July 19, 2024

Why Do Innovation Clusters Form?

Recently I attended a very interesting conference about high-tech innovation and public policy, with experts in various fields. (Such a conference will be either boring or fascinating, depending on who exactly is invited. This one was great.)

One topic of discussion was how innovation clusters form. “Innovation cluster” is the rather awkward term for a place where high-tech companies are concentrated. Silicon Valley is the biggest and best-known example.

It’s easy to understand why innovative people and companies tend to cluster. Companies spin out of other companies. People who move to an area to work for one company can easily jump to another one that forms nearby. Support services develop, such as law firms that specialize in supporting start-up companies or certain industries. Nerds like to live near other nerds. So once a cluster gets going, it tends to grow.

But why do clusters form in certain places and not others? We can study existing clusters to see what makes them different. For example, we know that clusters have more patent lawyers and fewer bowling alleys, per capita, than other places. But that doesn’t answer the question. Thinking that patent lawyers cause innovation is like thinking that ants cause picnics. What we want to know is not how existing clusters look, but how the birth of a cluster looks.

So what causes clusters to be born? Various arguments have been advanced. Technical universities can be catalysts, like Stanford in Silicon Valley. Weather and quality of life matter. Cheap land helps. Some argue that goverment-funded technology companies can be a nucleus – and perhaps funding cuts force previously government-funded engineers to improvise. Cultural factors, such as a general tolerance for experimentation and failure, can make a difference.

Simple luck plays a role, too. Even if all else is equal, a cluster will start forming somewhere first. The feedback cycle will start there, pulling resources away from other places. And that one place will pull ahead, for no particular reason except that it happened to reach critical mass first.

We like to have explanations for everything that happens. So naturally we’ll find it easy to discount the role of luck, and give credit instead to other factors. But I suspect that luck is more important than most people think.


  1. Video Converter OS X is currently the most powerful converter for mac os x users which allows you to convert video file between all popular video formats such as convert FLV to 3GP, DAT to 3GP, MOV to MPEG, AVI to MOV, WMV to MP4, H.264 video.

  2. One recent suggestion I saw somewhere was that Silicon Valley, in particular, got started because the inventor of the junction transistor set up shop there, then was a real bozo (and a racist!). Experts from far and wide would come to work for him, then quickly decide to leave and start their own business nearby or join one of the existing ones. Drawn in and then dispersed to the surroundings by a long range attracting, short range repelling individual.

  3. Another important reason can be immigrants and how they tend to congregate to a certain area due to family and friends network. Innovation centers attract highly educated immigrats who in bring along their acquaintences. Of course, money is also a factor.

  4. Well, I wouldn’t quite say that if funds are not sufficient then *nothing* can be done. For example, thanks to partial funding, the particle physicists community once got a huge, ten-billion-dollar hole dug in Texas to go jump in or something. Not exactly what they’d originally envisioned, mind you. 😉

  5. I agree that money is a big factor in promoting innovation. So is the ability of people to “flock together” in ways that increase the formation of companies and groups.

    Coincidentally, I addressed some related issues on one of my own recent blog posts titled “Comparing Eras of Innovation: 1950’s Aerospace Advances and Today’s Web 2.0” (http://www.ddmcd.com/aerospace.html). There I compared and contrasted today’s “web 2.0” phenomena with post WW II aerospace innovation. Government was very important following WW II both in funding and in guiding innovative developments in aerodynamics, jet propulsion, electronics, and related areas.

    Today’s Web 2.0 world doesn’t necessarily have the same Government boost but rather is a reponse to market opportunities and the ease with which new developments can be created and widely disseminated.

    In both cases, though, “innovation” can be a very messy process. For all the successful aircraft and technologies that came after WW II, there were many failures as well — aircraft that never progressed beyond the prorotype stage, jet engines that never delivered necessary power or reliability.

    We see the current proliferation of tools on the Web also resulting in a number of failures. The list of successful and monetized applications will look very different 12 months from now than it looks today.

  6. to my experience, luck are just the factors that we cannot or don’t want to see. and don’t bring up quantum mechanics now … please.

    didn’t read the other replys.
    sry if i restated something that already has been said.

  7. Well, if we take as given that:
    a) Innovation is a positive feedback process (innovation draws students, tax breaks, other innovators, everything else discussed here)
    b) Power laws are created by choice and more choice generates steeper power laws
    c) The average participant in a power law situation ends up below average

    Then clusters will happen whether we want them to or not and generally whoever gets ahead is going to get way ahead.

    So, I’m with Ed, although I’d phrase it slightly differently. We can spend all kinds of time measuring correlations but not get any closer to being able to predict where or why clusters will form. Instead, we should look at what factors encourage innovation — which, of course, has been done: diversity, low cost of learning, lots of boundaries between networks — and then hope you get lucky.

  8. Paul Graham wrote an essay about this that you might find interesting:

  9. In the United States the government creates innovation clusters by a combination of research spending and, usually military, contracting. Even the Continental Congress, during the Revolutionary War, was funding the likes of Eli Whitney to develop guns with interchangeable parts which led to new levels of accurate machining using what was called “armory practice” and turned the Connecticut river valleys into hotbeds of innovation. Guns, clocks, tools, rockets, helicopters, submarines and a host of other machines and parts were developed in this area.

    Dayton, the center of army procurement, and still an important part of logistics command, was the hot spot in the late 19th century and early 20th. Government contractors and those who wanted to be developed ball bearings, airplanes, high torque electric motors, and other gizmos were developed directly and indirectly for government use.

    More recently, Moffett Field became the great center for the computer revolution, and once again it was the government as the big customer. The government had been buying computing machines since the 1870s, and after the war, they began to pump money into the system. Take a look at who the original customers were.

    It is actually hard to find an innovation center in history without government backing. Consider the German research universities and their chemistry and metallurgy, the British Navy doing basic research on chronometers, assembly line production and saltine crackers, the shipyards at Piraeus or Venice, or the great library of Alexandria and its community of scholars. There had to be a customer with a long time frame and a tolerance for failure. That was generally the government, because governments expect to be around for a while, and they can command the resources to make it happen.

  10. To seed an innovation cluster you need the right infrastructure (Physical infrastructure, People, Money) and some innovative entrepreneurs that take the risk. When the cluster is formed it starts attracting more people and more money: If there is a perception that place A is 5% better than place B, that can make the difference between a profit and a loss. So everyone will cluster towards A, bringing in money, encouraging investments in infrastructure.

    Yes, there are better and worse places for innovation; but it is hard to trace it back to the specific individuals that started a specific cluster in a certain region and explain why they succeeded and their competitors in other regions failed.

  11. I agree on the luck part. “The right place at the right time” is just a different name for it. An important thing to note is that luck is not a causative, but merely an enabling factor.

    None of this means that effort, determination, talent, etc. are not important. Those are the causative factors the effectiveness of which is enabled by luck.

    What we are trying to do in the rational paradigm is to explain everything in terms of causative factors only.

    Note on “luck” — depending on one’s philosophical leanings one can consider it nondeterministic “random” chance, or deterministic and unpredictable because of sufficient complexity, or unknowable “starting conditions” that are the inputs to the determinism. The important part is that it is practically unknowable before and after the fact. Well, that’s what makes it luck.

  12. Blind luck is a big part, as is “the right place at the right time.” Everybody talks about high tech, but there are other “clusters” out there. I happen to live in Houston, Texas. The big business here is “energy” (oil and gas, exploration and refining, etc.). Pretty much all of the “white collar” aspects of the energy business are well represented or headquartered in Houston, with a non-trivial chunk of the “blue collar” industry close nearby.

    I recently met a hedge fund manager who trades energy stocks. He could live anywhere in the world (so long as he has a fast Internet connection), but he chose to move to Houston so he could be closer to the companies he trades. Meetings are easier to arrange. Specialized skills are closer at hand.

    Did Houston do anything, in particular, to create an attractive business climate? Did Houston give great tax incentives or have an amazing climate or environment? Hardly! [1] Instead, Houston is simply the logical place to be if energy is your business, in precisely the same way that Silicon Valley is the logical place to be for other things. The challenge for an aspiring cluster is to figure out the industry for which you’re a “logical” place to be.

    [1] For an amusing, decidedly unofficial campaign for why you (yes you!) should come to Houston, check out http://houstonitsworthit.com

  13. Rich Gibbs says

    Let’s try again, to see if I can get the URL in there.


  14. Rich Gibbs says

    Paul Graham has an essay related to this topic, “Why Start-Ups Condense in America”. which might be of interest:

  15. I agree that many of these ideas are factors in the forming of clusters. What is important to point out is that the formation of dense clusters of similarity is not limited to the innovation culture. For example, where I live the population is largly outdoorsey, democratic, hippie people that like to sit around and play music (none of this is good or bad, its just the average demographic).

    Many factors influence the development of population clusters.

    I would also like to point out that the internet is cutting down on some of the clusters. Obviously the ability to connect with like-minded people across the internet means less of a requirement to live near them — nevertheless, clusters will continue to form as the enviroment will always be conducive to the behavior of the people in the cluster. The internet also adds a new clustering capability which has much smaller barriers of entry and exit. http://www.MakeZine.com is one such online innovation cluster.

  16. enigma_foundry says

    One of my kids’s soccer coaches did something venture/investment-related, and he explained one reason why innovation clusters sometimes fail to form. States like Florida and Texas have (or had, before changes in bankruptcy law),

    I think these effects are under recognized. I would add the stories of several Polish immigrants I’d know (one photogrpher, and two Architects) who had chosen to start their businesses in Canada because of the Health Care uncertainty in USA..

    • Newsweek helpfully includes a list of selective schools that would have made the list but were barred due to SAT scores. This excluded-schools list is topped by a mind-bending caption: (student coursework)

  17. enigma_foundry says

    There are many reasons why clusters form, and I’d suggest the following article which is a case study, but has some insights as well as some good maps which give an interesting perspective on how these clusters evolve. And how they evolve suggests answers to they “Why”

    Scott, Allen J., “High Technology Industry and Territorial Development: The Rise of the Orange County Complex, 1955-1984.” Urban Geography, 1986, 7, 1, pp. 3-45.

    I feel that the support for basic research as well as the development of a market for something are the two essential ingredients. Think of markets in terms of labor, as well as products. Certainly once a market starts to develop, certain types of business see an advantage in participating in the market, for being left out would be problematic.

    A more pertinent question for many American metropolises is ways to keep these clusters from disintegrating. This is may be happening now in Detroit, and it has happened already in East Saint Louis, IL.

    Also highly recommended is the following book:

    Scott, Allen J. Regions and the World Economy: The Coming Shape of Global Production, Competition, and Political Order. New York, NY: Oxford University Press, 1998.

  18. Trevor Pascal says

    I think a single entity (person or organisation) with ideas with potential are the initial catalysts. Once these ideas have sufficient backing from persons or organisations with “influence” then it is only a matter of time before the money and infrastructure follow.

    Other reasons include physical and human geography. Physical geography-how conducive is the physical environment for the formation of a “settlement”, including the quality and capacity of basic infrastructure e.g. electricity, telecommunication links, water, transportation, ports, etc.
    Human geography-quality and to a lesser extent, quantity of skilled human resource which could also be stated as “a conducive demographic profile to facilitate the innovation cluster”.
    Cultural and social acceptance of the “idealogy” of the innovation cluster. This point is important in the birth and subsequent growth, maturity and eventual demise of the cluster.
    I disagree with the “luck” point though because what the uninformed see as luck is what innovators and “bigthinkers” seat and plan looking at the causes and effects.

  19. Seth’s point on funding for basic research has been made many times. But the culture in which the people interact is pretty important as well. An underlying openness and energy (does the Bay Area through the 60s and 70s qualify?) will both attract people who share those ideas — people who are not only unafraid of the future but want to invent it — and motivate those who are there.

    The people you really need are the ones who can see the ideas in the research findings or components and can convince others to help fund/build them (the Fab Four @ Sun, Jobs/Wozniak, etc.). A culture that says yes more than no and tries things without worrying about success or failure. Garrison Keillor riffed on this a couple of weeks back, reminding us that the iPod wasn’t invented in Omaha: there are reasons for that. There are people in Omaha just as there are in San Jose and Cupertino and Redwood City. 

    I think a broad spectrum of social liberalism and tolerance, funding in research as opposed to specific products, and a light hand on the management of that gets you close to it.

    • “Public schools are ranked according to a ratio devised by Jay Mathews: the number of Advanced Placement and/or International Baccalaureate tests taken by all students at a school in 2004 divided by the number of graduating seniors.” (Art Coursework)

  20. Seth, I disagree. While money can be a factor, it isn’t really money that starts it off. It is really much more about the people involved. Especially in the tech industry, but in other industries as well, the money and the jobs seem to follow the people. Huge amounts of money poured into a University, a business, or a community often do not generate the results that we are talking about here.
    Motivated people innovate and create new products, thus generating new jobs in the relevant field. This attracts more investment money to that field and area of the country. This in turn, generates more jobs in that field. More and more people with the needed knowledge will move to that area and live there. Since many people with the technical knowledge needed live there, more businesses are likely to start there or move there.
    But it all starts with the individual people initially. Money could be the motivating factor/cause behind those initial innovators congregating there, but it often isn’t. I wouldn’t say it was luck either. All I’m saying is that there are many possible reasons why they initially congregated there. The only constant, is the people.

  21. dr2chase says

    I think you’re supposed to reference
    Annalee Saxenian‘s

    One of my kids’s soccer coaches did something venture/investment-related, and he explained one reason why innovation clusters sometimes fail to form. States like Florida and Texas have (or had, before changes in bankruptcy law) a strong shield for “your home” in bankruptcy. You can insulate yourself from risk by sinking a lot of money into your “home”, and for someone putting money into a company, this is regarded with great suspicion. There was a guy in Texas who spent some time claiming that the multistory building he owned was his “home”, and thus he was entitled to shield all of it from his creditors. So, given a world full of places to invest, they invest elsewhere.

  22. I suspect the key factor is not luck, but HUGE amounts of money for basic research.

    This is a bit trickier than it sounds, since the money can manifest itself in various ways. It can be a technical university that has a lot of grants. It can be a government project for another goal. It can even be the rare large corporation which sets up a big lab for long-term R&D.

    Since many, though not all, such huge pots of money are government public spending, I’d expect certain factions at such conferences would strongly resist this explanation (the Cato/PFF/Liberbabbling types especially so).