R&D Funding Levels Affected by External Events

Growth continues, but overall values dampened by government funding issues and outsourcing

click to enlarge 
The Battelle/R&D Magazine 2007 R&D Funding Forecast is the latest installment in a continuous series dating back to the 1950s. Over most of this period of time, the U.S. had generally maintained a position of overall global dominance in R&D investment. Over the past decade, however, the character of U.S. R&D funding and performance has undergone significant changes, with some noted as follows:
• Federal funding of R&D has been impacted by programs directed toward the global war on terror, and the development of techniques to fight non-traditional types of ground wars.
• Federal funding for R&D also has been affected by major federal budget deficits that accompanied the necessity of responding to the unanticipated damage associated with natural disasters.
• Changes in industrial funding have been influenced by major moves toward establishing or pursuing research programs in financially and intellectually attractive offshore locations.
• Other countries, particularly China and India, are making aggressive moves toward becoming formidable participants in the global R&D and technology arena, with resultant strengthening economic presences.

These may well be seen as factors that have had, and will continue to have, negative impacts on the health and strength of the U.S. R&D enterprise. Both established and growing areas of science and science-based economic activity are expected to provide a dampening force on what had been dominant sectors in the U.S. economy.

While electronics, biotechnology, pharmaceuticals, software development, process modeling and a host of other areas are receiving considerable attention, they are still expected to continue as growth fields. Indeed, analysts from Battelle, Columbus, OH, and the editors of R&D Magazine forecast that, in 2007, total R&D spending in the U.S. will increase to $338 billion, a gain of 2.85 percent over 2006 levels as shown in the source-performer matrix (Table 1). This matrix — derived from the
click to enlarge 
National Science Foundation's (NSF's) National Patterns publications, serves to describe the distribution of R&D dollars in the U.S. and the manner in which the various types of institutions interact. In addition, the collection of such data over a long period of time (supplemented by breakdowns into the categories of basic and applied research, and development) provides a historic base that is useful for identifying trends and for supporting research.

A brief description of the interacting categories follows:
• Industry-supported research is performed in industrial, academic, and non-profit institutions. In actuality, industry also supports R&D that is conducted in some federal laboratories, with such initiatives having been encouraged by various federal programs over the past two decades. However, the total amount of such effort is relatively small.
• Academic institutions are a source of funds, almost all of which they spend in their own laboratories. Included within the category of academic funds are those monies that accrue from investments, sales, donations and the like. In addition, many state government expenditures in R&D are subsumed into academic sources, even though it's known that some state resources represent support of work that is performed by non-academic institutions.
• Non-profit organizations are classified by the NSF as organizations that provide funds for both academic and non-profit institutions' research, although it's believed that they also support work performed by industry.

Each of these categories reflects the realities of well-established practices; each provides an input to understanding the character of the overall R&D enterprise, and each adds another dimension to developing an improved means for assessing the impacts of decisions relative to R&D practices.

The flow of federal funds
Since the NSF began maintaining statistics on R&D, these data have provided a basis from which one can examine the changes in the patterns of funding and gain some insights into the evolution of the overall R&D enterprise. For example, an inspection of the pattern of expenditures of federal R&D dollars among the five classes of performing institutions (federal laboratories, industrial labs, academia, non-profit organizations and the FFRDCs — federally funded R&D centers) reveals that, over the past half century, there has been a steady decline in the proportion of total federal dollars that has been spent in industry facilities, with the principal beneficiary being academia.

When examined in more detail, it's apparent that the increase in academic share was strongly influenced by academia's growing presence as the preferred performer of basic research, at the expense of the federal labs and industry. Non-profit organizations maintained a relatively steady relative role.

The performance of applied research has been somewhat more erratic with tradeoffs in relative performance by both the federal labs and industry vacillating over the most recent 20-year period,
click to enlarge 
while all other sectors (especially academia) enjoyed a general growth. Finally, the distribution of support for development has shown steady growth of the relative performance by the federal labs and the FFRDCs, again at the expense of industry.

That such a shift in the distribution of R&D work should occur is likely related to the practice, over the past several years, of acquiring COTS (commercial off-the-shelf) systems for a variety of mission agencies. In such cases, the eventual supplier assumes the front-end costs of R&D, with these costs being recovered later through the successful marketing and sales of the subject components. On the surface, the practice shifts the risks from the purchaser to the supplier, and permits the diversion of what would nominally have been called government R&D costs to another category, such as systems acquisition or the equivalent.

Distribution of industry funds
Inspection of the distribution of industry R&D dollars among the three performing sectors — industrial labs, academia, and non-profit organizations — generally provides a much more clear-cut and understandable picture, but with a major caveat. With the exception of basic research, the preponderance of work is performed by industry, with only minimal fractions expended by academia or non-profit organizations. As expected and reasonable, industry-supported basic research is still dominated by industrial performance, although the academic share has been as high as 20 percent and the non-profit share as much as eight percent of the total.

There are two principal points that should be made regarding these observations, one dealing with industry-supported basic research and the other addressing the issue of industrial outsourcing of R&D work.

The basic research question
The classic definition of basic research is such that it is usually defined as being that type of activity that pursues the expansion of knowledge and has as its objective the understanding of phenomena just for the satisfaction of understanding. The usual definitions of "basic research" are devoid of reference to application, and look only toward broader or deeper understandings of physical, chemical and biological phenomena.

Such a definition is generally inconsistent with the objectives of profit-making private industry. To be sure, some of the major industrial laboratories of the past pursued this definition of basic research, and we are surely better off
click to enlarge 
for it. Work at IBM, Bell Labs, General Electric, DuPont and countless others set the stage for fundamental discovery and subsequent applications that have transformed the economy. For the most part, industrial basic research of that character and at that level has not been maintained, but has been supplanted by "directed basic research." That is to say, that which is presently thought of as "basic" is now much more aligned with exiting or planned applications and product lines. Hence, it is somewhat of a distortion to combine industrial support of basic research with that which is funded by the other sources.

Changes total R&D support and outsourcing
The period from 1995 through 2001 was characterized by continued average increases in expected R&D expenditures coupled with increases in the extent to which outsourcing would be pursued. This was followed by a period of lower, but recovering, expected R&D expenditures and decreased outsourcing. In the most recent two-year period, both the anticipated total R&D and the total outsourcing have increased.

Over much of the early part of the period covered, capital spending changes were relatively average and unchanging. Starting in 2001, the anticipated capital spending began to decrease rather strongly, followed by a recovery that began in anticipation of 2005 activities and beyond. Without further detail and response, it's not possible to assess whether such increases are associated with domestic or off-shore facilities, although there are many well-publicized cases of significant investments in the off-shore programs — in both the conduct of R&D and the construction of bricks-and-mortar research campuses.

Distribution of R&D
Among the three categories of R&D support to existing businesses, new-business projects and directed basic research, the emphasis over the past 13 years has consistently been directed toward new business projects. The strength of that enthusiasm had waned somewhat when considering R&D activities for the 2003-2005 period, but has regained strength in the most recent two years. On average, with the exception of these same two most recent years, support for existing businesses has slightly outranked the efforts on directed basic research. The concurrent revival of concentration on both new business projects and directed basic research suggests either a general new confidence or a greater promise (and perhaps great cost effectiveness) associated with the increase outsourcing referred to above.
Final reflection
Federal attempts to fund R&D come at a time when the federal government's support and appreciation of research and its potentials is being overshadowed by the impacts of a budget beset with the cost of military engagements, unanticipated relief funds and huge deficits over many years to come. Unless we take major steps now — not five years from now — we are going to find ourselves, as a country, in exactly the same position we had when the first Sputnik went up in 1957: caught with a significant technological surprise!

Jules Duga and Martin Grueber are senior research scientists at Battelle. Tim Studt is Strategic Editorial Director of Scientific Computing and Editor in Chief of R&D Magazine. They may be reached at

Strong Industrial R&D and Getting Stronger
Tim Studt

U.S. industry spends more on R&D than any other organization, government or country on the face of this earth — these commercial organizations spend $219 billion and they do $244 billion worth of work, according to our 2007 R&D funding forecast. That's 64 percent more than the entire country of China — government, industry and academia combined — spends on R&D, yet China is the second largest R&D spending entity. A lot is said in this report about the basic research that academia performs and the problems with the federal government's declining support of R&D (for a special case in 2007), but let's not forget that industrial R&D spending is up more than $7 billion this year — there are only 17 countries in the world that spend more that on R&D — and actual performance of R&D by U.S. industry is up nearly $8 billion.

As is noted elsewhere in this report, the amount of monies spent by industry has become partially diluted as companies look to outsource a portion of their R&D work to foreign countries for a variety of reasons, including lower costs (75 percent of the responses when focusing on China), an expanding market (68%), closer to manufacturing sources (65%), closer to customers (58%), building collaborations (42%), and ensuring the quality of products (40%). Still, only about 10 percent of the respondents
click to enlarge 
from an R&D Magazine reader survey indicated that they would outsource some of their R&D work to foreign countries like China, India, Korea and others. This ratio is likely to increase in the future, but most indicators reveal this change will be a very slow increase and may even plateau or reverse itself at some point in the not-too-distant future.

Table 2 reveals the level of R&D spending and its comparative scale to other industries (total industry spending reflects the amount spent by companies overall — both in the U.S. and in other countries). It's interesting to note that these five primarily technology-based industries have an average R&D growth of approximately six percent, about four percentage points over the expected inflation rate for 2007 and almost a full percentage point more than the growth cited in our 2006 R&D funding forecast.

This is not to say that industrial R&D growth is not without its problems. For one, the U.S. automotive industry continues to see an eroding market share to foreign suppliers. But these same foreign competitors also have built numerous manufacturing facilities throughout the U.S. and have established very significant R&D design, development and testing facilities within the U.S., as well. The U.S. aerospace marketplace, while partially built upon defense and homeland security markets, is also very healthy as commercial airliner sales continue to improve, communication systems and imaging satellite sales continue to escalate, and other high-speed transport systems are funded to provide alternatives that could offset the escalating price of foreign petroleum products.

It's interesting to note the islands of differences within industrial groupings, such as that within the chemical industry which is mostly flat (DuPont), except when it comes to becoming involved with high-growth agriculture (Monsanto, St. Louis, MO) or military product areas (Dow Chemical, Midland, MI).

The biopharm industry is fraught with issues, including not having enough new products close to the end of the development pipeline (to offset those multibillion dollar products which are just starting to come off of patent protection). The overall drug development strategies that many pharmas undertook over the past ten years have mostly been unsuccessful. The largest pharmaceutical company, Pfizer, New York, NY, is rumored to be preparing to make massive cuts to its R&D programs this year (as much as 30 percent), along with numerous other cost-saving actions to ensure that the remaining products support the company's operations.

The biotechnology portion of this industry continues to grow with market leaders Amgen, Thousand Oaks, CA, and Genentech, San Francisco, CA, continuing their very aggressive R&D programs. Stem cell research continues to evolve and a more liberal political climate offers a glimmer of more funding and "allowances" than has been seen in the past six years.

The information technology industry is also a mixed bag of specialized companies. IBM overall continues to thrive in the R&D marketplace and continues to make scientific and engineering breakthroughs that support its overall market strategies. HP, Palo Alto, CA, as well appears to be thriving in the computer marketplace, especially in the high performance computing areas. Other companies like Sun Microsystems, Santa Clara, CA, and Silicon Graphics, Sunnyvale, CA, have made strategic miscalculations and their R&D efforts continue to erode as the companies struggle to survive.

The semiconductor marketplace appears poised for dramatic growth as various elements within support the emergence and development of new technology — low-cost, high-efficiency photovoltaic systems that are approaching the cost efficiencies of petroleum-based energy supplies. Conventional semiconductor systems are maintaining their overall development pace with appropriate R&D funding increases, although low-cost off-shore facilities continue to compete for market share. The increasing complexity of next generation technologies and their respective devices, however, continues to increase the market entry barriers that will limit the number of competitors that existing suppliers will be faced with.

The telecommunications industry continues its problem-plagued story with most U.S. companies seeing their R&D budgets continue to shrink in one of the most competitive marketplaces.

Other industries show promise for continued improvements in their R&D investments, such as energy, nanotechnology-based materials science and any industry or sub-industry involved in sustainability areas. Overall, the U.S. industrial climate is strong and looking to improve its position over the next 12 months with continued and even stronger investments in R&D in 2008.

Tim Studt is Strategic Editorial Director of Scientific Computing and Editor in Chief of R&D Magazine. He may be reached at