Last year a paper [pdf] by the Kauffman Foundation, the leading entrepreneurship think-tank in the US, showed that research universities and other postsecondary institutions are important for metropolitan entrepreneurship, but are not the key catalyst in spurring such activity. Instead, the most fertile source of entrepreneurial spawning is "the population of existing companies, which has implications for economic policymaking and economic development strategies." Meanwhile, the received wisdom on clusters, that physical proximity within city-regions is key for innovation, is also under attack.

Michael Porter, the Harvard Business School academic and author of the 1990 book, 'The Competitive Advantage of Nations,' has defined a cluster as, "A geographically proximate group of interconnected companies and associated institutions in a particular field, linked by commonalities and complementarities (external economies)."

Prof Porter told BusinessWeek that "the more there are no barriers, the more things are mobile, the more decisive location becomes. ..Now that globalisation continues to power forward, what has happened is that clusters must become more specialised in individual locations. The global economy is speeding up the process by which clusters get more focused. There is a footwear cluster in Italy, for example, where they still produce very advanced products. The design, marketing, and technology still are in Italy. But much of the production has shifted to Romania, where the Italians have developed another cluster. All of the production companies actually are Italian-owned. Taiwan has done the same by shifting production to China. The innovation is in Taiwan, but its companies are moving aspects of their cluster that don't need to be in Taiwan."

However, Vivek Wadhwa, an Indian-American technology entrepreneur and academic, writing in The Washington Post, dismisses Porter's "outdated cluster theory," which "lies at the heart of what is wrong with these common prescriptions. He observed that geographic concentrations of interconnected companies, specialized suppliers, and service providers gave certain industries a productivity and cost advantage. His legions of followers postulated that by bringing these ingredients together into a “cluster,” regions could artificially ferment innovation. They just needed to build the right infrastructure and bring together chosen industries."

Prof Wadhwa adds:

Most of the top-down cluster-development projects in the United States and around the world have died a slow death in relative obscurity. Politicians who held the press conferences to claim credit for advancing science and technology are long gone. Management consultants have cashed in their big checks. Real estate barons have reaped fortunes, and taxpayers are left holding the bag."

Richard Bruton, Ireland's enterprise and innovation minister, who is an avid consumer of what  Prof Wadhwa calls "snake oil," remains in place for the time being, relying on faith rather than evidence. 

Wadhwa says that a recent analysis of 1,604 companies in the five largest Norwegian cities underscores what’s missing from this prescription for a knowledge economy: people. The prerequisite for a regional innovation system is knowledgeable people who have the motivation and ability to start ventures. To succeed, these people need to be connected to one another by information-sharing networks. Basic infrastructure is always needed, but fancy science parks and big industry are just nice to have.

He adds that the study, conducted by Rune Dahl Fitjar, of Norway’s Centre for Innovation Research at the International Research Institute of Stavanger, and Andres Rodriguez-Pose of the London School of Economics and Political Science, found that the key drivers of innovation in Norway are the communication channels that local entrepreneurs maintain to the outside world and their open-mindedness toward foreign cultures, change and new ideas. Companies that are “regionally minded” - - that maintain ties only with players within the same cluster - - are four times less likely to innovate than the globally connected. The study found that regional and national clusters are “irrelevant for innovation.”