Harvard professor Michael Porter takes a different view.83 He says that the most innovative cities have lots of companies in the same industry, which all learn from each other, drawing on a deep reservoir of specialised local experience. He cites wine-making, medical equipment and tailoring as industries which tend to be concentrated in cities or regions, where any minor improvement by one firm is seen and quickly copied by all the others.
In spite of their contrasting opinions, note that both Jacobs and Porter understand the importance of weak links in the transmission of ideas and innovation. Jacobs stresses the cross-fertilisation of ideas and practices from one industry to another; Porter emphasises the spread of myriad tiny improvements from one firm to another in the same industry. Strong links within the firms would not work the magic of innovation in either case. But Jacobs’ view is more in the spirit of the small-world idea–weak links spanning more diverse industries should facilitate more radical innovation and growth.
So who is right? A group of economists led by Ed Glaeser examined data from 170 US cities over thirty years to find out.84 The evidence clearly supported Jacobs–cities grew faster where there were several industries. Towns dependent on just one industry tended to contract.
It seems that a city which already boasts many different industries provides a hospitable home for new ones, and it can always shift its focus as some industries decline and others rise. Consider the fact that, outside the United States, most innovation and production in information technology tend to happen in the world’s largest and oldest cities–Paris, Moscow and St Petersburg, Tokyo–Yokohama, Shanghai and Beijing, São Paulo–Campinas, Buenos Aires, Mexico City. The only non-US technology innovation centres that don’t conform to this pattern are in the UK and Germany. But on closer inspection, even they cannot be said to go against the grain totally. In Britain, the main centres are the M4 ‘corridor’, which starts just twenty miles west of London, and Cambridge, both of which draw on the capital’s pool of talent and are relatively close to it. In Germany, Berlin was the country’s technological power-house until 1945. It was only then, with Russian and American troops advancing on the devastated capital, that Siemens moved to the safer haven of Munich.
Innovation seems to thrive most wherever there is the greatest diversity–where weak links are plentiful relative to strong links. Glaeser’s study also found that industry grew fastest in those cities with the greatest number of competing firms, and that small firms innovated more than large ones. His team’s research also showed that cities with increasing proportions of people born outside the United States, such as Los Angeles and New York, were the most successful, with wages increasing fastest for all residents, irrespective of whether they were born in the USA or elsewhere. Diversity, the economists concluded, created wealth through innovation.
These findings suggest that innovation is easier where there are coincident networks than where there is one isolated firm. The real generator of invention is not the individual firm, nor even the network of competing firms and their suppliers, but rather a series of overlapping networks. The unit that really matters is the total commercial network in a particular place. The town with a few large companies, such as Detroit, may contract for want of diverse input, for lack of fresh weak links. By contrast, the ‘sundry city’–large, miscellaneous and open to external influences, the incubator of weak links–will always be able to renew itself. Networks proliferate and cross-pollinate; individuals, firms and the city as a whole benefit.
Yet Michael Porter’s view–that innovation comes from firms in the same industry being cheek-by-jowl–also works, in its own way. He formulated this idea after observation of regions and businesses that apparently benefited from specialisation in one industry. Santa Clara County–a half-rural enclave in northern California, better known as Silicon Valley–packs the world’s heartland of electronics innovation into a very small region. The same is true of tailoring (New York, London), medical equipment and services (Boston), high fashion (Paris), fashion shoes (Milan), leather goods (northern Italy), pharmaceuticals (New Jersey), banking (New York, London, Frankfurt), diamond polishing (Antwerp), publishing (New York, London, Frankfurt), Islamic banking (Bahrain), aerospace (Toulouse), clean technology (Copenhagen), gambling (Las Vegas, London, Macao), racehorse training (Newmarket) and fine watchmaking (southern Switzerland). Firms that are located in these magically concentrated regions tend to do much better than firms elsewhere.
So why does Porter’s analysis work, and yet diverse cities work even better? Part of the answer lies in perspective. From the viewpoint of an industry, regional concentration is good; but from the viewpoint of a city, industrial diversity is better. In both cases, we are talking about networks driven by face-to-face weak links.
It seems remarkable, given the alleged ‘death of distance’, that certain very profitable industries are still concentrated in very small geographical enclaves. In Silicon Valley, success is local, physical, personal, contagious. Four-fifths of the world’s revenues and profits in financial services come from a few square miles in Manhattan, Frankfurt and the City of London. Half of all US breakthroughs in pharmaceuticals are made in New Jersey, which is home to only 3 per cent of Americans. In the UK, a large majority of the country’s fine jewellers are found in a few short streets in Hatton Garden, London.
So why is this the case? Back in 1890, the British economist Alfred Marshall provided an answer that still rings true today:
Great are the advantages which people following the same skilled trade get from near neighbourhood to one another. The mysteries of the trade become no mysteries, but are as it were in the air…Good work is rightly appreciated, inventions and improvements…have their merits promptly discussed: if one man starts a new idea, it is taken up by others and combined with suggestions of their own; and thus it becomes the source of further new ideas.
85
Note: innovations are promptly ‘discussed’. For all the faxes, emails, phone calls and videoconferencing we have today, the key element in the spread, amplification and adaptation of good new ideas is face-to-face discussion, just as it was a century ago. Knowledge is ‘in the air’. This doesn’t mean cyberspace; it means very local air. Anyone who has worked in Silicon Valley knows how apt Marshall’s description is. People learn most from the other people they meet, and most of the time it’s unconscious, natural and almost unavoidable–it happens at the gym, on the train, at the bar, in the back yard, in hundreds of casual encounters and conversations each week. It happens through weak links, through chance encounters, through meeting new contacts in other firms, because the local community is a hive of mutual learning and common aspiration. As Manuel Castells says, ‘Silicon Valley kept churning out new firms, and practicing cross-fertilisation and knowledge diffusion by job-hopping and spin-offs. Late-evening conversations at the Walker’s Wagon Wheel Bar and Grill in Mountain View did more for the diffusion of technological innovation than most seminars in Stanford.’86
But Silicon Valley was not America’s original centre of high-tech excellence. That was the area along Route 128 outside Boston, where such firms as Digital Equipment, Apollo Computer, Prime Computer and Wang Computer set up shop. However, many of the Route 128 companies were acquired or went bust, and the area lost its pre-eminence in computer design and manufacturing as firms such as Intel, HP, Sun Micro Systems, 3Com, Silicon Graphics and Cisco all based themselves in Silicon Valley.