Gone ... Or Not?

By Keith McDowell

“The Greatest Generation” left us with more than victory in World War II and the passing of the Great Depression. They created an era of prosperity and the advent of a large middle class. Even more, they brought us an era founded on innovation driven by corporate research laboratories such as Bell Labs, RCA Labs, Xerox PARC and many others. And they built a nationwide infrastructure of national laboratories and the world’s greatest research universities, all working together to make America the leader of the World.

But as we enter the second decade of the twenty-first century, many question the American model and whether it is relevant and sufficiently adaptive to the new global competition and Friedman’s mantra of “the world is flat.” President Obama in his speech to the National Academy of Sciences on 27 April 2009 best captured the spirit and mood permeating the past decade. It is worth repeating his comments:

A half century ago, this nation made a commitment to lead the world in scientific and technological innovation; to invest in education, in research, in engineering; to set a goal of reaching space and engaging every citizen in that historic mission.  That was the high water mark of America's investment in research and development.  And since then our investments have steadily declined as a share of our national income.  As a result, other countries are now beginning to pull ahead in the pursuit of this generation's great discoveries.  

I believe it is not in our character, the American character, to follow.  It's our character to lead.  And it is time for us to lead once again.  So I'm here today to set this goal:  We will devote more than 3 percent of our GDP to research and development.  We will not just meet, but we will exceed the level achieved at the height of the space race, through policies that invest in basic and applied research, create new incentives for private innovation, promote breakthroughs in energy and medicine, and improve education in math and science. 

Few would disagree with his analysis or fail to support his promise of devoting 3 percent of GDP to research and development. It’s the heart, soul, and core of any long-term innovation strategy. But do we have the R&D capacity to absorb and effectively spend such an investment, especially in the industrial sector? Indeed, many believe that the industrial R&D laboratories such as Bell Labs are “missing in action” and a creature of the twentieth century.

In the cover letter of 20 November 2008 to President Bush for the PCAST report entitled University-Private Sector Research Partnerships in the Innovation Ecosystem, the second paragraph states that “Additionally, we have observed a decrease in the number and size of industrial basic research laboratories.” That’s the prevailing view of America’s science leadership!

On the business side, Adrian Slywotsky in a Business Week article of 27 August 2009 entitled Where Have You Gone, Bell Labs? states that “Our growth engine has run out of a key source of fuel – critical mass, basic scientific research.” As his title suggests, he is referring specifically to the putative demise of industrial basic research laboratories. But he offers a solution and one that I agree with: “Today’s challenges require the government to unleash a series of highly focused, aggressively managed projects supported by a growing research investment in a dozen or more leading companies that in the aggregate reproduce the cumulative impact of Bell Labs, RCA Labs, Xerox PARC, and others.”  He further argues that “We need them. Soon.” 

Notwithstanding the demise of Bell Labs as an innovative force, this begs the question of whether American industry is truly reducing its commitment to R&D, especially basic research. Is America experiencing “a decrease in the number and size of industrial basic research laboratories?” So far, I’ve been unable to find a source of hard data that validates this claim, other than anecdotal stories. For example, the Center for History of Physics of the American Institute of Physics undertook in 2003 “a five-year study of the history of physicists in industry” and reported results in the July 2009 issue of Physics Today in a story by R. J. Anderson and O. R. Butler entitled Industrial R&D in Transition. While the results are skewed toward industries in which physicists play major roles, they are nonetheless informative. Here is a short version of the findings related to industrial research:

• Management is “Struggling to find the best mix of longer-term research and short-term development.”
• Radical R&D funding and organizational changes are endangering centralized labs.
• Management is looking to “external sources … for innovative technology” following successes in the Asian or early Japanese model.
• There are no standards for preservation of records and America is losing precious “know how.”

Confirming the notion that industrial basic research is declining, the Physics Today article states that “interviewees at all the laboratories we visited described a sharp transition over the past two decades toward shorter-term projects and more control by the business side of operations.” 

In a similar vein Marguerite Reardon in a CNET news article describing industrial research laboratories such as AT&T’s Bell Laboratories, Xerox’s Palo Alto Research Center (PARC), and IBM’s Watson Research Center stated that “The labs are still around, but some experts say the labs conduct basic research on a much smaller scale than they used to.”

The message from conventional wisdom is clear and consistent. There is a decline underway for industrial basic research laboratories. But lacking hard data for the number, size, description and funding of industrial basic research laboratories, are we merely looking at a phenomenon associated with the physical sciences, the domain of the twentieth century giants, or does it cross all sectors of science and technology?

I decided to examine the question on the basis of industry basic research expenditures relative to the GDP of the United States. Using the yearly industry basic research expenditure data available from the NSF S&E Indicators report and taking fourth quarter GDP data available at the website forecasts.org/data/data/GDP.htm, I computed the percentage of the yearly industry basic research expenditure to the fourth quarter annual GDP (for example, the fourth quarter of 1953 would be dated as 1954-01-01 in the GDP table). Each annual percentage was then compared to the 1953 percentage such that a value of 1.00 represents an industry basic research expenditure percentage against GDP equivalent to that of 1953. The resulting curve is given in the following plot.

As we can see from the plot, industry expenditures on basic research as a percentage of GDP have moved around over time and become somewhat erratic over the past decade. We don’t see a recent pattern of less money being spent on basic research as a percentage of GDP. Instead, the ratios are commensurate with the previous highs. These data do not indicate the decrease that might be expected from the letter to President Bush or the conventional wisdom. Of course, this doesn’t tell us whether there are fewer or more industrial basic research laboratories. Is conventional wisdom being driven by a maturation of the tradition physics-based laboratories? Or do the traditional laboratories need a new beginning? Perhaps the “lablets” model advocated by Slywotsky and others including the author is the way to go.

“The Times They Are A Changin’,” as Bob Dylan prophesied. And the changes augur well for industrial basic research. Witness the recent endeavors by Google in the arena of wind power and their investment of $100 million in the world’s largest wind farm. As Katie Fehrenbacher says in her article, Google Invests $100M in (Another!) Wind Farm, “Google has put more than $350 million (a jaw-dropping amount for an Internet company) into clean power.” Maybe the Cassandra’s of conventional wisdom are looking in the wrong places.