Here’s a simple question for you…which metro areas did prospered the most during the past business cycle? (2000-2008) Were the winners the highly-educated communities that make up the Creative Economy? Or did someone else zoom ahead?
I asked myself these questions when I was preparing for a talk that I was giving at the Rochester Institute of Technology on innovation and economic development. Being a man of numbers, I calculated the gains in real-per capita income for all metro areas. Who do you think was #1, and who do you think was #366 (out of 366)?
A bit surprising, isn’t it? The common themes are guns and oil. The big gains in the #1-ranked Houma region are mainly connected with the increase in oil drilling, since BLS data shows that wages in the mining/oil industry in Terrebonne Parish, where Houma is located, soared from $58K a year to $78K from 2005 to 2008. #2 Jacksonville (NC) is the location of Camp Lejeune. Fayetteville (NC). #5 Fayettville (NC) is home to Fort Bragg, one of the larget military bases in the world. #6 Killeen is obviously home to Fort Hood. #8 Odessa, Texas, is riding the oil boom.
Now let’s look at the metro areas which were the biggest losers in real per-capita income, 2000-2008.
Uh, oh. This is not the list you might have expected, in a world where brains and innovation are supposed to be important. There’s Silicon Valley at the top (or the bottom) of the list, where incomes didn’t recover from the popping of the tech bubble that peaked in 2000. But other tech-type metro areas, such as Raleigh and Austin were hit hard as well.
Brains and education did not seem to count too much in success in the last business cycle. Overall, the top ten cities, measured by growth in per capita income, had an average college graduate rate of 17.7% The bottom ten cities had a college graduate rate of 31.8%.
Is this inverse relationship between growth and education going to persist into the future? Impossible to say. My personal view is that the lack of rewards for education–which show up in the individual income statistics as well–is correlated to the lack of commercially-successful breakthrough innovations, which would immediate sop up all the excess college graduates.
To put it another way, innovative industries tend to locate where they can get a lot of college graduates. That means high education areas attract new companies, boosting growth.
But without innovation, the whole economic development dynamic changes. You can’t attract growing innovative companies because they are few and far between. For their part, companies are more likely to view cost as a main consideration in deciding where to locate. Goodbye San Jose and Austin, hello China and India.
Mandel on Innovation and Growth 
Thanks for picking up on this. We’ve used it as a hook–as unfortunate as it may be–to get people here in Austin thinking more about the relationship between education, workforce development, and economic development. I posted about this as well a couple of weeks ago on my blog: http://civicanalytics.com/austin/austin-salaries.
Good point.
Yeah, I’m agree with you
Obviously San Jose was going to drop from 2000, the top of the tech boom. I’m not sure what you hope to learn from this data, all it highlights is sectors that gained or lost and the metros most associated with them, such as the big drops in Michigan caused by the auto bust. Also, by highlighting relative changes and not absolute levels, you ignore the fact that San Jose is still one of the highest-paid metros, because it was so high-paid to begin with. Your education-innovation analysis is hopelessly outdated. Education has never been that important, the good news is that it’s basically getting ignored now in favor of experience, although experience is still measured very crudely. Further, your backwards-looking education-innovation clustering theory has been rendered irrelevant by the internet, where anyone can work with everyone from anywhere. Google has small teams that work all over the world and communicate electronically, many other companies do some version of this far-flung distribution of work. Education is made even more irrelevant as nobody can gauge the value of a degree from anywhere in the world, making employers focus even more on actual experience. Finally, your notion that innovation keeps the college-educated employed is hysterical. If these people are so smart, shouldn’t they be innovating while they’re having trouble finding work? Oh I see, they can’t innovate but they must wait for innovation to fall from the sky and keep them occupied, makes perfect sense. 😉
If it was just San Jose, I’d agree with you.
I don’t see why the decline of other tech-heavy markets in parallel with San Jose causes you to disagree. If tech salaries were inflated during the boom, then we would expect to see income declines in tech markets.
I have two points of disagreement. First, is that it appears that the decline in tech markets is more likely due to inflated tech salaries returning to normal or market levels. Additionally, just noting that income in tech-heavy markets is down the most says nothing about the return by education level overall. How did markets with lots of biotech or finance perform?
Second, it is easier for markets with smaller starting income level to post higher growth rates. If an oil discovery or a new factory or a new company office is opened in a small market, that is going to have disproportionate effect on growth in that area. The discovery of $100b worth of oil in a small market with a per capita income of $15k is going to make a big splash in that market. Additionally, less developed economies tend to grow faster as they catch up to the rest.
I think the numbers you are seeing are the result of confounding factors, even if the results are consistent with other data. It would be interesting to see growth rates in the 10 most educated markets vs. the 10 least educated. It would also be interesting to see absolute increases instead of percentage increases.
Ajay
In your posts, you have repeatedly criticized the link between education and innovation. I accept that your argument may apply to many areas of the economy. The case you seem to be making is that it applies broadly – that we don’t really need all those engineers and scientists with graduate degrees who design medical imaging equipment, model internal combustion engines or design new computational geometry algorithms.
There are many aspects to innovation and the type of innovation you seem to be referring to is important. However I would suggest that aggressively following the model that you believe in would eventually result in less and less meaningful innovation, eventually getting to the point where minor changes to a user interface is considered innovative. If the innovators that you champion do not have meaningful new ideas coming from upstream on which to base new products and services, they will end up just reimplementing existing technologies repeatedly and will find their work becoming commoditized. This naturally would have a negative impact on GDP growth.
Luke, a couple of responses. First, engineers and scientists who design medical imaging devices or new car engines are few and far between. Just because we neeed a few people to do heavy-duty work doesn’t mean most people would benefit from even a fraction of what a highly specialized engineer would learn. Second, you retail a fallacy that education or the university environment have much to do with innovation, Nassim Taleb has said that empirically most innovation actually starts in business and then works back to the backwards universities. I’m not making any claim favoring one type of innovation over another, I’m claiming that all innovation would be better served without the horribly backwards education systems used throughout the world today.
“At right angles you people talk!” Tom Yoda-ed.
Mandel, you’re the very model of a 1960’s policy wonk. You look at numbers which summarize human lives, you see patterns, you ask if they’re _good_ patterns, and wonder what we might do to improve those lives. Ajay’s attitude is “I’m all right, jack!”
And Ajay and people who think as he does outnumber you. They will contribute more to politicians who share their views, they will outvote you. Likely they will outearn you — they will not understand your interests, or your modes of thought. And they will furnish the middle class for whatever sort of society the American republic builds in the 21st century, or perhaps for all time to come.
Welcome to the future, guy.
shupp, I’m not even sure what your comment is supposed to mean, that I and most everybody else don’t care about statistical trends and Mike is going to be ignored by us? Far from it, I pointed out the specific flaws in Mike’s metrics and analysis: you are the one apparently ignoring the holes in Mike’s post, since you cannot make a single specific argument about it. Always easier to just lump everyone into some imagined herd than make actual arguments, I suppose.
1960s policy wonk? Nah, then this blog would be full of proposals and suggestions for fixing the world. Truth is, I just try to look at the world as it is, without assumptions or political preconceptions.
Ajay,
I love your stuff. Don’t be a troll, though. Michael deserves better.
The hard data from Australia a decade ago was that there was no more certain relationship than that between socio-economic status and educational attainment.
The the loss of a causal link between education and se status (if that is what we are looking at) is a really interesting phenomena.
Is it a failure of education, or evidence of a failure of a civilisation?
Ha, I don’t think you know what a troll is. Please don’t use such terminology if you’re not even sure what it means, it just makes you look confused. The reason you’re confused about the “correlation” between education and income is because you’ve been looking at cooked data served up by schools and their pushers: here’s an example of how when you dig even a little bit, the lies fall apart. A failure of civilization? In that those who wasted the most money on education are entitled to higher income, no matter what? What a joke, that’s where the aphorism comes from that the C students boss the B students at work, while the A students teach. 😉
This is an interesting post, but I think it’s mostly just playing with numbers. It’s saying that certain metro areas had decreases from their previous highs, and those metro areas tended to contain better educated people. What it DOES NOT say is that within each of those areas, more educated people were still no doubt more likely to be employed, and at higher salaries. So, in the end, one thing I know it does NOT mean is that we should abandon the notion that more education is a good thing, whether measured by individual incomes or metro areas.
Also, I don’t buy the “guns and oil” argument. Certainly successful oil drilling drives up incomes, but how precarious is that? I mean, isn’t that the area of La. where the geyser is spouting? As to guns, I believe that those same areas with big military bases are also WIRED areas: Fayetteville and Jacksonville, NC, and Manhattan, Kansas. Of course, the presence of the such large military installations may in fact drive up innovation, which is really the point the original author is trying to make.
As a general rule, I’m still OK with the byword of “find your regional competencies “– workforce-wise and “culturally” – and go with them, and go to school!
The results are consistent with a narrowing of the education gap in this period, which is what the Census income numbers show. I’d say that innovation and education are still the right long-term strategies for economic development–but they didn’t work in this period.
Lawton, OK is home to Fort Sill…
Measuring peak to trough?
2000 was definitely a peak. The next peak was either 2007 or 2008–it’s a judgement call.
Ajay – A large part of the Ann Arbor, MI numbers, I’m willing to bet, are due to Pfizer closing a very large research plant circa 2005. It was a very prominent employer of Ann Arbor’s upper middle class, and there was little work anywhere else in the area for the type of specialists Pfizer employed, so a very large number of experienced, well-paid research scientists left Ann Arbor, likely driving a portion of those numbers. Ann Arbor has never had much manufacturing, and has been fairly insulated from the collapse of the auto industry. A fairly large part of the city’s economy is based around the University of Michigan, with various small tech businesses employing graduates. I think Ann Arbor fits to Mandel’s analysis, even if he didn’t mention it.
Agreed! Have seen this happen with a friend and a bunch of people in Ann Arbor.
Mike, a very though provoking post. An even more interesting question would be, “what happened to innovation?”
Here is my hypothesis. In the wake of the tech-bubble bursting, the efficient allocation of capital was skewed by the “housing bubble.” I bet if you shorten your time horizon to 2002 – 2005, the largest income gains would be seen in So. Cal, FL, AZ, and NV. Flipping houses, pushing liar loans, installing granite countertops didn’t require innovation and didn’t reward education. Capital was drained from the innovation economy, making it more difficult for start-ups to acquire funding.
At the same time, you had a crop of recent graduates who entered college/grad school in the late 90s/early 00s in computer science to chase instant millions in the dot-com boom flood the market with highly educated, well qualified candidates just when the sector was contracting.
If we look at this data in 5 years, I bet you’ll see the big finance and housing boom hubs at the bottom of the list. The 00’s saw a crush of students majoring in finance and economics, going to b-school, etc. all in the hopes of landing making millions in hedge funds and on fixed income trading desks. We lost another decade of potential science and technology innovators to financial alchemy.
I’m not convinced that we’ve returned to an era of rational and effective capital allocation. We need to move from an era of financial engineering to one of mechanical, chemical, and bio-medical engineering.
SpeciousRiches
Great point. I hadn’t read your comment before writing something similar below.
As for “glut of recent graduates” (including then-recent graduates coming to the US on H1B etc.), this alone cannot be it. There certainly is, and has been in dotcom times, a glut of “old experienced hands” i.e. engineers in various tech specialties in their 40’s-50’s with the corresponding years of experience under their belt. During the dotcom “IT worker shortage” (“artificially” created by boatloads of investor money sloshing around in the economy) companies who weren’t first-rate brand-name employers had to hire a lot of staff who was older, with unconventional career history, or “interesting characters”, all of which will usually be shunned in a “normal” economy. That included people with “personal issues” who were good but difficult to work with when your standard is “exchangeable cogs who can be pushed around at will”. When looking at technical merit, i.e. what those people could deliver compared to (inexperienced) “recent graduates” (domestic or abroad), that didn’t work out too badly. When the cuts came, those were more or less the first out the door.
But in the aggregate, they all add up to a workforce/skills oversupply.
Interesting as all change occurs at the margin and these are the significant margins. I would say it is a shortage of valuable ideas and results from them. Innovation cycles tend to be long with discoveries, research, and development taking a long time and often are a matter of serendipity, obsession, perseverance, and desperation than planning and profits. It is questionable how durable the gains of guns and oil will be over time while the educated will be positioned to take advantage of what innovation does occur. It may still take another decade for substantial progress in innovation as it has a long gestation period and often arises from the confluence of other innovation occurring at the same time. Are we even assured it will occur? Perhaps not, but it is better to believe so as the alternative is not pleasant.
“Paradigm change” innovation comes in spurts. The combination of some recent innovations will be “good enough” for a while. Every paradigm change leads to a new wave of legacy installations (in the case of equipment) and legacy methodologies/skills that have to be amortized. This creates a barrier for new paradigms. The new paradigm (if compelling enough) will gain traction when the old paradigm is either discredited in some way or its support base wanes (decision makers and available workforce invested in the old paradigm retiring or kicked out from business). In addition to this, the new paradigm may have (perceived?) unaddressed disadvantages. For example, “cloud computing” and generally internet based business raises the issue of privacy (on the consumer side) and confidentiality as well as control over and availability of critical business data. For communications we have VPN, but the issues with data storage and application hosting are not plausibly addressed.
Excuse me if I missed it in an earlier post, but how are you computing *real* per capita income? That suggests you have a metro-specific inflation measure and, sorry, I’m not aware of one that covers 366 metro areas.
And shouldn’t you also deal with regression to the mean? The 8 highest gaining areaqs would seem to be areas that started from pretty low values; the 8 lowest-gaining areas started, I would guess, from pretty high values.
It would be interesting to add to the chart current incomes instead of just delta-from-2000.
Look at some of the businesses that boomed with the housing bubble: construction, realtors, mortgage brokers. What education or training level was needed for those? Think about a good mortgage broker, or underwriter. They’d be refusing to lend money and losing their jobs. This past bubble, call it a business cycle if you want, rewarded incompetence and/or malfeasance.
To me, these tables reflect the economic stability that major universities bring to the surrounding areas. They are generally immune from boom and bust cycles.
Is there any possibility of this data coming from extremely high growth in population outstripping supply of high paying jobs? Austin, for one, has almost doubled in population in the last 10 years, and I wonder if some of the per capita loss of income might have to do with too many (highly educated) people coming to Austin to come to a “hot” city. Innovation may still be active in a case like that, just not active enough to keep up with such high growth.
Brennan, you are on the right track, at least in Austin’s case. We’ve seen tremendous growth in technology services in the last 10 years, which in part has absorbed some of the losses of higher-wage jobs in computer and semiconductor manufacturing here. But the fastest growing industries by far are in lower-wage jobs that result from population growth: retail, restaurants, and other local services. No question that entrepreneurship and innovation are still alive and well here, but we’re not seeing the returns yet in the form of large, sustainable gains in high-wage jobs. The joke making the rounds now is that we’re becoming the land of consultants.
In 1999, we took a very long cross-country trip, out one way and back another. The entire country looked incredibly prosperous … except for the “oil patch” of Odessa, Texas, Midland, etc. The big gains in per-capita income since then certainly have something to do with starting from a depressed base.
Poor Flint, MI. That city never catches a break, does it?
No, it doesn’t, but certainly Flint, with only about 12% of the adult population having 4-yr degrees, also does not fit the argument that inversely correlates education and economic growth. On the other hand, Flint is long on manufacturing experience, but that does not seem to be paying off right now.
Flint needs both more graduates and more companies that can use their skills.
I live in Manhattan, KS. It is rather shocking to see a statement like…”meanwhile, in much more educated and vibrant cities like Raleigh and Austin…”
The university is the only reason Manhattan exists. At least half of the community is engaged in one form or another with the university.
What is consistent with the thesis of the article is that since 2000, Ft. Riley, on the western border of the area, has expanded greatly. A majority of civilian management and officers now live in Manhattan rather than other adjacent communities. This has driven a housing and retail boom.
Further growth is being driven by the relocation of several research labs to KSU / Manhattan. Most of these will be federally funded, and will employ a college educated work force, included a large number of PhD level research leads.
So the take-away from Manhattan. KS is that military and public-sector growth is, indeed, a major driver of employment. And when injected into a small community, can result in fairly dynamic economic activity and salary growth.
From my perspective, the data from Manhattan, KS, would run, at least in part, counter to the idea of the article.
I agree–the data from Manhattan KS partly runs counter to my thesis. Still, nine out of ten ain’t bad.
From 2001 San Jose-Sunnyvale-Santa Clara is the poster child for labor mobility vacating an areas failing economy. 30,000 high end tech employees and whole teams of tech management relocated into other areas and into other employment behind the Dotcom bust. Then out to 2008 we had the transfer of tech services and most of the remaining production into Asian centers of tech production. The work of executives like Carly Fiorina transferred 10’s of thousands of domestic high tech jobs from the Bay Area.
The Bay Area at the same time remained a major center of real tech wealth created within the domestic economy. You discounting that level of infrastructure build and tech wealth and innovation that was generated, because its not centered in domestic labors gains. That wealth was created with the leverage of foreign labor. That is a systematic shift in labor that cannot be recaptured. Anything new in tech goes there for production because of the cost of production and return on product. Try finding more services that will expand the domestic economy, and a tax structure to gain investment in them, because all manufacturing and production and tech innovation will be shared with world centers.
San Jose-Sunnyvale-Santa Clara in particular is a bad choice for the example.
Why is it a bad example? The measure I used was personal income, which is more than wages–it includes dividends and other capital income (though it does not include capital gains).
I agree with your facts (and projections), but I don’t understand your conclusion about the bad example.
OTOH, the “Silicon Valley” tech workforce is to a large degree not local. Graduates are from anywhere, prominently from abroad. There is probably little correlation with local graduation rates.
Mobility of professionals in general (not sure that tech is even the largest phenomenon) is probably pretty high. People grow up somewhere, often go to school somewhere else (at the most renowned institution that will accept them?), then take a job wherever the jobs and growth potential are supposed to be at the time.
Later in life they can listen to a lot of admonitions of not having made prudent choices. (No sour grapes (yet?) on my part here but I’m reading the public discourse.)
There are some ideas for energy-efficient search in a house or you can do when you move into a house.
Even if it is not absolute green hospitality more than a normal high
I really like this blog and article, Please continue the great work. Regards!!!
La qualite des films en streaming sur ces le contenu peut etre de mauvaise qualite et fractionne en plusieurs sequences.
Vous souhaitez regarder vos films, series et
mangas en illimite?
Good website!