• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Companies' Shift to 'Application' Research and Development May Cut Innovation

Posted on: Monday, 3 November 2003, 06:00 CST

Nov. 3--Taking stock of research and development projects at General Electric Co. three years ago, its new chief executive, Jeffrey R. Immelt, visited the company's 100-year-old central research lab in Schenectady, N.Y. What he found was a "high-tech job shop" of 1,000 research projects, Immelt recalled recently. GE since has streamlined the operation to about 20 core projects with shorter time horizons and greater payoff potential.

The idea is to focus on "projects that are meaningful to the company three to five years out," said Immelt, who succeeded Jack Welch and is known as an advocate for innovation.

GE's actions are not unusual. Across corporate America, companies are reassessing and restructuring their research efforts at a time of slower growth and shorter product life cycles. While some have boosted their research spending and others have scaled back, most have been deemphasizing longer-term basic research in favor of efforts that will bear fruit in the marketplace in the next few years.

In the past, "most of the research and development labs were practicing science for the sake of science," said Jane L. Snowdon, manager of embedded applications for IBM Research in Yorktown Heights, N.Y. "Now about 25 percent of our researchers are working with customers on very tough challenges."

The shift to what is sometimes called "application R&D" -- which is often focused on business processes, along with other trends such as companies relying more heavily on acquisitions for their research -- has prompted some concern about the future of innovation. The ability to innovate is seen as a key engine driving the US economy, especially at a time when not only blue-collar manufacturing jobs but also white-collar jobs such as software programming are migrating overseas.

"Short term, it's a great strategy," said Adrian J. Slywotzky, vice president at Mercer Management Consulting in Boston. "Long term, it's a disaster."

Overall industrial research and development spending dipped to $198.5 billion in 2001, the most recent year for which figures are available, from $199.5 billion the previous year, according to a survey released last month by the National Science Foundation. It was only the third annual decline recorded in the past 50 years, and came largely because federal funding of business research slipped to $15.4 billion from $17.8 billion.

The survey was taken for the science foundation by the US Census Bureau. In the past, the foundation has not funded a survey on the mix of longer- and shorter-term research projects. But with increased interest in that subject, it will ask the Census Bureau to begin surveying for that data in the coming year.

Nathan Myhrvold, who founded Microsoft Research in the early 1990s and is now running an inventors' consortium, deplored the current state of corporate research in a speech at MIT's Emerging Technologies Conference this fall. Claiming that research operations at IBM Corp. and Bell Labs are "shadows of their former selves," while high-tech giants like Oracle Corp. and Cisco Systems Inc. have skimped on research, Myhrvold said the trend toward accelerating applied research at the expense of long-term projects is "a disastrously stupid thing to do."

IBM's Snowdon said her company continues to do exploratory research, in fields like pervasive computing and wireless technology, at a network of labs that employ more than 3,000 people in six countries. But she acknowledged that a growing focus of IBM research is shorter-term "on demand innovation," and customer-focused projects, such as real-time distribution systems or customized data mining that enables retailers or airlines to understand buying patterns in their markets.

"It's really driven by business problems that customers have, where they can't buy something off a shelf," Snowdon said.

Historically, the US government has played a key role in underwriting industrial research, through funding industrial labs and commercial spinoffs of military projects, from jet propulsion to the Internet, said Kei Koizumi, analyst for the American Association for the Advancement of Science in Washington.

After the Cold War, however, federal funding began tapering off and corporations stepped up to play a larger role, he said. Koizumi added the trend could be changing again now, as some corporations retreat from research expenditures and the government boosts spending on defense, homeland security, and even life sciences.

Research and development efforts at corporations have gone through several "paradigm shifts," said Ryan M. Stark, senior vice president of technology at Nortel Networks Ltd. in Ottawa.

"Seven or eight years ago, things were built in a glass house -- on site, internally," Stark said. "With the Internet boom, we changed the model. The pendulum swung to more of an acquisition model."

Nortel, along with competitors such as Cisco and Lucent Technologies, snapped up dozens of networking and telecom start-ups in New England and elsewhere in the late 1990s. Technology gleaned from such acquisitions, together with in-house research and collaborations with universities such as MIT, has created a "hybrid model" of research, Stark said. "We've moved to a short-cycle ecosystem" for research and development, he said.

While corporate labs like Xerox PARC in Palo Alto, Calif., once spun off technology "that will give you a pop in 10 years" -- including in Xerox's case the computer mouse and the graphical user interface -- few private labs are making those kinds of strides today, said Mercer's Slywotzky. "Managers have been saying, 'We can't afford to do this basic stuff. We've got to get products out the door,"" he said.

Sometimes, getting products out the door means locating research operations closer to a target market. Oracle Corp. last week said it was opening its second Chinese research and development center in Beijing. Last year, Oracle established an R&D beachhead in the southern Chinese center of Shenzhen that today employs about 100 workers. Both offices work on applications that adapt Oracle's database software for use by Chinese businesses and government agencies.

Some companies have moved to boost their research spending in a slow economy, seeking a competitive advantage. Microsoft Corp. will increase its R&D outlays by 8 percent to $6.8 billion in the current year and hire up to 5,000 researchers worldwide, said Mike Kosek, the Microsoft general manager for New England. Not surprisingly, Microsoft is exhorting its customers to invest in the fruits of such research efforts.

"Not investing in information technology might not be in the best interest of organizations," Kosek warned.

-----

To see more of The Boston Globe, or to subscribe to the newspaper, go to http://www.boston.com/globe

(c) 2003, The Boston Globe. Distributed by Knight Ridder/Tribune Business News.

GE, IBM, 6680, NT, CSCO, LU, XRX, ORCL, MSFT,

More News in this Category