at the right time can literally change the entire economic trajectory of a failing region. The legion of cities that experienced rebirth after Rust Belt malaise suggests that stars can’t be the entire story, however. Saskia Sassen, for example, takes a very different view based on structural changes from globalization increasing demand for advanced producer services linked to historical economic functions in select global cities. This was not a superstar phenomenon, though we see tremendous innovation.
And of course it’s hard to just say “go get me some stars.” The stars mentioned helped launch new clusters in new industries. It’s not clear that they’d start a new hub in an established industry with already established hubs. So hiring the top biotech guns right now might not be the best idea. Also, one can perhaps only recognize stars in retrospect. I think again of the story of Mark Andreessen, who was in Illinois but ended up getting run out of town. Tomorrows stars are very likely people who don’t look like it today. I won’t pretend to have all the answers here, but these is clearly an area to think about.
While I think Moretti had a lot of great things to say, I think he missed a few points as well. First, in addition to a divergence between the haves and have-nots, there’s another divergence we’ve been between what I call “vertical” cities like the Bay Area, Boston, and New York, and “horizontal” cities like Houston, Austin, and Charlotte. Vertical cities are producing stunning growth in output and wages, but very few jobs. The horizontal cities are producing lots of new jobs, but little or no per capita output or wage growth.
Moretti describes both the Bay Area and Austin has tech innovation hubs, yet these cities have staggeringly different profiles. For all its tech growth, the Bay Area has added almost no net new jobs since 1990 (4.6% total growth). During the same period the US as a whole grew jobs by 20%, so the Bay Area radically underperformed. Meanwhile Austin doubled its employment.
On the other hand, the Bay Area grew per capita personal income much more strongly than Austin. In fact, during the 2000s, Austin actually suffered a severe erosion in relative incomes, as its PCPI dropped from 108.1% of the US average to 97.0% (This after a 16 percentage point gain in the 90s!) Clearly being an innovation hub didn’t help there, and Moretti doesn’t address these issues.
He does hit on how restrictive housing policies drive up prices in places like California, and thus effects the purchasing power of those high salaries. But I would like to have seen more on the different philosophies of places like SF and Boston vs. Austin and Raleigh-Durham. These are not all just vanilla “innovation hubs.”
Moretti himself seems to side with the global city crowd, being a bit enamoured of the vertical style economy. He mentions how high costs, for example, don’t deter businesses from locating in these places since talent matters more than costs. That’s why, for example, Wal-Mart located its e-commerce division in the Bay Area not Bentonville. This trumpeting of defects as virtues (e.g., treating high costs and huge regulatory burdens as actual signs of health) is common, but the job creation record of those hubs suggests most businesses don’t find them attractive places to locate. Rather, they are like boutiques, specializing in a narrow, high end niche. Certainly America can, and perhaps should, sustain a few such places. The Bay Area may be the creative capital of the world. New York is New York. But these are not templates to follow.
Moretti also focuses in on fairly obvious examples like Silicon Valley and Detroit. As a research economist, I would have expected some more interesting choices. There weren’t a lot of a-ha’s in here. It seems a bit superficial when it comes to examples, and may actually mislead some people. For example, Fargo, North Dakota has been doing very well thanks to not just the energy boom (more on that later) but also because Microsoft has a huge operation there (reputedly its second largest in the US) because of the purchase of Great Plains software. America is littered with places apart from the obvious that are doing well, and more exploration of these would have given the book more depth. As it is, it can seem a bit superficial.
This superficiality shows up, for example, in the way Moretti highlighted companies that do acquisitions ostensibly just to acquire the talent. He cited the example of Sam Lessin, founder of drop.io, whose company was acquired by Facebook and shut down. The idea was that Facebook really wanted Lessin because he was just an amazing talent. However, as Gawker reported, Mark Zuckerberg had a longstanding relationship with Lessin going back to when Zuckerberg was still in college. Lessin’s father is an investment banker, and Lessin was apparently instrumental in making a number of early introductions for Zuckerberg’s nascent Facebook empire. The story isn’t quite what it might have seemed.
Also, back to Fargo, Moretti doesn’t mention anything at all about the domestic energy boom we’ve experienced in America in recent years. He managed to cover trends like urban manufacturing, but overlooked fracking? Assuming environmentalists don’t manage to choke off exploitation of our vast energy resources, the energy industry should be entering a growth super-cycle. (Even George Monbiot has thrown in the towel on peak oil).
I wonder too whether the trends Moretti cites, while valid over a 40 year cycle, are less true today. One way economists measure clusters is by using a metric called Location Quotient. This measures the concentration of employment in a particular industry in a specific industry relative to America as a whole. But the math works for lots of things. So we can look, for example, at literal clusters of talent by looking at the location quotient of college degree attainment. Here is a map of changes in the location quotient for college degree attainment from 2000 to 2011.
This is certainly interesting. Many of Moretti’s talent hubs actually are less concentrated in brainpower relative to America today than they were in 2000. Out of 51 metro areas with more than a million people, Austin ranked 50th on this metric. San Jose and San Francisco were 42nd and 43rd respectively. (Pittsburgh was #1, incidentally).
Now the changes are very small and could be statistical noise. But this doesn’t look to me like the vast divergence we saw looking over that 40 year span. Instead, what it looks like to me is that a number of cities that weren’t in the game at all back in the day have started to turn the corner on talent. This foots to my anecdotal view of the livability of cities. Back in 1990 moving to say Indianapolis from a tier one city would have been getting exiled to Siberia. Today it’s shocking how little you give up. This is true of other strong talent performers like Fargo and Des Moines as well. And you’re definitely seeing at least some level of innovation based economic activity showing up in all these cities.
The strong may well be getting stronger, but those who were behind are growing their brainpower at an even faster rate. I wonder if perhaps innovation/knowledge work has reached the point on the maturity curve (as Moretti talked about for other industries) where it has stopped concentrating and started de-concentrating?
Lastly, I would urge caution in looking at Moretti’s maps. I know Indianapolis extremely well so wanted to see how it fared in his rankings. I was very surprised to see it rank in the bottom group on basically every map. According to Moretti, Indy’s share of workers with a college degree could not exceed 20.4% Yet it has a college degree attainment rate of 31.1% These aren’t the same metric, and I don’t have the worker share handy, but that big of a gap seems suspect. Indy is showing as lower than far less educated metros in Indiana like Evansville and Terre Haute. Indy also showed as “no data” on the patent map, which seemed odd. I wonder if perhaps there was a coding error in the map generation? I’m not sure, but it looked possibly off. I tried to get clarification on this point, but wasn’t able to, so I’d simply suggest being careful with the maps.
Despite a few things I thought could have been more developed or stronger, I think that, as a book making the case for the innovation economy and what it means, The New Geography of Jobs is a strong one, and I again would suggest it to people in cities that have not yet fully found their place in the new century.