All those forecasts of the end of cities have yet to come to fruition. Perhaps the following chart, via Richard Florida (The Atlantic) helps to explain why:


Category: Psychology, Science

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

31 Responses to “Human Capital “Density””

  1. YY says:

    Whoever decided to measure per mile most likely didn’t finish even high school….

    Why not measure it per population! Believe me the results will come out quite different!

  2. DeDude says:

    Normalizing to “per square miles” makes the whole thing pretty useless. A chart of population density would look pretty similar. You need the percent of inhabitants with a college degree to make any sense of whether the numbers are an issue of population density or population education levels.

  3. Tom K says:

    When will we see the college grads vs. donuts sold chart? It would be about as meaningful as this.

  4. The reason density per area matters, at least to this author, is noted in an article he wrote for the Atlantic a while back:

    “The University of Chicago economist and Nobel laureate Robert Lucas declared that the spillovers in knowledge that result from talent-clustering are the main cause of economic growth. Well-educated professionals and creative workers who live together in dense ecosystems, interacting directly, generate ideas and turn them into products and services faster than talented people in other places can. There is no evidence that globalization or the Internet has changed that.”

    In the author’s view, 100 PhDs living close to each other and interacting daily are more valuable than 100 PhD’s spread out across a less-populated area where they’re unlikely to interact or generate many ideas together. In his view it’s not “talented people per capita” but rather it is “talented people in close quarters with each other” that generates wealth. Whether there may or may not be lots of other people around them (the per capita view) is less relevant to him.

    Taleb echoes the same view in The Black Swan, chapter 13, where he writes: “Work hard, not in grunt work, but in chasing such opportunities and maximizing exposure to them. This makes living in big cities invaluable because you increase the odds of serendipitous encounters — you gain exposure to the envelope of serendipity… Diplomats understand that very well: casual chance discussions at cocktail parties usually lead to big breakthroughs — not dry correspondence or telephone conversations. Go to parties! If you’re a scientist you will chance upon a remark that might spark new research…”

    I haven’t crunched the numbers to try to prove or disprove the hypothesis, but the view from 100,000′ suggests that the hypothesis has some merit.

  5. pmorrisonfl says:

    I’m surprised the Research triangle area cities didn’t rank more highly.

  6. powersjq says:

    @Michael Gat: Nice follow-up. Jane Jacobs makes a similar point in her book, _The Wealth of Cities_. In it, she makes a compelling case that agriculture (yes, agriculture!) was invented in cities, then exported. It’s the _unplanned_ encounters between people that advance ideas, the _unplanned_ linkages between this bit of technology and that bit, that make an economy hum. Cities have always been the places where innovation happens because they make it easier for such serendipitous encounters to happen. The only kinds of technology that matter in the development of cities are the ones that make face-to-face encounters easier (think: skyscrapers and transportation) or more pleasant (think: cafes and cocktail bars).

  7. AGG says:

    It seems that college educated people can squeeze together with less violence resulting. Although this may be correlation without causation, consider that many people that own a rifle or a shot gun prefer rural living. Those people tend to overwhelmingly be high scool educated or less. They are the ones who claim to have it “all figured out”. They are the ones who never doubt their entitlement, politics or loyalties. They are also the easiest to manipulate. They get mad as hell when you tell them this but the fact is that people with low levels of education in the USA are uneducated by choice. After college there are few people that stop reading and learning. After high school, too many stop reading and learning. Those who do continue to read, learn and question are those most likely to achieve harmony with theire fellow humans in a large city. It also needs to be said that a small percentage of college educated people stopped reading and learning after college annd are as bigoted and closed minded as the most uneducated person around.
    Of course, you could question the whole concept of the chart and say HEY, they’re just going where the jobs are, duh.

    I wonder if any social scientist or university psychology department have studied human behavior in elevators. I find it fascinating how some people head straight for the corner, others mill around, others face the wall and few, if any, make eye contact. There are so many social dynamics coming into play in an elevator. There is the potential of ahaving to spend hours with these strangers within inches of you if the elevator gets stuck. There is the “opportunity” to experience various body odors, perfumes and cologne. Then there is the great view of footwear. How many people can tell you what an elevator’s cieling looks like?
    It would be most enlightening to study this stuff. There is a definite tension everyone experiences in an elevator. There is an adjustment to proximity levels that everyone is making for the convenience of riding an elevator. We are will to forego a lot of personal preferences for the sake of an elevator ride.

  8. I suspect the Research Triangle ranked where it did because of the way he chose to aggregate the data and the size of the area (including outlying suburbs and rural regions) that were included in the “metro area.” He does include the full data for anybody who wants to slice or dice it on their own.

    Realistically, I can think of lots of places that would probably rank a lot higher if you included only the “core” city and dropped lots of the outlying suburbs/exurbs that are often all but disconnected from the core. But it is an interesting cut at the numbers.


  9. Thor says:

    AGG – Yes, believe it or not, behavior in elevators has been studied extensively. Next time you’re in an elevator stand the wrong way (with your back to the doors) and see how people react.

  10. Just an example of how subjectively cutting the data may alter things:

    San Francisco and San Jose are aggregated into a single metro area per the author. Oakland, which presumably includes the entire East Bay north of San Jose, is its own metro area, even though it’s just as close to the other two as they are to each other and arguably better connected to SF than San Jose is. Aggregating the entire SF Bay area would unquestionably result in it scoring lower, though most likely would just move it from the #1 spot to the #2. (Although I can also see NYC scoring lower depending on how much of Westchester, Fairfield, Nassau, and Northern NJ you chose to include as part of the “metro area.”


  11. constantnormal says:

    Paul Kedrosky and Barry must be at the same gathering in SF, ’cause Paul is running a similar/related story to this, but Paul is looking at the declining usage of cars to teens and wondering if this is due to the alternative networking provided by the internet.

    Barry’s chart points out the (purported) multiplier effect of large numbers of educated people in close proximity to each other, which implies a higher level of networking among those folks, and consequently, greater (in theory) productivity.

    If this is in fact the case, then today’s marked increase in connectivity between physically separated people — via the internet and cellphones — will go a long way towards flattening this chart. It would be nice to see some measure of intellectual productivity per square mile for these cities, both in 1978 and now. I’ll bet that such a chart would show the same sort of flattening that Kedrosky’s chart of car use by teens does.

  12. I don’t read Kedrosky’s comments that way. He states clearly: “my guess is that the decline has at least as much to do with economic considerations, given rising youth unemployment, and higher gas/car/ insurance prices (in inflation-adjusted terms). That’s not to say that online stuff isn’t making physical presence less important, because it is, but economics matters at least as much, or more.”

    Commenters to his piece note that in many states it’s become much tougher to get an unrestricted license at 16, in many other states insurance becomes unaffordable with an < 18 year-old driver in the household, etc. And of course, this dovetails nicely with the stuff from the Atlantic showing that our world is becoming more urban and the young are flocking to the cities where drivers licenses are far less necessary. I'm 45 and driving much less myself. So it's hardly surprising.

    And as noted above, the existence of better communications does not seem to have changed the fact that in-person communication seems to matter a lot. My own view is that social networking tends not to be serendipitous in the way that real-world interaction does. You don't accidentally end up standing next to somebody who you have no connection to and getting involved in a conversation when you're in a social network, yet those tend to be the connections that matter most.


  13. constantnormal says:

    Yeah, but I don’t think you can separate the economic effects the way that he does — I think they are part and parcel of the same thing, the flattening of hierarchies by pervasive communications networks.

    This is not to say that you don’t have spikes arise out of dense networks — Stephen Johnson’s book “Emergence”, and Albert-Laszlo Barabasi’s book “Linked: How Everything Is Connected to Everything Else and What It Means” show that emergent poles of influence always arise in networks, but with the rise of dense communications networks, the importance of physical proximity rapidly decreases as the number of connections increases.

    As for the utility of an interaction with someone you have no direct communication to at a party, I would say that is exactly the same situation as a chance remark by a commenter that you have absolutely no connection to in a blog forum. And there can be hundreds of thousands of participants in a blog forum, while a party is pretty well constrained by physical space and the economic costs of attendance. I certainly don’t deny the immense value of chance interactions, but I fail to see the essential need of physical presence for such interactions to occur.

  14. hammerandtong2001 says:

    @ Michael Gat

    Not that it’s a piece of high scholarship — coincidentally, yesterday Ad Age published an interesting piece about the “decline of car culture” in America. Interestingly, the number of drivers under the age of 30 in the US has plummeted, and it appears to be a secular trend which started in the mid-90′s — well before the widespread use of digital services.

    I think kids these days view automobiles with the same disinterest as I might view a refrigerator. Gone is the zeitgiest of the “open road” — they all wanna hang together in “urban areas”.

    I was 17 when I bought my first car — a 1964 Triumph Spitfire 4, Mk I. It didn’t run. It cost me more to have it towed to my dad’s garage, than I paid for the thing in the first place. In 3 months it had a new engine, new transmission, new clutch — I drove it for 2 years and got every penny I spent on it back in a $400 sale. I loved that car.


  15. patient renter says:

    Oakland does quite well, considering.

  16. Well, the Oakland metro area most certainly includes Berkeley, the Biotech-heavy Emeryville/Alameda areas, the nuclear physicists out in Livermore, the various tech companies out in Pleasanton, etc.

    I suspect much of the reason for Oakland’s low score compared to the San Jose-San Francisco corridor is that it includes a lot of open space between Walnut Creek/Pleasanton/Livermore and the Bayside cities. Lots of open space, parks, golf courses, etc. can skew the numbers. L.A. is likewise affected by the presence of large chunks of open space within the boundaries of the metro area.

  17. @constantnormal

    My disagreement with that perspective is that I’m unlikely to run into somebody who has little in common with me while reading or commenting on a blog, while I’m much more likely to when going to my neighborhood bar (especially in a dense and relatively diverse neighborhood). The current business proposal I’m working with came from the interaction of a tech/finance guy (me) a media person and a semi-employed actor. A relatively common enough encounter in a bar in Venice Beach, but not one that I’m ever likely to have on any of the blogs, websites or social networks that I frequent online.

    Social networking, I believe, tends to channel us towards the familiar and connected. We rarely reach out beyond the “friend of a friend” or “reader of the same stuff” level. Neither of the two people who sparked this recent idea would have been likely to be found in my online experience. But at a bar with the wine and beer flowing, things happen that are a real challenge to replicate in an online environment. I mean seriously, when is the last time you decided to go online and seek out people completely different from you in the hopes of getting an idea you didn’t have before? Yet those tend to be the most powerful connections and ideas.

  18. Stav says:

    Oakland surprised, but the biggest shock to me was Miami. Fantastic place in February, but you never get teh sense that too much gets accomplished there (and no I don’t isolate myself on SoBe), and it is statistically one of the poorest cities in the country.

  19. dsawy says:

    A bunch of Phd’s in a close concentration do not mean that wealth creation will result.

    The question that is not being asked here is “PhD’s in *what*?”

    100 Art History PhD’s in close proximity aren’t going to create a whole lot of wealth. Neither are 100 English, Sociologists, “education” Phd’s, etc.

    100 PhD’s in economics probably aren’t going to create a whole lot of wealth when you cluster them either. It took packing only two Nobel-winning PhD’s into one hedge fund to cause LTCM to crater. Packing 100 economics PhD’s into a small space is probably going to result in a lot of bluster, noise and bickering, but not much actual wealth creation. The example of this is Washington DC.

    BTW – the sloppy methodology of the authors is one of the reasons to hold economics PhD’s in disdain.

    San Francisco stands out because of the proximity to Silly Valley. That’s it, full stop. Pull SF out as its own county (which completely contains and defines the borders of SF the city), and you could probably spin this correlation differently than they do. All you need to do is take out the many PhD’s in Palo Alto, Menlo Park, Redwood City down to Morgan Hill. Very few people like commuting from SF down to the Valley, so while there’s wealthy people in SF, the people actually doing the creative work in Silly Valley don’t often live there.

    You’ll find that there’s a very large geographical dispersal of PhD’s that are working in Silicon Valley (some who are telecommuting and living in places like Santa Cruz, La Honda, etc) and live in the suburb area of the valley, and a counter-argument to the “density” argument postulated. They don’t need to live in close proximity as NYC urban advocates fantasize – all we needed for an exchange of ideas was to go eat lunch at “Tom’s” – aka “The Little Garden” on El Camino Real in south Palo Alto. It is no longer there, but it used to be *the* hang-out for techies from PARC, HP Labs, Intel, cisco, etc. Take away Tom’s, sell it off for real estate during the boom and the exchange of ideas slowed down significantly among a group of very bright people.

    And while this is all nice factoids for now, as someone who was an engineer in Silicon Valley, I’m here to tell you the heyday of Silly Valley is over. Some good ideas will still come forth, some capital will still flow towards the Valley, but it will never again be like the period of 1975 to 2000. Think of Silicon Valley in that 20+ years as being like the radio industry in the US from the 20′s through the 40′s. I could elaborate why, but that’s rather off topic from this thread.

  20. purple says:

    As other people mentioned, this chart is basically measuring population density.

    Quite a bit of disdain for liberal arts majors here. Glad to know that Shakespeare and Hemingway didn’t create ‘wealth’.

    This is what happens when lobotimized engineers hold to much sway.

  21. Patrick Neid says:

    This chart has been part of San Francisco’s urban legend for 25 years.

  22. dsawy says:

    Neither The Bard of Avon, nor Hemingway, had a degree in anything to my knowledge, much less a do-nothing PhD degree in liberal arts.

  23. anian says:

    I live on a small island in south Puget Sound – roughly 175 acres, population about 650+/-. The island is connected to the mainland by a bridge. There are appoximately 220 occupied lots, roughly half of which are waterfront. The demographic mix includes a wide range of doctors, lawyers, indian chiefs, butchers, bakers, candlestick makers, restauranteurs, hoteliers, car dealers, retired military, small business owners, a nationally reknown nautical watercolorist, a Hollywood producer, several employees of the Frank Russell company, real estate agents, mid-level managers, etc, etc. Of course there is a neighborhood association. In 1996 the association determined that the bridge would outlive its usefulness in 20 years. Costs for a new bridge were extrapolated to 2016. An additional annual assessment was levied to provide capital. An investment committee was formed to invest the assessments so that all the money needed for the new bridge would be in hand in twenty years. After fifteen years the investment committee populated by a doctor, two lawyers, a dentist, a retired Air Force Lt Colonel (who survived a second career as a bean counter), a retired airline pilot and several other sundry overeducated nitwits has managed (inflation adjusted) to actually LOSE about 1 1/2% per year by prostrating themselves before the altar of Jack Bogle. Although numerous alternative investment strategies were proposed, they were stonewalled, ignored or dismissed out of hand. If only Richard Florida had done just a smidgen of research here he may well have come away with a less idealized conclusion. Florida obviously and erroneously mistakes advanced formal education with advanced common sense. Put a thousand monkeys at a thousand typewriters and they may in time duplicate Shakespeare, but I’ll put my money on gibberish.

  24. That is the opposite of the creative destruction referenced by Florida.

    You ha d a group of people thoroughly entrenched in the status quo. They were consumers of conventional wisdom, no matter how wrong it was . . .

    That is not how a new firm comes into being, new business models and technologies develop.

  25. I agree with some of the comments, especially with the one by @DeDude. My point is that these figures are likely to change depending on what reference you take and that is the reason why sometimes charts like this can be manipulated or used for a certain purpose. This is correct, I have no doubt about it but it is put that way so it can serve the purpose of this specific article. And, to be honest, if this could be seen in another way, in order for the post to show reality, it should include all the possible references and therefor all the different charts.

  26. m1ek says:

    RTP scores very low because its overall population density is very low – it’s one gigantic exurb. And as for Silicon Valley, it may be somewhat more suburban than I would like, but it’s still drastically more urban than the typical entry on the rest of the list.

    Some people’s carefully guarded oxes are obviously getting gored here.

  27. TomS says:

    Has anyone come across education density data (on a per pop basis) on a more granular level (eg ZIP codes)?

    While retirement is still far away I would like to start identifying very rural settings which also have a highly educated population. A year ago I visited friends of a friend in a valley in western Canada. It was stunningly beautiful and remote AND half of its very small population seemed to consist of retired / semi-retired college professors, PhDs. Thanks.

  28. James says:

    At least one potentially serious flaw to the degrees/land area approach is it assumes a uniform distribution of degrees over a region. This strikes me as being inherently flawed since certain neighbhorhoods and regions will most certainly have higher concentrations of degrees, and this could vary significantly from city to city. As an example of how things can go wrong, take city A with 100 sq miles of area and 100,000 degrees, and city B with 200 square miles and the same number of degrees ALL CONCENTRATED IN 100 SQ MILES OF THAT CITY. According to the author city A has a degree density of 1000 and city B has a degree density of 500. Yet, both cities have all of their degrees within a 100 sq mile region, and so both should be equally fruitful.

    This argument comes to mind with regard to Seattle, which is considerably larger (landwise) than Boston and San Francisco. Many of Seattle’s degrees are concentrated in certain areas of the city – a characteristic that is lost using Richard Florida’s degrees/land area methodology.

  29. bdg123 says:

    Only someone with a college degree could come up with that map. I am quite dubious of the rationale behind this analysis. College degrees are by no means an accurate measurement of intelligence that matters. ie, Creativity. I could argue the rote and often tedious topics of college, or for that matter, any dogmatic educational structure is counterproductive to creativity. Other than the sciences where one needs a foundation of skills in mathematics or biology or whatnot, what does college really do to foster inventiveness or creativity? How many artists learn their creativity from college? How many breakthroughs in science? The artist and the scientist capable of breakthrough inventiveness are on in the same. College is simply a tool for that type of scientist. The vast majority of people going to college are clearly incapable of this artistic expression regardless of how many physics classes they take.

    From my experience, college is pretty much worthless outside of the sciences because the curriculum encourages mimicry rather than free thought that is required from an inventive mind. What we do know about the vast majority of college-degreed people is that they are able to learn through mimicry. That is a far cry from creativity. In fact these brain processes are generally mutually exclusive as are the people who excel at one or the other. Or, put another way, most college curriculum could be shown to stifle the free mind. Just as our K-12 curriculum does. It’s also a prime determinant of why so many boys are diagnosed with ADD. W(ho)TF wants to listen to Charlie Brown’s teacher yap all day? That stimulates my cerebral cortex about as much as watching paint dry. Our teaching methods are the problem. Not the kids. We are often most inventive in spite of our educational system because our society has historically embraced individualism.

    What creates a wealthy society is a self-determined society not a college-educated society. One where free thought and free association is embraced as a foundational requirement for the creativity of the individual human mind. Can then people brainstorm on that creativity to come up with business ideas? And is that beneficial? Sure it is. But going to college to then sweat away in some corner office of a bureaucracy is not how creativity or invention is fostered. Except for the basic sciences, most invention or human expression requires no college curriculum.

    What ever happened to the America of entrepreneurs? It has been replaced by the American dream of going to work for someone else. That ‘someone else’ does the vast majority of thinking for us and makes our minds fat, dumb and lazy and ultimately stifles inventiveness. Just as the Soviet Union. The corporatocracy is the Soviet Union of the United States. A college degree used to slave away for the Orwellian corporation makes us stupid if we are to be frank.

  30. HoldYourHorses says:

    Very good point on the lacking degrees per capita, but it still would have to be normalized for population density.

    I have another one, only count employed degrees. Then, see how many are applying their degree, not necessarily explicitly, but a PhD in PhysEd @ $tarbucks shouldn’t. (this is a metric I’ve been curious about for a while). LA might have loads of LibArts, Arts, and Literature/English degrees, but it’s got a lot of waitstaff to hire them.

  31. AndrewBW says:

    Just as a point of reference, the Richard Florida post Barry links actually points to a post by Rob Pitangelo at his blog Extraordinary Observations. Here’s the original source: