What is a Cluster?

According to Porter (who I was lucky enough to have as a visiting professor during an M.B.A. program that I started in the late 1970s while studying at the Universidad Técnica Federico Santa María in Santiago, Chile) in his November-December 1998 Harvard Business Review article on “Clusters and the New Economics of Competition,” a cluster is a geographic concentration of interconnected companies and institutions in a particular field.
Clusters include governmental and other institutions (such as universities, standards-setting agencies, thinktanks, vocational training providers and trade associations) that provide specialized training, education, information, research and technical support.
“A cluster’s boundaries are defined by the linkages and complementarities across industries and institutions that are most important to competition,” Porter wrote. “Clusters promote both competition and cooperation. Rivals compete intensely to win and retain customers. Without vigorous competition, clusters will fail.”
According to Porter, clusters affect competition in three ways by:
1. Increasing the productivity of companies based in the area
2. Driving the direction and pace of innovation
3. Stimulating the formation of new businesses, which expands and strengthens the cluster
Porter also recommends that governments working with the private sector should promote cluster formation by reinforcing and building on existing and emerging clusters rather than attempting to create entirely new ones.
Another perspective on clusters (not just biotech here) was recently published by Forbes in its May 26, 2003 article on “Cluster Creation” by Ross Devol, who is the director of regional and demographic studies at Milken Institute . According to Devol, there are four key ingredients in successful clusters:
1. Innovation
2. Human capital
3. Globalization
4. Dynamism
With reference to innovation, knowledge is shared more efficiently in close proximity. The resulting economic activity tends to cluster geographically and is less likely to escape to other regions.
A leading measure of this is patenting activity. Devol says Rochester, N.Y. (Kodak land) is No. 1 among the nation’s top 50 metropolitan areas with 2.3 patents per thousand workers versus 1.5 in the San Francisco Bay area and 0.8 in San Diego. Another key metric is academic R&D relative to total employment . In this category, the leaders are Raleigh-Durham, Boston and Greater Washington, D.C.
In the realm of human capital, talent and experience draw ambitious companies. A composite score based on a weighted average of advanced and bachelor’s degrees ranks Greater Washington, D.C. as the most educated metro area in the U.S. followed by San Francisco, Denver and Minneapolis.
Another metric is percent of the workforce (managerial, professional and technical jobs) with the resultant ranking Washington, D.C. (48 percent); Hartford, Conn. (46 percent); and Minneapolis-St. Paul (45 percent).
In globalization, success turns on the ability to exploit international opportunities. Two metrics here are the share of recent immigrants as a percent of the working population and the export sales per manufacturing employee. In the first metric, San Francisco is the leader (26 percent) followed by Miami (17 percent), New York (15 percent), Los Angeles (12 percent) and San Diego (5 percent).
In the latter metric, Seattle is the leader ($129,000) followed by Miami ($127,000); Richmond, Va. ($86,000); and San Francisco ($80,000).
The dynamism ingredient focuses on the ability to start and sustain companies. Although the number of new businesses created is a good metric, Devol states that a powerful measure is the number of IPOs per $10 billion of gross metro product. The leader here is San Francisco followed by Seattle, West Palm Beach and Boston.
Another metric is the number of jobs in fast-growing companies as a percent of total employment. The leaders here are Orlando, Las Vegas, Charlotte and San Francisco.
If we cross the above with the Brookings study, there certainly are some common metro areas that stick out. Still, there are a lot of surprises. Also, we must take into account that much of the Brookings study was based on data accumulated during the late 1990s, which was then analyzed and from which conclusions have been drawn.
But what has happened since? According to the Forbes article, the bad news is that the biotech clustering effect around the leading centers has concentrated even more, which we can see below:
City
Life Science Firms
Employees
Revenue
VC Money Raised
SF Bay Area
819
85,950
$4.1 billion
$1.2 billion
San Diego
499
29,500
$1.8 billion
$0.5 billion
Boston Area
485
47,370
$6.7 billion
$0.4 billion
However, you pay a price for operating and living in these areas (or do you?). An April 15, 2003 ranking by Genetic Engineering News of total annual costs to operate a biotech company’s facility looked at the cost of operating a 100-worker biotech R&D center for 48 American cities:
Rank Biotech Facility Location Annual Operating Cost % vs. Leading Area

1    San Jose, Calif.          $12.1 million          100%
2    San Francisco             $11.9 million           98%
3    Fairfield County, Conn.   $10.7 million           88%
4    Boston                    $10.6 million           88%
5    Nassau/Suffolk Cty, N.Y.  $10.3 million           85%
6    Princeton, N.J.           $10.2 million           84%
7    Chicago                   $10.2 million           84% 
8    Philadelphia              $10.1 million           83%
9    San Diego                 $10.0 million           82%
10   Middlesex/Somerset, N.J.  $10.0 million           82%
19   St. Louis                  $9.2 million           76%
22   Cleveland                  $9.2 million           76%
28   Madison, Wis.              $9.0 million           74%
29   Milwaukee                  $9.0 million           74%
30   Kansas City                $8.9 million           73% 
33   Raleigh-Durham, N.C.       $8.7 million           72%
34   Columbus, Ohio             $8.7 million           72%
35   Des Moines, Iowa           $8.7 million           72%
36   Minneapolis-St. Paul       $8.7 million           72% 
48   Montreal, Canada           $8.0 million           66% 

Note: Cities displayed in red are in the Midwest

Is Chicago and the Midwest a Cluster?

In Loffler’s presentation, the argument was presented that Chicago doesn’t yet have a biotech cluster. During the Q&A afterward, I suggested that our relevant biotech cluster was not the city of Chicago but the entire Midwest, which when lumped together would appear to have the critical mass seen in other areas.

She responded that the cluster must be geographically concentrated to have the positive cluster effects and that by driving several hours or flying by plane to another area defeats the benefits of this concentration (the interplay of both competition and cooperation that increases productivity).