Creating Jobs in Clusters
David Nicklaus
Of the Post-Dispatch

Economic development officials here are embracing the concept of industry clusters -- geographic concentrations of interconnected companies. While St. Louis has a few established clusters, two others -- plant and life sciences and information technology -- are emerging. Officials seek to build on the region's strength in those areas and make St. Louis attractive to promising young companies.

Twenty years ago, economic development largely meant chasing smokestacks. Throw enough tax breaks at a big company and it would put a factory with hundreds of jobs in your town.

But along the way, the process got very expensive. In the 1980s, Illinois spent $250 million, or $86,000 per job, to lure a Mitsubishi auto plant to Normal. And some cities learned the hard way that, once the incentives expired, the plant for which they paid so dearly could just close up shop.

Cities still compete to lure big corporate prizes, of course -- witness the recent four-city tournament for Boeing's corporate headquarters, or the 1998 contest in which St. Louis beat Dallas to land a new MasterCard International campus. (It's being built in O'Fallon, Mo., with $40 million of state loans and tax credits.)

But most economic development organizations, including the St. Louis Regional Chamber and Growth Association and the Missouri Department of Economic Development, say they've learned that it's more cost-effective to grow your own jobs. Putting public money into things such as infrastructure and venture-capital formation can make the area attractive to a plethora of promising young companies.

But what kinds of infrastructure for which kinds of companies? To focus their efforts, officials at the RCGA -- and in many other cities -- have turned to an academic concept called industry clusters.

A cluster -- a concept popularized by Harvard business professor Michael E. Porter -- is a geographic concentration of interconnected companies. The very existence of a cluster gives the companies in it a competitive advantage, Porter theorizes. If you wanted to start a mutual fund firm in Boston, home to giant Fidelity Investments, you could tap into a network of lawyers, accountants and other service providers who understand the industry. If you wanted to start the same firm in the middle of South Dakota, such expertise would be harder to come by.

A firm in a well-established cluster also has an easier time finding workers with industry-specific skills. "You want to locate in the same area if you're all using the same skills, so you can get the local community colleges to educate people in those skills," said Jackson A. Nickerson, assistant professor of organization and strategy at Washington University's Olin School of Business.

But while local officials try to build clusters, Nickerson said, they should keep their eyes open for innovative firms that want to be in an area with a good quality of life, but don't want competitors hiring away the best people. "If you're particularly good at producing workers with specialized skills, you may not want to be part of a cluster," he said.

Five clusters in the region: Three in place; two emerging

In St. Louis, the RCGA has decided to focus its efforts on five industry clusters. Three of them -- advanced manufacturing, financial services, and transportation and distribution -- are established clusters that already have a critical mass here. Think of the Boeing Co. aircraft plant, the headquarters of A.G. Edwards Inc. and Edward Jones, and the TWA hub at Lambert Field.

The other two -- plant and life sciences and information technology -- could be classified as emerging clusters. Each accounts for about 21,000 jobs, which is just 1.6 percent of the St. Louis work force.

Walter Plosila, Battelle vice president in charge of St. Louis cluster studies

The Battelle Institute, a Cleveland-based research group, completed a study of St. Louis' plant and life sciences cluster last year and is now studying the local information technology industry.

In both clusters, St. Louis has plenty of competition. Plenty of cities would like to duplicate the success that San Diego and Baltimore have had in building biotechnology industries, and even more places would like to think of themselves as the next Silicon Valley.

But Walter Plosila, the Battelle vice president in charge of the St. Louis cluster studies, says developing a cluster doesn't mean copying another city. "People are not, in fact, duplicating each other and doing the same thing. It's all about niches," he said.

Capitalizing on plant science strength

In life sciences, for example, most other cities are focusing on drug-discovery companies. While several promising companies have been founded based on medical research at Washington University and St. Louis University, plant research appears to be St. Louis' true claim to fame. The Battelle study says that with Monsanto Co., the Missouri Botanical Garden and the Donald Danforth Plant Science Center, St. Louis is already the nation's leading center of plant science research.

To build on that strength, the study says, St. Louis needs to do several things:

* Polish its image as a center of life sciences research.

* Develop locally run seed-capital and venture-capital funds that invest mainly in life sciences companies.

* Develop more laboratory space.

* Start educational programs to produce workers with the skills that biotech companies need.

* Make quality-of-life improvements, including a rejuvenated downtown, so St. Louis will be more attractive to talented scientists.

* Try to attract a federal research institute for food and agricultural science.

Plosila said he's impressed with what has happened in St. Louis in the nine months since he delivered his life sciences study. "The St. Louis community really is mobilizing around the strategy," he said.

Within weeks after the study came out, St. Louis Community College at Florissant Valley announced a new educational program offering a certificate in biotechnology. The RCGA has launched a marketing campaign to brand the region as the BioBelt and has put together a group of business and academic leaders called the Coalition for Plant and Life Sciences.

Even in a difficult money-raising climate, local institutions have launched one new seed-capital fund, the state-backed Prolog Ventures, and two new venture-capital funds, Oakwood Medical Investors and RiverVest Venture Partners. Those three funds total nearly $75 million. At least two other groups are still trying to raise money, and if they all hit their targets, the total will far exceed the $100 million venture-capital goal set in the Battelle study.

Developing lab space is proving to be a thornier problem. The area has two technology incubators, the Center for Emerging Technologies and the Nidus Center for Scientific Enterprise, but almost no lab space available for companies that can't get into the incubators or have outgrown them. In areas like San Diego, where the biotech industry is better established, real estate developers build lab space because they know there's a market for it.

"That space is important," said Donn Rubin, executive director of the Coalition for Plant and Life Sciences. "If the companies are ready to graduate from the incubator and that space isn't available, those companies are going to go someplace else."

He said the coalition is talking to developers about the need for lab space and is looking at innovative approaches used in other states. Maryland, he said, provides tax credits to developers for such projects, and New Jersey has a technology-financing authority with eminent domain powers.

The biggest potential headline-grabber in the life sciences study -- a federal agricultural research institute, modeled on the National Institutes of Health -- is little more than a dream at this point, although local officials hope it's an idea that will be taken seriously in Washington.

"We are in the process of developing a strategy, testing the waters, developing some potential champions in Congress," Rubin said. "It would be tremendously important in terms of the image of St. Louis in plant sciences and the number of scientists it would attract here. But it is indeed a long-term strategy."

Computer literate work force is a resource

St. Louis' assets in information technology aren't as prominent as those in the life sciences. Washington University does some important networking research, and the area has seen a flurry of start-up activity in the telecommunications and Internet fields, but St. Louis doesn't have any large, well-known hardware or software firms.

What it does have, partly because of the number of large corporate headquarters in town, is a relatively computer literate work force. Battelle counts 45,000 St. Louis workers in information technology occupations, even though more than half of them work for nontechnology companies. That's 3.5 percent of the regional work force, compared with a national average of 3.7 percent.

But Battelle says St. Louis has an above-average concentration of the highest-skilled technology occupations. The concentration of computer systems analysts here is 68 percent above the national average, and the concentration of software engineers is 31 percent above average.

Outside the top technology centers such as Silicon Valley or Boston, Plosila said, the opposite is usually true: Highly skilled people tend to be rare, and most of the technology work force consists of lower-skilled occupations such as programmers.

"It's the legacy of the communications industry in the region," Plosila said, particularly the fact that SBC Communications Inc. had its headquarters here until 1993.

The failure of several technology start-ups over the past year has created an "I told you so" atmosphere in St. Louis, where the traditional measure of success is landing a good job with a big company, not getting in on the ground floor of a successful tech company. Still, Plosila said, more than 20 percent of the region's technology jobs are at firms that are 5 years old or younger.

"It does suggest that there's some dynamic growth still going on," he said. "You hear about the firms going under, but growth that's spread around, you don't hear about so much."

The information technology cluster is valuable not only for its own growth potential, but also for the links it has with other clusters. Advanced manufacturing, for instance, depends on sophisticated computer software and hardware, as well as on other technologies such as lasers.

Leadership in bioinformatics

And biotech couldn't exist without info-tech. Researchers in genomics -- including the Danforth Center's work on plants and Washington University's work on the Human Genome Project -- generate vast amounts of data that must be stored and sifted through.

That means computing resources that are truly cutting-edge. The Danforth Center's computer has 1,040 processors and is capable of doing 335 billion mathematical operations per second.

The computer at Washington University's Genome Sequencing Center isn't as fast, but is one of the top two or three academic installations in the world, said Jim McCarter, a postdoctoral fellow at the center. The only other data centers that compare, he said, are at the Massachusetts Institute of Technology and at Cambridge University in England.

Washington University has had a program in computational biology -- a discipline that has come to be known as bioinformatics -- for more than a decade. "Other places are scrambling to catch up," McCarter said. "It seems like every university in the country is trying to build a program in bioinformatics."

In addition to his academic appointment, McCarter is president of Divergence LLC, a start-up company that is researching ways to kill nematodes -- which are parasitic worms that harm people, plants and animals -- without harming the host organism or the environment.

Because humans and parasites "look an awful lot alike at the level of the genes," Divergence needs a combination of biology and computing expertise to identify a nematode gene that doesn't exist in humans, and then find a lethal agent that acts on that gene.

Most of the data Divergence is using is in the public domain because it was created with government-funded research at the Genome Sequencing Center. But because of the company's bioinformatics expertise, McCarter said, "We feel that we can very competitively mine that data and see things other people cannot see."

Bioinformatics can also be an important building block for St. Louis' high-tech industry, McCarter said. "These people are very rare. They are either biologists who happened to be hackers on the side, or they were computer scientists who fell into biology."

back to articles index

back to top