The 14th Annual International Convention of the US Biotechnology Industry Organization (BIO) was spectacularly big. At this year's meeting, held in Chicago between April 9 and 12, the 176,000-net-square-foot exhibit hall housed no less than a 1,000-square-foot indoor transgenic cornfield (nearly as much transgenic crop as in the whole of the European Union), an Indy Racing League racecar with accompanying driver, a Ford F-150 Flexible Fuel Vehicle, a Harley-Davidson motorcycle and exhibitors from 50 US states and 62 nations.

The conference has turned into something of an Olympian schmooze endurance test for the 19,479 delegates. For the business forum alone, according to official statistics, 4,260 people representing 1,476 companies participated in 11,018 meetings. Nature Biotechnology is not quite sure how the other 15,219 delegates spent their time.

Viewed through BIO's rose-tinted glasses, biotech is booming in every US state, every Canadian province, each Swiss canton, German Länder, British shire, and French département and city, and in every newly economically progressive nation of the former Eastern bloc or East Asia. There is BioValley, BioVallé, BioVale, BioDale, BioPole, BioPolo, BioCity, BioUrbis and BioVillage. And growing bigger clusters is easy: just take a map, draw a ring round the existing ones and call it a mega-cluster—Biocountry, Biobloc or Biocontinent, perhaps, or the Council of European Bioregions.

The question is, with so many clusters vying for biotech, is there enough business to go around? According to a report released at the conference by the nonprofit organizations Battelle Memorial Institute and the State Science & Technology Institute (SSTI), the answer is certainly “yes.” Growing The Nation's Biotech Sector: State Bioscience Initiatives 2006 reckons that in the United States, the 'biotech industry' comprises more than 40,000 companies employing 1.2 million people.

The reality is less impressive. Biotech globally is hardly profitable, and whatever profits and revenues are made are generated by a few hundred large companies, most of which are in the United States. The vast majority of companies are loss-making basket cases, their ability to contribute to a national or regional economy heavily dependent on external finance from investors or government.

The point is not to belittle biotech's economic potential, but simply to be realistic. Biotech companies may aspire to become Amgen and bioregions may aspire to be San Diego or Boston. But those aspirations cannot be achieved by emulation. The next 'Amgen' cannot afford to make all the mistakes Amgen did. And just because the biotech industry grew as a cluster in San Diego, it doesn't mean that it has to do that everywhere from here to eternity.

Thirty years after the foundation of Genentech, venture capital with expertise in biology is not scarce, growing companies from scratch is not a journey into the unknown, our understanding of biology is better, and electronic communication and information exchange are routine and cheap. In short, the need for close geographic networks is diminishing.

Biotech clusters of the near future are going to be virtual. They will be built on shared needs and proximity of interests, not merely on neighborliness. BIO, and similar meetings elsewhere, should be where these networks are initiated, not merely where the biggest clusters are put out for display.