But, there are many other factors that produce the silicon valley network effect, in particular access to venture capital, experienced financiers, entrepreneurs and engineers, and proximity to mature (public) technology companies. With so many factors at play, finding a statistical correlation with a single factor such as the quality of the local university is challenging.
NYTimes: ...If Stanford can hatch world-famous companies around Palo Alto, politicians assume, their colleges can, too. But with so many trying to spin universities away from their traditional academic focus into engines of economic development, it is worth considering whether investing in local universities can achieve that goal.
This strategy is based on the view that research done by professors can form the basis for local start-up companies and that the graduates of the university can supply the entrepreneurs and employees.
But advocates should remember an old maxim of economic development: Beware of investing in things that can move. As it turns out, graduates and research ideas both tend to move around a lot.
Subsidizing teaching is problematic as a development strategy because graduates frequently move out of state.
A study by the economists John Bound, Jeffrey Groen and Gabor Kezdi of the University of Michigan and Sarah Turner of the University of Virginia, “Trade in University Training: Cross-State Variation in the Production and Stock of College-Educated Labor,” (http://cat.inist.fr/?aModele=afficheN&cpsidt=15840383) found little evidence of people staying in places because they went to college there.
The more likely smart people are to leave, the more money their state is spending on helping another area’s economy develop. Marc Andreessen, for example, invented the Web browser while at the University of Illinois, but then founded Netscape in the actual Silicon Valley rather than starting a new one in Urbana.
Texas may subsidize science teaching at the University of Texas, El Paso, but the chance that its graduates will stay and transform the local area into Silicon Rio Grande is remote.
So if a state’s subsidies to graduates from a university will not create new Silicon Valleys, how about subsidizing the research? There is no question that academic research has hatched many of today’s booming technology industries. But scientific and engineering ideas also travel quickly.
Recall the 1980’s, when Japanese companies rose to prominence by producing things that had been invented in the United States, like photocopiers, computer memory chips and video recorders.
In a recent study of the determinants of the creation of high-tech firms in different locations, “Movement of Star Scientists and Engineers and High-Tech Firm Entry” (www.nber.org/papers/w12172), Professors Lynne G. Zucker and Michael R. Darby of the University of California, Los Angeles, examined the importance of what they called “disembodied discoveries.”
They looked at such factors as having successful patents at universities or where highly influential science articles had originated. They found little evidence that the ideas helped local businesses any more than businesses in other areas.
For every Stanford in Silicon Valley, there seem to be several Purdues in West Lafayette, Ind., or Cornells of Ithaca, N.Y. — places filled with path-breaking discoveries but with local economies that are seldom seen as the next Silicon Valley.
The one thing the study does find to be consistently associated with high-tech start-ups is the presence of star scientists — not the ideas, which can be copied, but the scientists themselves. This seems to be the one way in which a university can be used as an engine of business growth.