AMERICA’S BASIC RESEARCH · vi The Committee for Economic Development is an independent research...

R E S E A R C HP R O S P E R I T Y

T H R O U G HD I S C O V E R Y

A Policy Statement by the Research and Policy Committee of the Committee for Economic Development

A M E R I C A ’ SB A S I C

Copyright © Committee for Economic DevelopmentAll rights reserved

Library of Congress Cataloging-in-Publication Data

Committee for Economic Development. Research and Policy Committee.America’s basic research : prosperity through discovery : a policy

statement / by the Research and Policy Committee of the Committee for Economic Development.

p. cm.

Includes bibliographical referencesISBN 0-87186-130-51. Research—Economic aspects—United States. 2. Research—United States.

HC110.R4C63 1998001.4’0973—dc21 98-24814

CIP

First Printing in bound book form: 1998Paperback: $18.00Printed in the United States of America

COMMITTEE FOR ECONOMIC DEVELOPMENT477 Madison AvenueNew York, N.Y. 10022(212) 688-2063

2000 L Street, NW, Suite 700Washington, D.C. 20036(202) 296-5860

http://www.ced.org

CONTENTS

Page

RESPONSIBILITY FOR CED STATEMENTS ON NATIONAL POLICY vi

PURPOSE OF THIS STATEMENT ix

CHAPTER 1: INTRODUCTION AND SUMMARY OF FINDINGS AND RECOMMENDATIONS................................................... 1

Findings ................................................................................................................................................. 2Summary of Recommendations ......................................................................................................... 4

CHAPTER 2: UNDERSTANDING THE CONTRIBUTION OF BASIC RESEARCH...................................................................................... 6

The Benefits of Basic Research................................................................................................................. 6The Economic Impact of Basic Research ............................................................................................... 10Publicly-Funded Basic Research Is Critical to Private Sector Innovation......................................... 12The Future Impact of Basic Research...................................................................................................... 13

CHAPTER 3: THE EVOLVING AMERICAN BASIC RESEARCH SYSTEM........................................................................................ 15

Today’s Basic Research Institutions ........................................................................................................ 16The Critical Value of The Individual Researcher .................................................................................. 27The International Context for American Basic Research ..................................................................... 29Conclusion.................................................................................................................................................. 31

CHAPTER 4: MAINTAINING U.S. LEADERSHIP IN BASIC RESEARCH: THE CED PRESCRIPTION ..................................................... 32

Improving the Allocation of Resources for Basic Research................................................................. 32Preserving Our Capacity to Do Basic Research in the Future............................................................. 38Principles for University Research in the Marketplace........................................................................ 44International Challenges and Opportunities for American Basic Research.......................................................................................................................... 45Conclusion.................................................................................................................................................. 47

APPENDIX 1 – OVERVIEW OF RESOURCES FOR BASIC RESEARCH 48

APPENDIX 2 – UNIVERSITY PATENTING GUIDELINES 57

iii

iv

CASE STUDIES ............................................................................................................................ 64Merck & Co., Inc.Columbia UniversityPfizer Inc.Procter & GambleHarvard UniversityXeroxIBMBBN

MEMORANDA OF COMMENT, RESERVATION, OR DISSENT........................................ 80

A M E R I C A ’ SB A S I C

R E S E A R C HP R O S P E R I T Y

T H R O U G H

D I S C O V E R Y

vi

The Committee for Economic Developmentis an independent research and policy organizationof some 250 business leaders and educators.CED is nonprofit, nonpartisan, and nonpolitical.Its purpose is to propose policies that bringabout steady economic growth at high employ-ment and reasonably stable prices, increasedproductivity and living standards, greater andmore equal opportunity for every citizen, and animproved quality of life for all.

All CED policy recommendations must havethe approval of trustees on the Research andPolicy Committee. This committee is directedunder the bylaws, which emphasize that “allresearch is to be thoroughly objective in character,and the approach in each instance is to be fromthe standpoint of the general welfare and not fromthat of any special political or economic group.”The committee is aided by a Research AdvisoryBoard of leading social scientists and by a smallpermanent professional staff.

The Research and Policy Committee does notattempt to pass judgment on any pending

specific legislative proposals; its purpose is to urgecareful consideration of the objectives set forthin this statement and of the best means of accom-plishing those objectives.

Each statement is preceded by extensive dis-cussions, meetings, and exchange of memoran-da. The research is undertaken by a subcommittee,assisted by advisors chosen for their compe-tence in the field under study.

The full Research and Policy Committee par-ticipates in the drafting of recommendations.Likewise, the Trustees on the drafting subcom-mittee vote to approve or disapprove a policy state-ment, and they share with the Research andPolicy Committee the privilege of submittingindividual comments for publication.

Except for the members of the Research and PolicyCommittee and the responsible subcommittee, therecommendations presented herein are not necessar-ily endorsed by other trustees or by the advisors,contributors, staff members, or others associated with CED.

RESPONSIBILITY FOR CED STATEMENTS ON NATIONAL POLICY

vii

Chairman

JOSH S. WESTONHonorary ChairmanAutomatic Data Processing, Inc.

Vice Chairmen

IAN ARNOFPresident and Chief Executive OfficerFirst Commerce Corporation

ROY J. BOSTOCKChairman and Chief Executive OfficerThe MacManus Group

BRUCE K. MACLAURYPresident EmeritusThe Brookings Institution

CLIFTON R. WHARTON, JR.Former Chairman and Chief Executive OfficerTIAA-CREF____________________________________

REX D. ADAMSDeanThe Fuqua School of BusinessDuke University

ALAN BELZERRetired President and Chief Operating

OfficerAlliedSignal Inc.

PETER A. BENOLIELChairman, Executive CommitteeQuaker Chemical Corporation

GORDON F. BRUNNERSenior Vice President and DirectorThe Procter & Gamble Company

FLETCHER L. BYROMPresident and Chief Executive OfficerMICASU Corporation

DONALD R. CALDWELLPresident and Chief Operating OfficerSafeguard Scientifics, Inc.

PHILIP J. CARROLLPresident and Chief Executive OfficerShell Oil Company

JOHN B. CAVEPrincipal Avenir Group, Inc.

CAROLYN CHIN Executive Vice PresidentReuters America

A. W. CLAUSENRetired Chairman and Chief

Executive OfficerBankAmerica Corporation

JOHN L. CLENDENINRetired ChairmanBellSouth Corporation

GEORGE H. CONRADESExecutive Vice PresidentGTE Corporation and PresidentGTE Internetworking

RONALD R. DAVENPORTChairman of the BoardSheridan Broadcasting Corporation

FRANK P. DOYLERetired Executive Vice PresidentGE

T. J. DERMOT DUNPHYChairman and Chief Executive OfficerSealed Air Corporation

W. D. EBERLEChairmanManchester Associates, Ltd.

WILLIAM S. EDGERLYChairmanFoundation for Partnerships

WALTER Y. ELISHARetired Chairman and Chief Executive

OfficerSprings Industries, Inc.

EDMUND B. FITZGERALDManaging DirectorWoodmont Associates

HARRY L. FREEMANPresidentThe Freeman Company

RAYMOND V. GILMARTINChairman, President and Chief Executive

OfficerMerck & Co., Inc.

BOYD E. GIVANSenior Vice President and Chief

Financial OfficerThe Boeing Company

BARBARA B. GROGANPresident Western Industrial Contractors

RICHARD W. HANSELMANRetired ChairmanGenesco Inc.

RODERICK M. HILLSPresidentHills Enterprises, Ltd.

MATINA S. HORNERExecutive Vice PresidentTIAA-CREF

GEORGE B. JAMESSenior Vice President and Chief

Financial OfficerLevi Strauss & Co.

JOSEPH E. KASPUTYSChairman, President and Chief

Executive OfficerPrimark Corporation

CHARLES E.M. KOLBPresidentCommittee for Economic Development

ALLEN J. KROWERetired Vice ChairmanTexaco Inc.

RICHARD KRUIZENGAPresidentISEM, Southern Methodist University

CHARLES R. LEEChairman and Chief Executive OfficerGTE Corporation

A. V. LIVENTALSVice President, Strategic PlanningMobil Corporation

ALONZO L. MCDONALDChairman and Chief Executive OfficerAvenir Group, Inc.

HENRY A. MCKINNELLExecutive Vice PresidentPfizer Inc.

JOSEPH NEUBAUERChairman and Chief Executive OfficerARAMARK Corporation

JOHN D. ONGChairman EmeritusThe BFGoodrich Company

JAMES F. ORR IIIChairman and Chief Executive OfficerUNUM Corporation

VICTOR A. PELSONSenior AdvisorSBC Warburg Dillon Read Inc.

PETER G. PETERSONChairman The Blackstone Group

NED V. REGANThe Jerome Levy Economics Institute

JAMES Q. RIORDANStuart, Florida

MICHAEL I. ROTHChairman and Chief Executive OfficerThe Mutual Life Insurance Company of

New York

HENRY B. SCHACHTDirector & Senior AdvisorLucent Technologies Inc.

ROCCO C. SICILIANOBeverly Hills, California

MATTHEW J. STOVERGroup President, Information

Services GroupBell Atlantic Corporation

ARNOLD R. WEBERChancellorNorthwestern University

LAWRENCE A. WEINBACHChairman and Chief Executive OfficerUnisys Corporation

MARTIN B. ZIMMERMANExecutive Director, Governmental

Relations and Corporate EconomicsFord Motor Company

RESEARCH AND POLICY COMMITTEE

*

*Voted to approve this statement but submitted memorandum of comment, reservation, or dissent, see page 80.

viii

ChairmanGEORGE H. CONRADESExecutive Vice PresidentGTE Corporation and PresidentGTE Internetworking

MELVYN E. BERGSTEINChairman and Chief Executive OfficerDiamond Technology Partners

GORDON F. BRUNNERSenior Vice President and DirectorThe Procter & Gamble Company

FLETCHER L. BYROMPresident and Chief Executive OfficerMICASU Corporation

DONALD R. CALDWELLPresident and Chief Operating OfficerSafeguard Scientifics, Inc.

CAROLYN CHINExecutive Vice PresidentReuters America

JOHN DIEBOLDChairmanJohn Diebold Incorporated

CHRISTOPHER D. EARLSenior Vice PresidentPerseus Capital, LLC

PATRICIA O’DONNELL EWERSPresidentPace University

KATHLEEN FELDSTEINPresidentEconomics Studies, Inc.

RAYMOND V. GILMARTINChairman, President and Chief Executive

OfficerMerck & Co., Inc.

CAROL R. GOLDBERGPresidentThe Avcar Group, Ltd.

JUDITH H. HAMILTONPresident and Chief Executive OfficerFirst Floor Software

NOAH T. HERNDONGeneral PartnerBrown Brothers Harriman & Co.

ROBERT C. HOLLANDSenior FellowThe Wharton SchoolUniversity of Pennsylvania

JAMES P. KELLYChairman and Chief Executive OfficerUnited Parcel Service of America, Inc.

FRANKLIN A. LINDSAYRetired ChairmanItek Corporation

COLETTE MAHONEYPresident EmeritusMarymount Manhattan College

HENRY A. MCKINNELLExecutive Vice PresidentPfizer Inc.

WESLEY D. RATCLIFFPresident and Chief Executive OfficerAdvanced Technological Solutions, Inc.

HOWARD M. ROSENKRANTZPresident and Chief Operating OfficerUnited States Surgical Corporation

NEIL L. RUDENSTINEPresidentHarvard University

GEORGE E. RUPPPresidentColumbia University

JONATHAN M. SCHOFIELDChairman and Chief Executive OfficerAirbus Industrie of North America, Inc.

ALISON TAUNTON-RIGBYPresident and Chief Executive OfficerAquila Biopharmaceuticals, Inc.

JAMES A. THOMSONPresident and Chief Executive OfficerRAND

CHANG-LIN TIENNEC Distinguished Professor

of EngineeringUniversity of California, Berkeley

LINDA SMITH WILSONPresidentRadcliffe College

MARGARET S. WILSONChairman of the BoardScarbroughs

Ex-Officio Members

FRANK P. DOYLERetired Executive Vice PresidentGE

CHARLES E.M. KOLBPresidentCommittee for Economic Development

JOSH S. WESTONHonorary ChairmanAutomatic Data Processing, Inc.

Non-Trustee Members

CHARLES P. HOLTVice PresidentJoseph C. Wilson Center for Research and

TechnologyXerox Corporation

PAUL SCHIMMELProfessor and MemberThe Scripps Research InstituteSkaggs Institute for Chemical Biology

LARRY K. SMITHFormer Counselor for Secretary of

Defense, William J. Perry

FRED W. TELLINGVice PresidentCorporate Strategic Planning & PolicyPfizer Inc.

CHARLES VESTPresidentMassachusetts Institute of Technology

Guests

LEWIS BRANSCOMBProfessor EmeritusHarvard University

THOMAS CLARK, JR.President and Chief Executive OfficerCarver Federal Savings Bank

GREG FARMERVice President, Government Relations &

International TradeNortel

JOHN MONTJOYSenior Vice President and

General CounselGTE Internetworking

HERMINE WARRENPresidentHermine Warren Associates Inc.

PROJECTDIRECTORSWILLIAM J. BEEMANVice President and Director of Economic

StudiesCommittee for Economic Development

SCOTT MORRISSenior EconomistCommittee for Economic Development

ADVISORSCLAUDE BARFIELDResident ScholarThe American Enterprise Institute

for Public Policy Research

M. ISHAQ NADIRIJay Gould Professor of EconomicsNew York University

RICHARD NELSONProfessor, Department of EconomicsColumbia University

PHILIP M. SMITHPartnerMcGeary and Smith

EDITORIALADVISORHOWARD GROSSHG Communications

SUBCOMMITTEE ON SUSTAINING AMERICAN BASIC RESEARCH

ix

Basic research is a critically important—yet often undervalued—source of American eco-nomic growth and prosperity. Advances in fun-damental science and engineering knowledgeresulting from basic research have made possi-ble most of our recent technological progressand the resulting improvements in incomes andquality of life. Our nation now invests a mere 0.4 percent of Gross National Product on basicresearch.

As this policy statement is being released,American science and scientists are acknowl-edged as preeminent in the world. Many of ourscientific endeavors do receive significant polit-ical, public, and financial support. But compet-ing demands for scarcer federal dollars, shiftingeconomic and social priorities, political pres-sures, and short-term corporate earnings pres-sures pose long-term threats to the continuedstrength of America’s basic research and its con-tributions to our prosperity.

This policy statement takes a broad look atAmerica’s basic research enterprise and lays outthe processes and systematic reforms neededto meet emerging risks to the valuable outcomesfrom our investments in basic research. CED’sTrustees undertook this important project in thefirm belief that significant progress with manyof society’s problems and new discoveries willprimarily depend upon fundamental scientificinsights derived from basic research.

ACKNOWLEDGMENTS

This policy statement was developed by an outstanding group of business and academic

leaders from some of the nation’s top research-oriented companies and research universities. Thebreadth and depth of experience representedin this working group is extraordinary, and weare very grateful for the time, effort, and care eachmember put into the development of this state-ment. We are also grateful for the attention givento this report by numerous scholars, advisors, andreviewers.

Special thanks goes to subcommittee chairman,George H. Conrades, Executive Vice President,GTE and President of GTE Internetworking, forhis leadership and for the intellectual rigor anddrive he brought to this project. Thanks are alsodue to subcommittee members Raymond V.Gilmartin, Chairman, President & CEO of Merck& Co., Inc., for his strong personal commitmentto this project and to Christopher D. Earl, SeniorVice President of Perseus Capital LLC, for his helpin crafting the introductory chapter of this report.The clear, cogent, and perceptive analysis in thisstatement is due in large part to our co-projectdirectors, William J. Beeman, CED’s Vice Presidentand Director of Economic Studies, and CED’sSenior Economist Scott Morris, who addedimmeasurably to the success of this study. Thanksalso to Amanda Turner in CED’s Washington officefor her assistance on this project.

We are also grateful to The New YorkCommunity Trust for supporting this project.

Josh S. WestonChairmanCED Research and Policy Committee Honorary ChairmanAutomatic Data Processing, Inc.

PURPOSE OF THIS STATEMENT

1

Asked why American scientists have won somany Nobel prizes, a former secretary-generalof the Royal Swedish Academy of Sciences onceremarked, “No other country has invested as muchmoney in research over the years as the U.S. It’sas simple as that.” Indeed, America’s long-stand-ing endowment of basic research has been over-whelmingly successful, providing Americansociety not only with the fruits of new knowledge,but also with the practical benefits of economicgrowth and improvements in the welfare of itscitizens.

CED believes, however, that the success ofAmerican basic research is not simply a matterof money. Rather, as we argue in this report,that success has grown from a uniquely Americanorganization of the basic research enterprise.That organization has relied on an abiding faithin the superiority of a free market in ideas andentrepreneurial competition over top-down deci-sion-making in ensuring the quality and effi-ciency of research efforts.

We believe it is essential to uphold the inte-gral role of government in supporting basicresearch, as industry continues to focus on R&Dwith specific product-directed goals. The largeeconomic returns from investments in basicresearch show it to be an extremely productiveuse of the taxpayer’s money. We are encour-aged by recent bi-partisan proposals to increasethe nation’s investment in basic research. But wecannot take today’s political support for grant-ed, especially as we look towards a future ofgreater resource constraints. American excel-

lence in basic research is truly a national treasure,but its supporters must be vigorous and articu-late in its defense.

Basic research—conducted in academic insti-tutions, federal laboratories, private companies,and nonprofit research institutions—has pro-vided the intellectual and technological founda-tion for innumerable practical inventions that areintegral to American technological and eco-nomic leadership. Industries as diverse as phar-maceuticals, defense, electronics, and aerospacehave relied on basic discoveries fueled by gov-ernment grants.

A critical factor in the translation of basicresearch into functional applications has beenAmerica’s unique entrepreneurial spirit. From small start-ups to large multinational corpora-tions, American ingenuity has excelled in con-verting new knowledge into practical andprofitable products. A common misconceptionis that fundamental research is conducted in anivory tower, with no regard for practical bene-fits. On the contrary, a consistent virtue of U.S.basic research has been the pursuit of funda-mental knowledge with a sharp eye out fordownstream applications. American entrepreneurshave been distinguished by their ability to cap-italize effectively on new knowledge whereverit arises.

Like any far-reaching enterprise that com-prises hundreds of institutions and thousands ofworkers, America’s basic research establishmentmust constantly renew itself in response tochanging conditions in global economic, politi-

Chapter 1

INTRODUCTION AND SUMMARY OF FINDINGS AND RECOMMENDATIONS

2

cal, and scientific markets. This enterprise mustalso recognize the legitimate expectations of thesociety that supports its efforts.

The basic research enterprise faces impor-tant questions about the priorities and balanceof its basic research missions, the consistency ofgovernment support, the global disseminationof new knowledge, and the collapse of ColdWar rationales for massive investment in defenseresearch. At the same time, the institutions thatperform basic research must maintain the excep-tional quality of their faculties, grapple withever-increasing costs, and confront the complexchallenges of the expanding ties between cor-porations and universities.

The goals of this report are first, to set forththe compelling case for basic research and its ben-efits to society, and second, to make recom-mendations to policymakers and practitioners.We endorse the strength of American universi-ty-based research and its tradition of excellence.But we also advocate a ceaseless quest to mea-sure output against investment, and resultsagainst expectations. This means extending theuse of peer review and competition for researchgrants. We argue for an end to political ear-marks for research, and we are concerned about“mission creep” in those sectors of the basicresearch establishment—particularly certain ofthe Department of Energy’s national laboratories—that have completed or lost their mandates. Inthis regard, we are concerned about the growingtendency of government to directly fund thedevelopment and commercialization of tech-nologies which, with few exceptions, is proper-ly the function of the private sector.

This report also reaffirms positions taken inprevious CED policy statements that have spe-cial relevance to basic research. We place greatemphasis on improving K-12 education to ensureour future supply of outstanding scientists. Wealso insist on the importance of curbing federalentitlement spending, so that future investmentsin basic research will not be undermined by thebudgetary effects of inexorable demographicpressures.

We take the long view: Although a few scien-tific breakthroughs find immediate applications,yields on basic research are typically realized far

in the future. Frequently the greatest benefits arethe least anticipated. The virus research initiat-ed by the War on Cancer in the 1970s deliveredits most significant benefits—both unintendedand unexpected—in the treatment of AIDS in the1990s; only now is this research yielding new drugsthat will transform clinical oncology in the 2000’s.

The medical, environmental, social and mil-itary challenges of the 21st century will demandsolutions that can only emerge from a healthy andproductive basic research system. CED firmlybelieves that maintaining excellence in basicresearch is essential to America’s continuedprosperity and global leadership.

FINDINGS

1. Basic research in science and engineering hasmade a major contribution to the growth ofthe U.S. economy. Economic returns oninvestments in basic research are very high.In addition, the returns to the nation frombasic research investments are substantial-ly higher than the returns to private firms,since advances in fundamental knowledgetend to be widely dispersed and exploitedin innovations that deliver substantial eco-nomic benefits over a lengthy period.

2. Basic research performed in major researchuniversities is typically correlated withstrong economic activities in their neighbor-ing locales. For example, there are morethan 1,000 MIT-related companies inMassachusetts, with world wide sales of morethan $53 billion. Similar developments havetaken place in California’s Silicon Valleyand the Research Triangle of North Carolina.

3. The federal government has long been themost important source of support for basicresearch. Government funding of basicresearch exceeds that of private industry inboth absolute amount and as a share of itstotal R&D activities. Of the nearly $63 bil-lion that government spends on R&D annu-ally, $18 billion goes to basic research, whilejust $8 billion of industry’s total R&D spend-ing of $133 billion does so. However, the long-

3

term future of federal funding is clouded bythe likely budgetary effects of impendingdemographic pressures.

4. Publicly-funded basic research is criticalto private sector innovation. Although pri-vate industry conducts basic research, theseefforts are primarily to “fill-in-the-gaps”within broader programs of applied researchaimed at new product development. Industrydepends on the intellectual foundationsprovided by basic researchers in the nonprofitand public sectors for innovative productsand services. A recent study found that 73percent of research publications cited byindustrial patents were derived from gov-ernment-funded research.

5. American science is not conducted in an“ivory tower.” Even basic researchers whoare exploring fundamental problems of the-oretical physics, advanced materials, ormolecular biology are doing so with theexpectation that their work will be relevantto the development of new chips, compos-ite aircraft, or cancer drugs.

6. Federal funding is directed principally notto institutions, but to individual investiga-tors who compete directly for governmentgrants. These investigators represent thebackbone of the American basic researchenterprise. An essential strength of theAmerican basic research system is the allo-cation of grants through a rigorous and com-petitive peer review process.

7. The most important American institutionsconducting basic research are the nation’s200 major research universities. These insti-tutions are characterized by highly com-petitive allocation of funds, a tradition ofexcellence, and a brain trust of highly trainedand motivated faculty, post-doctoral fel-lows, and graduate students. The wide,unrestricted dissemination of researchresults has been important to the broadbenefits of university-based basic researchfor our society.

8. Passage of the Bayh-Dole Act in 1980 explic-itly allowed recipients of government grants

to retain title to inventions made usinggovernment funds. This law has stimulat-ed intense growth in university patentingand subsequent technology transfer frombasic research institutions to industry. As aresult, industry is increasingly involved incollaboration with, and sponsorship of,university-based researchers. While cer-tain fields are comparatively untouchedby Bayh-Dole, the biotechnology industryin particular has been a major beneficiary.

9. When managed skillfully, university-indus-try relationships can benefit both the researchinstitutions and companies, while societyreaps rewards from the efficient transferof technology into useful products. CEDbelieves, however, that these relationshipsshould be conducted according to guidelinesthat protect the primary basic research mis-sion of the universities.

10. The priorities of the federal governmentare changing. With the end of the ColdWar, Pentagon requirements for basicresearch have shifted and contracted. As aresult, the missions of the massive federallaboratory system have changed, and insome cases, disappeared. The federal lab-oratories continue to play important rolesin defense, health, and energy. But some, par-ticularly among the national laboratories atthe Department of Energy, have not actedforcefully to eliminate work in areas nolonger relevant to their missions, nor toexpose themselves to merit-based peerreview processes.

11. Deficiencies in science teaching in primaryand secondary education threaten our futuresupply of outstanding young researchers.These deficiencies may also limit the capac-ity of women and minorities to pursuecareers in basic research. Moreover, con-tinued societal support for basic researchdepends on an informed electorate whorecognize the benefits of a strong basicresearch system.

12. As other nations build their economicstrength, they will inevitably also invest

4

in basic research capabilities. While theUnited States continues to be the world’sleader in the generation of new knowl-edge, other countries are contributing morethan in the past. We have nothing to fear fromthese trends. So long as basic research is freelydisseminated around the globe and theU.S. continues to capitalize on the practicalapplication of new knowledge, global expan-sion of basic research on balance will ben-efit the U.S. economy and society at large.

SUMMARY OF RECOMMENDATIONS

1. Policymakers in Congress and the Admin-istration, informed by a national policydebate, should set broad national prioritiesfor basic research that reflect the needs ofsociety at large. Scientists have an impor-tant role to play in informing the debate, butsuch priorities are appropriately set by ouraccountable, elected political leaders.

2. Federal support for basic research shouldcontinue to be diverse in its sources and objec-tives. Efforts to impose central control or toconcentrate resources in a single research area should be resisted.

3. Within the broad priorities established bypolicymakers, the primary mechanisms forallocating federal basic research funds in allagencies and to all institutions should bebased on scientific merit determined throughpeer review. In general, support should begiven to individuals and not to institutions.Political earmarks for basic research are anunproductive use of scarce resources.

4. Because federal support is essential for athriving basic research enterprise, the long-term federal budget outlook is critical. Basicresearch should be a high priority in thefederal budget in the decades to come. Thiswill require reforms of the federal entitlementprograms that otherwise will grow explosivelyin response to demographic pressures a fewyears hence.

5. The most productive recipients of federal basicresearch funds are the nation’s research uni-

versities. Their leadership and productivi-ty should continue to be a guide for otherinstitutions receiving federal support.

6. Individual investigators are increasinglyweighed down by the complex demands ofseeking grant support. Mechanisms shouldbe devised to allow researchers to competefor longer-term funding, and administrativeburdens from granting agencies should bereduced.

7. Funding agencies are increasingly attempt-ing to reduce payments for indirect costs andto shift costs to research institutions. We rec-ommend reform of the system for deter-mining indirect costs to ensure simplicity,fairness, and reductions in the costs of com-pliance.

8. CED calls on Congress and the Admin-istration to clearly determine the missionsof the Department of Energy’s national lab-oratories and decide what realignments ofmissions and functions are necessary. Theactivities of the national labs must be jus-tified on the basis of strong missions, peer-reviewed determinations of scientific merit,and efficient structures for managementand oversight.

9. With few exceptions, government should notbe in the business of directly funding thedevelopment and commercialization oftechnologies, which is properly a functionof the private sector. Exceptions generallyoccur in cases where the funding serves aclear procurement function for governmentmissions, such as defense technology needs.

10. The federal government should continue toplay a major role in funding large-scaleinfrastructure projects that are used exten-sively by many researchers, such as exper-imental energy generation facilities andthe Hubble Space Telescope.

11. As CED has recommended in previousreports, the United States must raise acad-emic achievement in math and science ingrades K-12. To improve learning in math andscience, we urge the adoption of national stan-

5

dards, policies to increase teacher knowledgeand skills, and upgrades in classroom cur-ricula, facilities, and teaching materials.

12. The training of graduate students is anindispensable role of the research universi-ties. The federal government should makegraduate student training a higher priorityand increase its funding of scholarshipsand training grants. Research universitiesshould explore ways to reduce the time andexpense required to obtain a doctorate. Sincea majority of Ph.D. graduates will not returnto the research universities as faculty, grad-uate schools should offer training programsand mentorships to prepare their studentsfor employment outside of academe.

13. Industry-university relations and univer-sity patenting and licensing should bedirected towards maximizing benefits forthe society at large. As a general principle,

new knowledge from university basicresearch should be freely disseminated. Incases where new knowledge has commer-cial potential, however, patenting and licens-ing may be appropriate and in the publicinterest. However, these technology trans-fer activities should not dilute or compro-mise the basic educational and researchmissions of the university.

14. The United States should expand its effortsto benefit from international collaborationand the globalization of basic research. Tothis end, public and private policies shouldensure that the United States is an attractiveplace for researchers to live and work. Ourimmigration policies should be further liberalized to allow foreign scientists and engi-neers more long-term and permanent visas as well as more short-term visits to actas consultants, collaborators, and visitingscholars.*

*See memorandum by James Q. Riordan, page 80.

6

America’s basic research is one of the nation’sgreatest assets. Advances in fundamental sci-ence and engineering derived from basic research,combined with strong economic incentives fortechnological change and innovation, have madean enormous contribution to the economic pros-perity and social progress enjoyed by U.S. citi-zens. Achievements in basic science andengineering have also contributed to our nation’srole as economic, social, political, and militaryleader, which has helped bring about an histor-ically unique period of relative peace and stabilityin the world. It is clearly in the national inter-est to maintain and to strengthen America’scommitment to basic research and to improvethe productivity of resources used in basicresearch.

As in the past, the economic prosperity offuture generations will depend critically on pre-sent day efforts to sustain this country’s his-toric and fundamental commitment to basicresearch. Indeed, this commitment is a tremen-dous source of hope for the future; the solu-tions to many of society’s greatest challengesand the key to exploiting new opportunities—such as the cure for cancer and AIDS, unlockingnew productive potential from natural and man-made resources, and the antidote to loomingenvironmental concerns like global warming—will depend upon fundamental scientific insightsderived from today’s basic research.

It is important that policymakers and thepublic understand the benefits of basic research.Without an appreciation of these benefits, it will

be difficult to maintain our national commit-ment to the basic research enterprise, and thusmaintain our nation’s leadership in the future.

THE BENEFITS OF BASIC RESEARCH

Although basic research—experimental ortheoretical work undertaken to add to the knowl-edge of fundamental science and engineering—accounts for only 15 percent of total R&D in theUnited States, it has been a major factor in theimprovement of technologies, living standards,and life styles. For non-scientists, the contribu-tion of basic research may not be obvious becauseof the complicated “trail” between research, dis-covery, and innovation. One reason the trail is com-plicated is that fundamental knowledge derivedfrom basic research is generally widely sharedand often exploited by scientists and entrepre-neurs who were not involved in the originaldiscovery. Moreover, the potential value of newdiscoveries from basic research may not beimmediately evident even to the discoverersand consequently the associated innovationsoften occur several decades after the initial dis-covery. By contrast, there may be only a fewmonths or years between applied and develop-ment research and commercial innovations. (Fordefinitions of basic, applied, and developmentresearch see Box 1, page 7.) Consequently, theresults of applied research can easily be observedin new or improved products and processesintroduced by the business firms that sponsoredthe research. Of course, basic, applied, and devel-

Chapter 2

UNDERSTANDING THE CONTRIBUTION OF BASIC RESEARCH

7

Traditional DefinitionsScientists and R&D sponsors have traditional-

ly divided R&D activities into three categories:basic, applied, and development research. Al-though it is often difficult to place a specificresearch project exclusively in one of these cate-gories, the data on R&D activity published bythe federal government continue to follow thisdivision. These three categories may bedescribed as follows:

• Basic research is experimental or theoreticalwork undertaken to add to fundamental sci-ence and engineering knowledge. Thisknowledge is often drawn upon in subse-quent basic or applied research. Althoughbasic research can be exploratory, withoutany particular application in mind, the vastmajority of basic research is directed towardachieving new science or engineering knowl-edge in areas of interest to funders.

• Applied research includes investigationsthat draw from basic research or otherapplied research to create new knowledgethat in turn can be used to develop new orimproved products and processes.

• Development research draws on existingknowledge gained from basic and appliedresearch and from practical experience for thepurpose of creating innovative new productsor processes, as well as incremental improve-ments for existing products or processes.

The Discovery Process and Alternative Definitions

Despite the traditional distinctions and bound-aries between basic, applied, and developmentresearch, it would be a mistake to draw the con-clusion that the best description of the discoveryand innovation process is a linear model whichbegins with basic research, proceeds to appliedresearch and ends in development. Discoveryand innovation often do not proceed in such asequential and uni-directional fashion.

Analysis of individual cases demonstrates that

technological advances occur in an interactive,interwoven process where breakthroughs takeplace both before and after basic research. Thereare numerous examples of technological break-throughs occurring well before the basic sciencewas understood (see bullet on “Xerography”page 9). This interactive discovery and innova-tion process—often described as a “chain link”or “continuous” model incorporating feedbackloops—appears to be more characteristic of theR&D process than the linear model.

Given this deeper understanding of the dis-covery process, and its rejection of a linearmodel, characteristics other than the traditionaldefinitions have been proposed to delineatebasic research. For example, it has been suggest-ed that the time period between research and itseffects on output is one useful point of distinc-tion. There have been many instances where along period—often 10 to 20 years—has elapsedbefore new knowledge derived from basicresearch had any influence on output. However,there are also cases where the time lapsebetween basic research and innovation is veryshort. Moreover, many business firms haverecently instituted reforms in the researchprocess designed to shorten this time lag.

An additional distinction between basic andapplied research is based upon the extent towhich the results are shared with others.Fundamental knowledge tends to be widelyshared, because, unlike most products, its useby another individual does not reduce its avail-ability to those who made the discovery (seeBox 2, page 12). By contrast, the dispersion ofthe results of applied research is generally morecircumscribed by patents or secrecy so that thefruits of discovery can accrue to the firms spon-soring the research.

The Myth of Untargeted andUnmanageable Research

Perhaps because individual discoveries areoften serendipitous and the discovery process

Continued on page 8

BOX 1

CHARACTERISTICS OF RESEARCH AND THE DISCOVERY PROCESS

8

opment research are all vitally important foreconomic progress. However, basic research isthe foundation of most technological change, asexplained below.

Although the critical role of basic research inthe development of new technology is some-times easily observed, as in the case of biomed-ical research undertaken explicitly for the purposeof developing new drugs, more often the rela-tionship between fundamental science and newor improved goods and services is quite complexand indirect. Only retrospective examination ofspecific cases makes evident that basic researchhas been the foundation of many revolutionarytechnological innovations. The following aresome examples:

• Lasers owe their heritage to basic research bya number of scientists including: AlbertEinstein, the first to recognize in 1917 thetheory of “stimulated emissions;” CharlesTownes of Columbia University who in 1958discovered how to create a focused microwavebeam; Townes and Arthur Schawlow (of Bell

Labs) who published the theory of how stim-ulated emissions would work with shorterwavelengths, including those in the spec-trum of visible light; and Theodore Maimanwho constructed the first laser at Bell Labs in1960. Today, laser applications have a wide vari-ety of applications, including surgery, telecom-munications (lasers with fiber optics), printers,precision drills, and other machine tools.

• X-rays were initially and fortuitously discov-ered in 1895 by William Roentgen, as he wasexperimenting with cathode rays. Since then,many scientists and engineers have devel-oped diverse uses for x-rays. The best knownapplications are in the medical field. In morerecent years, the value of x-rays has beenenhanced greatly by the mathematical con-tributions of physicist A.M. Cormack (forwhich he won a Nobel prize). Cormack’s workcontributed to the development of computerizedaxial tomography (CAT), which has revolu-tionized medical imaging and diagnosis. Today,CAT scans produce three dimensional x-ray

itself is inherently complex, it is sometimesassumed that basic research cannot be targetedtoward objectives or managed effectively. But infact, R&D investments, including basic research,can be and generally are, carefully managedand directed to achieve specific objectives.

An institution or an individual invests inbasic research within a particular context, in asetting in which the research investment isaimed at the creation of valuable results. In auniversity, research makes a contribution to theeducation of students, and to fields of knowl-edge, which may impact the economy, the envi-ronment, the health of the populace, or thesecurity of the nation. Government laboratoriesexist in the context of a particular governmentmission, be it health care, energy, defense, agri-culture, etc., which affects the value context ofthat institution’s basic research efforts. Thesame is also true in industry research, where theobjectives and applications are clear.

Management of all research, including basicresearch, is aimed at maximizing the value thatis likely to be created from the research invest-ments, where the notion of value is particular tothe mission and setting of the institution.Implementing this principle forces the institu-tion and the individuals within the institution,to develop a clear understanding of what valueis, and what can be done to increase it. Choicesof research areas, for both basic and appliedresearch, reflect this understanding of value.Realistic measures of progress and of successare developed in the value context, andemployed to influence the course of research. Inshort, a set of processes, deeply involving theresearchers themselves, but including the other stakeholders is created, affecting basic researchat many stages—in the allocation of fundingacross sectors, the decisions on resources forparticular efforts, and the measures of progressand results.

Continued from page 7

9

images. X-rays are also used for the detectionof internal stress defects in materials and in theassembly of small electronic microcircuits. X-rays also have security applications, such asin airport baggage inspection.

• Semiconductors were first discovered in 1886by German chemist Clemens Winkler, butremained little more than a laboratory oddi-ty for many years. Before World War II someuses were found for semiconductors (in radars,for example), but wide-ranging use waited untilthe advent of transistors in 1948. Semiconductorshave since created a revolution in electronics.Tiny electric circuits are now used in every imag-inable electric device—miniature radios, tele-vision, telephones, airborne navigational aids,diagnostic instruments, etc. Transistors weredeveloped at Bell Labs by American physicistsWalter Brattain, John Bardeen, and WilliamShockley. Just as the transistor replaced the vac-uum tube, integrated circuits and micro-processors replaced transistors.

• The Global Positioning System (GPS) owesits origin to theoretical research on atomicstructure and to the construction of an atom-ic clock. Work on an atomic clock, whichbegan before World War II, was initially under-taken by researchers interested in the effectsof gravity on time. These researchers cer-tainly did not anticipate the contribution of theirwork to a GPS constellation of 24 satellites thatmake it possible to provide extremely accu-rate information on location. GPS has had amajor impact on transportation and its economicimpact is anticipated to grow very rapidly(e.g., in the “smart” highway program).

• Treatment for HIV disease and AIDS—espe-cially the impact on HIV protease inhibitorson slowing disease progression and pro-longing survival—resulted from more than 10years of public and private sector funding ofbiomedical research in several areas includ-ing immunology, virology, and biochemistry(see Merck discussion in Case Studies sec-tion). Research at the NIH elucidated thepathogenesis of HIV infection and the slow butsteady deterioration of the immune system by

HIV that eventually leads to AIDS, and helpedcharacterize the molecular composition ofHIV. Research conducted by academic andother nonprofit laboratories demonstratedthe capacity of HIV to replicate and seednumerous reservoirs of virus throughout theimmune, central nervous, and gastrointesti-nal systems. Research conducted by severalpharmaceutical companies contributed fun-damental knowledge about the role of HIV pro-tease enzyme in the HIV life cycle—includingits molecular 3-D structure and the mechanismof the emergence of viral resistance—andeventually led to the design of several potentHIV protease inhibitors that prevent replicationof the virus and reduce viral levels in thebody.

• Xerography, invented by Chester Carlson in1938, had a profound impact on informationprocessing (see Xerox discussion in CaseStudies section), but that impact was delayedby a quarter of a century. Many advances intechnology, basic science, and engineeringwere necessary before Carlson’s discoverywas commercialized. Research in the privatelabs of Battelle Memorial Institute providedthe foundation for Haloid-Xerox (now XeroxCorporation) copier products, more than 20years after the original discovery by Carlson.Xerox acquired a license to the process in1947 and assembled a large team of scientistsand engineers to further advance the xero-graphic process, filling in the gaps in funda-mental knowledge that stood in the way ofnecessary product improvements. Xerox spon-sored both basic and applied research andinvolved scientists and engineers in manydisciplines, generating advances that led tothe modern high-speed copier.

• The Internet. When BBN and others builtthe ARPANET—the forerunner to the Inter-net—in 1969, its primary goal was to enablescientists and engineers across the countryto share ideas and information. But its effortsalso brought together more than a centuryof fundamental research and discovery. Thesize, speed, reliability, and cost-efficiency ofnetwork switching equipment were initially

10

based on advances in miniaturization of elec-tronics—particularly digital electronics—whichhad its origins in the development of logiccircuits in the 1850’s. Early circuits were builtwith various complex forms of electricallycontrolled switches, such as the Strowger step-by-step switch, which was the heart of thefirst completely automated telephone exchange.The invention of the transistor made possibleeconomical and reliable packet switching,which depends on a robust mesh of largenumbers of relatively inexpensive, dedicat-ed computers able to run unattended for longperiods. The links between these switching ele-ments also rest on the results of basic researchin information theory. In the 1940’s, ClaudeShannon, a scientist at Bell Labs, presented ameans of symbolically analyzing the behaviorof switching circuitry. Shannon’s work was close-ly related to the system of symbolic logicdeveloped by George Boole in 1848, and whichhas come to be known as Boolean algebra.Today’s scientists and engineers still rely on thesedevelopments as they continue to enhancethe Internet with remarkable speed.

Hundreds of additional cases could be citedto illustrate that the economic and social effectsof basic research are pervasive in our society.

The innovations described above and manyothers demonstrate certain characteristics oftechnological advances:

1. The benefits of basic research are large,widely dispersed, and frequently unan-ticipated;

2. New scientific discoveries and technolog-ical advances generally have a rich historyof basic science behind them, often build-ing on and extending the work of others;

3. Technological advances often combine dis-coveries in several fields—see the aboveexample where x-rays, mathematics, andcomputer technology were combined todevelop CAT scans;1

4. Knowledge resulting from basic researchtends to be widely dispersed and employedby researchers in other fields;2

5. Even applied and development researchcan help drive new basic discoveries:

• by developing new tools and instru-mentation for use in basic research;

• applied researchers frequently must stepback and perform basic research to fill gapsin fundamental knowledge that arerequired to achieve their practical goals;

• development researchers often make newdiscoveries in the course of their appliedwork that adds to our basic understandingof nature.

For further observations on the discovery andinnovation process see Box 1, page 7.

THE ECONOMIC IMPACT OF BASICRESEARCH

There is strong evidence indicating that basicresearch in science and engineering has made amajor contribution to the growth of the U.S. econ-omy. The study of individual cases shows that dis-coveries and innovations derived from basicresearch have led to new products and industriesthat employ thousands of workers nationwide. Ofcourse, technological change resulting from basicresearch often results in temporary displacementof workers in older industries, as well. And basicresearch is only one of several factors—such as theeducation of the labor force and the stock of cap-ital—that accounts for economic growth. Thus, iden-tifying the net impact of research expenditures onnational economic growth generally involves theemployment of sophisticated models of economicgrowth, an area of economic research that hasbeen ongoing over the last four decades.3 A con-servative interpretation of the results of this workindicates that total R&D accounted for 12 to 25 per-cent of the annual growth in productivity duringpost-World War II decades.4 This suggests that thecumulative impact on living standards has beenvery large. Moreover, a new view of economicgrowth suggests that R&D has dramatically larg-er economic effects.5

Basic research is a critical part of this contri-bution to growth. One reason that basic research

11

is so important is that advances in fundamentalknowledge tend to be widely dispersed, withdiscoveries extended by other scientists, andbuilt on in ways that often result in enormous economic benefits for society over a prolonged period. Also, as noted earlier, advances in appliedand development research activities are oftencritically dependent on earlier advances in fun-damental science and engineering derived frombasic research.

The Rate of Return on Basic ResearchInvestments

Another approach for gauging the impact onproductivity of basic research, and of R&D gen-erally, is to estimate the internal rate of return onR&D investments made by individual firms orindustries.6 Many such studies provide strong evi-dence of the economic benefits of R&D.7 Rate ofreturn studies measure both the increase in pro-ductivity experienced by the industries fundingR&D and the “spillover” benefits that improveproductivity in other industries. Although the rangeof estimated rates of return is quite wide, the con-sensus is that on average private returns arevery high relative to other investment opportu-nities—on the order of 20 to 30 percent annual-ly, or roughly double the average historicalreturn to stock market investments. Moreover,“social” returns on R&D investments—that is, thereturns to society as a whole—are substantiallyhigher than private returns.8 The social returnsto basic research are often particularly high due,in part, to the wide dispersion of fundamentalknowledge, which frequently leads to additionaldiscoveries and applications in diverse fields.

The difference between private and socialreturns on investments in basic research, oftendescribed as a “market failure,” is an importantjustification for public funding of research andfor tax policies to encourage increased research(see Box 2, page 12). Business investments inresearch are driven by expected private benefits.In order to maximize the benefits to society as awhole, the optimal government investment strat-egy is to provide additional funding for thoseresearch activities that offer social returns inexcess of private returns.

Although basic research investments are com-monly viewed as “high risk,” this best describesthe private sector’s, rather than the public sector’s,investments in basic research. Aprivate firm mustcapture some minimum level of benefit over a rea-sonable period of time to justify its investment.Because the results of basic research are oftenunpredictable (in outcome and in timing) andwidely dispersed, capturing adequate returns atthe firm level is difficult, and consequently risky.But for government, the returns to public invest-ments in basic research need not be captured byan individual entity—be it a researcher, univer-sity, or agency—in order to justify the invest-ment. Rather, so long as the results are widelydispersed and used by many different individu-als and institutions, the public benefits from thegovernment’s investments. The risk, then, is muchsmaller for these public investments than it wouldbe for a comparable investment by a private firm.

Further, the “portfolio” of basic researchinvestments which the government holds is farlarger and more diversified than that of anyindividual firm. As a result, risk associated withthe whole universe of government-funded pro-jects, which will include many winners andlosers, is much lower than would ever be pos-sible within a single firm.

Regional Economic Benefits

Basic research performed in major research uni-versities (and in other public and private labs)often has a large indirect impact on the econo-my of the regions where the universities arelocated. For example, a study by BankBostonshows that research conducted at MIT has hada large impact on the economy of Greater Boston,where numerous knowledge-based companiesare located.9 There are now estimated to be morethan one thousand MIT-related companies locat-ed in Massachusetts with world-wide sales of $53 billion. About 125,000 workers are employedby these companies in Massachusetts. World-wideemployment of these MIT-related companies isabout 353,000 workers. Of course, many otherregions of the nation, such as the Silicon Valleyof California and the Research Triangle of North

12

Carolina, have benefited similarly from the pres-ence of major research universities.

PUBLICLY-FUNDED BASICRESEARCH IS CRITICAL TO PRIVATE SECTOR INNOVATION

As described in Chapter 3, universities are thelargest venue for basic research, but it is also

performed in government laboratories, industrylaboratories, and in other private nonprofit insti-tutions. In several of the historical cases mentionedearlier in this chapter, the basic research under-lying technical advances was undertaken in pri-vate labs operated by business firms. Nevertheless,empirical evidence on industrial innovation inearlier decades found that many important newproducts and processes in the marketplace could

The federal government is by far the mostimportant funder of basic research (seeAppendix 1 for an overview of government andindustry resources for basic research). Whataccounts for this role? There are two generalexplanations, one defined by the needs of thegovernment itself, and the other by the inabilityof private markets to adequately respond to theneeds of the economy and society at large. First,government funds basic research, and R&D ingeneral, simply as a matter of procurement. Justas government purchases pens and pencils tocarry out administrative tasks, it also “purchas-es” research to support agency missions. In thissense, the Department of Defense funds researchin physics and computer science in order to gen-erate knowledge that will lead to better defensesystems. In general, this explanation appliesmore strongly to development activities than itdoes to basic research, which explains the pre-dominance of federal development spendingover basic research spending. That is, short-term, well-defined needs take precedence overlonger-term, less-defined needs.

But there is a second, broader category ofexplanations for why government is the primaryfunder of basic research. These explanationstranscend the narrowly-defined needs of thevarious federal agencies. As we discuss through-out this chapter, basic research has resulted incountless social and economic benefits in theUnited States. The very fact that the benefits ofbasic research are so broadly realized explainsthe central government role in its support.

In economic terms, the knowledge that results

from basic research is, by and large, a publicgood. Unlike private goods, a public good can beused simultaneously by any number of individ-uals without anyone’s use diminishing its sup-ply. For example, one person’s use of a scientificformula does not exclude another person’s use,nor does it diminish the formula in any way.This general characteristic has many implica-tions, one of the most important of which is thatit makes private ownership of a public good dif-ficult and economically inefficient.

Because an individual or firm cannot easilyreap all of the benefits of the scientific formula(or other potential outcomes from basicresearch), these private market actors tend tounderinvest in basic research activities from soci-ety’s perspective. As a result, a gap emergesbetween the prevailing level of private invest-ment in basic research and the level that wouldmaximize the benefits to society at large.Economists identify this gap as a market failureand point to government intervention as neces-sary to fill this funding gap and exploit the posi-tive externalities of basic research.

The Pfizer case study (see Case Study section)puts the distinctions of public and private sup-port for R&D in very practical terms as theyapply to drug innovation: “Industry and acade-mia each play vital yet different roles in [theinnovation] process, with industry focusing onintegrating basic science findings in a directed,applied process of managed drug develop-ment...This development enterprise is consider-ably different from the inquisitive, open-endednature of basic academic research.”

BOX 2

WHY DOES GOVERNMENT SUPPORT BASIC RESEARCH?

13

not have occurred without academic research.10

Today industry continues to perform basic researchin many areas (see Appendix 1 for an overviewof industry support for basic research), but height-ened competitive and financial pressures haveencouraged many firms to place more emphasison short-term applied and development work (seeChapter 3). Consequently, government-funded (primarily university-based) basic research isnow more critical to industrial technology.

The growing importance of publicly fundedresearch in industrial innovation was recently con-firmed in a 1997 study based upon an analysisof citations in patents issued to U.S. industry.11

U.S. patent law requires that the front pages ofpatents report “other references,” i.e., impor-tant prior art upon which the patent improves.The study examined 430,226 citations on thefront pages of 397,660 patents issued from 1987to 1988 and 1993 to 1994. This analysis revealed that73 percent of the research papers cited by industry patentsreferred to “public science”—government fundedresearch reported in papers authored by scientistsworking in academic, governmental, or other insti-tutions. Only 27 percent of cited materials wasauthored by industrial scientists. Moreover, acomparison of citations in the two time periodsshowed rapid growth in the dependence of pri-vate technology on public science.

Although much of the research undertaken inuniversities is intended to advance fundamen-tal knowledge in science and engineering, it fre-quently has very practical applications. Indeed,in recent years many talented scientists employedby research universities have participated incooperative research efforts with businesses andconsequently, a growing proportion of theirresearch has become directed at commercially valu-able objectives. Moreover, the major researchuniversities have very active technology trans-fer offices working to assist businesses to acquirenew technology. As noted in Chapter 3, whileresearch universities perform large portions ofthe basic research underlying commercial tech-nology, they also benefit society by trainingfuture scientists and engineers. In fact, the train-ing of future researchers is the most importantthing that universities can do to ensure a strongfuture for basic research.

Because of the characteristics of basic research,government is its primary funder (see Box 2, page12). CED strongly disagrees with the “newthinking” currently in vogue among some com-mentators12 which holds that government fund-ing of basic research is unnecessary in a freemarket economy and that a reduction in fund-ing for basic research would have little eco-nomic effect because business innovation reliesprimarily on existing technology. The evidenceon patent references described above indicatesthat this “new thinking” is seriously flawed.Technological development in industry is high-ly dependent on publicly funded basic researchundertaken in universities and elsewhere.

THE FUTURE IMPACT OF BASICRESEARCH

Basic research has been tremendously impor-tant to our economy and society throughoutthis century. But what of the next century? Doesbasic research promise to bring just as big apayoff to as many areas of our lives in thefuture? The answer, we believe, is a resoundingyes. The challenges and opportunities facingour economy and society in the years to comeare varied and plentiful. And in case after case,advances in fundamental scientific knowledgeare necessary before we can hope to see reme-dies or benefits.

For example, as computers and informationtechnologies become increasingly sophisticated,our expectations for future progress will beunrealized absent advances in fundamentalknowledge. The hopes for a “computer thatlearns” require significant scientific break-throughs—possibly in DNA research as well asin computing13—and not simply steady tech-nological improvements and tinkering.

DNA research itself holds great promise forbenefits in all aspects of human health and well-being. But moving from the current process ofmapping genes to that of understanding themaps themselves and the significance of genesequences is an enormous challenge that will occu-py scientists for decades.

Basic research will also be important in iden-tifying and defining social problems that vex us

14

today. For example, the current debate swirlingaround global warming is a measure of ourignorance in this area. With scientific progress wecan move beyond political contention to prop-erly defining problems and determining reme-dies, which themselves will likely be derived frombasic research.

Of course, the prospects for future scientific break-throughs in specific areas are largely unknowableto us today, and in many cases, appear almostunfathomable. Ideas that currently exist in theo-ry would, if brought to fruition, redefine the waywe exist as human beings and as a society.“Nanotechnology” is but one remarkable exam-ple. The ability to manipulate molecules sys-tematically, today a theoretical proposition, mayoffer the potential to live in a world of science fic-tion, where nature can be replicated in a million

different ways to any number of ends. Basicresearch itself might ultimately prove molecularengineering to be fantasy rather than reality; butspeculating about its potential underscores the dra-matic nature of scientific progress and the tremen-dous unknowns that are associated with it.

The list of potential outcomes in basic researchis endless, but in considering the opportunitiesand challenges of the future, one way to think ofbasic research is as low-cost insurance—cur-rently, basic research uses less than one-half ofone percent of GDP to create very significant long-term economic and social gains.14 We have onlythe slightest understanding of what lies ahead.But our greatest hope for capitalizing on unknownopportunities and avoiding unknown calamitiesis investing in the scientific knowledge that willmeet these unknowns in the decades ahead.

15

The success of basic research in this countryis due in large part to the unique characteristicsof the American research system. The Americanway of doing research is characterized by theflexibility and heterogeneity of its institutionsand disciplines, the intense competition forresearch funds, and the independence and creativityof individual scientists and engineers. More thananything else, the high caliber of our individualscientists and engineers accounts for the excep-tional quality of our basic research. In the wordsof one observer, American scientists are a “nation-al treasure chest.” Moreover, the economic impactof our basic research also reflects the respon-siveness of the U.S. economy to technologicalchange, innovation, and commercial needs.

The dynamic character of the U.S. basicresearch system is particularly evident now asthe entire R&D system is undergoing substan-tial change. The characteristics of this most recentwave of changes include the following:

• Because of increased competitive pressures, manybusinesses are taking steps to improve the man-agement of their research activities and the returnon R&D investments. This is reflected in (1)greater emphasis on applied and develop-ment activities, which have more timely andcertain payoffs, (2) efforts to become morecompetitive by reducing time-to-market; and(3) actions to reduce R&D costs, includingthe downsizing of large industrial labs.

• The priorities of the federal government are chang-ing. (1) The end of the Cold War, which re-

sulted in sharp cuts in defense appropriations,has placed basic research in areas traditional-ly funded by the Pentagon at a disadvantage,while funding for certain civilian priorities,particularly health research, continues to growrapidly. (2) Unless entitlement programs forthe elderly are reformed, entitlement spendingwill crowd out other discretionary expendi-tures, creating the potential for declines in fed-eral investment in basic research. (3) Withheightened concern for our international com-petitiveness, the federal government hasincreased funding of collaborative efforts by government and business to accelerate com-mercial technology, a change that threatens tosqueeze resources for future basic research.

• Modern technology is bringing forth radical changesin the research enterprise itself. The revolution ininformation technology has resulted in morerapid transfer of knowledge and increasedopportunities for one researcher to build uponthe discoveries of others. This trend is partic-ularly important in those fields of researchwhere modern technology permits rapid repli-cation of work that the original researchertook years to complete.

• In many cases university research has taken on agreater entrepreneurial character. Federal leg-islation enacted in the 1980s to facilitate eco-nomic development now permits universitiesto hold patents to, and license, work fundedby the federal government, and business hasincreased funding of research in universities.

Chapter 3

THE EVOLVING AMERICAN BASICRESEARCH SYSTEM

16

This trend is also explained by the ability ofindustry, particularly the biotechnology indus-try, to translate basic research into commer-cial products very quickly.

• Collaborative efforts between basic research insti-tutions—universities and business, business andgovernment etc.—and between disciplines aregrowing in importance. Rapidly rising researchcosts, particularly for capital equipment,have been an important motivation for insti-tutional collaboration. Inter-disciplinary col-laboration has been driven by the nature andcomplexity of the questions explored.International collaboration in research hasalso expanded, encouraged by the need to sharecosts, especially in projects with large capi-tal infrastructure requirements such as highenergy physics and space exploration.

• Increased foreign presence in R&D has intensi-fied in recent years, even in basic research. Othercountries are increasing their investments inR&D and basic research, including develop-ing countries which spent virtually nothing onR&D just a few decades ago. This trend isreinforced by the rapid dispersion of knowl-edge and the tendency of international cor-porations to locate research in countries whereskilled workers are abundant and specializedskills are available. The increased availabilityof opportunities for employment abroad hasalso reduced the number of foreign scien-tists—including those educated in the UnitedStates—choosing to locate in the United States.

This chapter briefly describes the qualities ofthe American basic research system that haveled to its success, as well as the evolving natureof these qualities: its institutional structure, thehuman infrastructure, and the system of funding.Finally, the chapter points to international trendsthat promise an increasingly global context forAmerican basic research in the years to come.

TODAY’S BASIC RESEARCH INSTITUTIONS

The current institutional structure for basicresearch is largely a legacy of important decisions

made by government and industry after WorldWar II and throughout the Cold War about howand where to allocate resources for basic research(see Box 3, page 17). One key characteristic of theAmerican basic research enterprise since thelate 1930’s has been the predominance of feder-al government funding relative to other fundingsources (see Box 2, page 12). Industrial and othersources of funding, though smaller, have alsoplayed important roles in supporting basicresearch (see Appendix 1 for an overview offunding patterns for basic research).15

Yet, financial resources alone do not accountfor the high quality of basic research in theUnited States. How, or to whom, those resourcesare allocated matters a great deal. Perhaps morethan any other country, the United States relieson competitive mechanisms in allocating fundsfor basic research. Competitive, peer-reviewedgrants to individual investigators are a hallmarkof the American system. The basic research insti-tutions that have succeeded in this allocationmechanism, primarily large research universities,have set the standard for quality in the post-Cold War era. On the other hand, research insti-tutions that have grown up under a differentregime, one defined by centralization and top-down management, have struggled to jus-tify their cost-effectiveness in an era of tightbudgets.

IndustryBasic Research Conducted by Industry

Industry is, by far, the largest funder of totalR&D in the United States, but its relative pres-ence in basic research is less significant (seeAppendix Figure 2 in Appendix 1). Basic researchconducted by industry is largely targeted at itsown proprietary product development—often fill-ing in the gaps remaining from publicly-supportedresearch and in areas of inquiry that are neces-sary to proceed with product development16—though a small amount of industrial basic researchis conducted for the government, particularly inthe area of defense.

Industry spending on basic research varies great-ly from sector to sector. A relative few, such aspharmaceuticals, are highly dependent on their

17

The basic research system that exists in theUnited States today—with world-class researchfunded and conducted by federal agencies, largecorporations, state governments, public and pri-vate universities, small businesses, and privatenonprofit research institutions—has come a longway from its infancy a century ago. At the turnof the century, scientific inquiry was still largelya European affair. At the time, the United Stateswas known for its individual inventors, theclever non-scientists tinkering in their homes todiscover a new product or process, but rarely afundamental scientific insight.

With the emergence of large, multi-productcorporations and the scientific and technologi-cal demands of World War I, the modernAmerican basic research enterprise began totake shape. In a drive to innovate, large firmsturned to scientists and engineers for expertise(most significantly in chemicals, petroleum, andelectrical machinery), establishing in-houseresearch laboratories or contracting out to inde-pendent labs. During these years, federally-sponsored research was largely agricultural,though the land grant universities did sponsorother important areas of research, often relatedto local and regional economic needs.

World War II was a watershed for Americanresearch, creating a public enthusiasm for scienceand technology that would carry over into thepost-war years. The aftermath of World War IIcreated an infusion of resources into basicresearch directed to military objectives, codifyinga system of public funding that flourished dur-ing the post-war period as it was enlarged toaddress civilian objectives as well. Under thissystem, basic research projects were funded bythe government and conducted in universities,government laboratories, and private companies.

Vannevar Bush’s post-war blueprint17 forpublic research led to the creation of theNational Science Foundation (NSF) and with it,a system of allocating federal dollars for all ofthe sciences based on scientific merit. Most ofthe federal resources devoted to research were

(and are) allocated through federal agency mis-sions. The federal government invested heavilyin research as a means to achieve supremacy inweapons systems and space technology.Government investment in health research alsogrew rapidly during this period, as public poli-cies sought to focus the nation’s scientific exper-tise on major diseases. Clearly, during theseyears, government funding for basic researchwas largely targeted, even if broadly so.

Whether funded by the NSF or by other fed-eral agencies, publicly-supported basic researchcame to be characterized as “individual investi-gator” research. For it was the individual inves-tigator who would “define problems and designthe best approaches to solving them.”18 In thecase of NSF grants, this process was predomi-nantly investigator-initiated. But even withinthe narrower confines of agency missions, basicresearch was largely driven by the expertise ofscientists, rather than through “task orientedmanagement by sponsors or ultimate users.”19

As the system for publicly-supported basicresearch took shape, many large U.S. firmsexpanded their own basic research activities.The Cold War era also marked an extraordinaryeconomic period during which many firmswere able to command significant marketpower. Companies like AT&T, Xerox, GeneralElectric, and Eastman Kodak used the resourcesfrom this market power to fund activities insome of the nation’s premier basic research lab-oratories. Indeed, work at industrial labs wouldlead to numerous Nobel prizes for scientificbreakthroughs.

The end of the Cold War marked a turningpoint for basic research in the United States. Bythe 1990’s, the federal labs, which sustained agreat deal of basic research after W.W.II, facedan uncertain future. Universities also preparedfor belt-tightening in response to an expecteddecline in federal support. As businesses facedan increasingly competitive and global market-place, they began to shift their R&D activitiestoward projects with short-term horizons.

BOX 3

THE HISTORICAL EVOLUTION OF TODAY’S BASIC RESEARCH SYSTEM

18

long-term basic research investments for new prod-ucts and processes, while others spend very lit-tle, if anything, on basic research. In the aggregate,industry places a much greater emphasis onshorter-term development activities in the allo-cation of its R&D dollars, leaving the majority ofthe nation’s basic research investments to the pub-lic sector.

But there is much more to the differencesbetween industrial basic research and basicresearch supported by the government (per-formed primarily in universities) than the rela-tive size of the investments. Basic research incompanies and basic research in universitiesmay both achieve new, fundamental scientificunderstanding, but they are characterized bydifferent motivations and different goals. The uni-versity researcher is attempting to ask and answerquestions of deep scientific importance, to gen-erate knowledge and theories with the mostpower to explain natural processes. The indus-try researcher has a very targeted goal: to gainunderstanding that is essential for further devel-opment of technology, which in turn, is expect-ed to drive development of new products. Whenuniversity researchers make an important basicdiscovery, they and their colleagues immedi-ately increase their efforts along similar lines toconfirm and amplify the discovery. Further, theacademic researchers will direct new effortstoward questions that are inherently most inter-esting. Once the company researcher has madea discovery, he or she attacks the next set ofproblems that must be solved for product devel-opment to continue, and may leave elaborationof the discovery to others (including universityresearchers).

Industry as Partner and Collaborator in BasicResearch

While industry performs a significant shareof the nation’s basic research, increasingly com-panies have also found it useful to support basicresearch projects at universities, and to collabo-rate with the individual investigators. Directindustry funding of “sponsored research” atuniversities, though small relative to govern-ment funding, has expanded significantly, from4 percent of total support for academic research

in 1980 to 7 percent in 1996 (see Appendix Figures8 and 9 in Appendix 1). Much of this expansionoccurred in the biomedical field where compa-nies have rapidly translated basic research dis-coveries into new product development programsin therapeutics, diagnostics and medical devices.

The willingness of industry to fund spon-sored research has been spurred, in part, by theBayh-Dole Act of 1980, a federal law whichallows recipients of federal grants to retain titleto inventions made in the course of researchfunded by the grant.20 In practice, this means thatan industrial sponsor may have the right tolicense proprietary rights not only to the workwhich is being sponsored, but also additional rightsto earlier inventions funded by governmentgrants. Thus, the industry sponsor may be ableto license a coherent “package” of intellectual prop-erty. In certain cases these rights may provide asufficient platform for the company to initiate itsown internal R&D toward the development ofnew products.21

At the major research universities in theUnited States, a sponsored research contractnegotiated between a company and the univer-sity’s technology transfer office has the follow-ing attributes: the company provides a specificamount of funds in return for the investigator’sagreement to perform an agreed-upon researchplan; the company receives an option to licenseany new inventions, which takes the form of afirst right of negotiation; the company agrees dur-ing the option period to fund any related patentapplications filed by the university; and thecompany is given a brief period in which toreview for patentability any written or oral pre-sentations (such as conference presentations orjournal articles) prior to their public disclosure.The latter understanding maintains the univer-sity’s right and need to freely disseminate itsresearch findings, while preserving both thesponsor’s and the university’s interest in thevalue of any proprietary inventions.

With appropriate safeguards for the role of theuniversity,22 industry sponsorship of basic researchcan bring benefits to both parties. The industrysponsor gains access to some of the most creativeminds in its field of interest, which can stimulateimportant breakthroughs. The company is also

19

able to license technological innovations andassociated patent rights that can be the basis forimportant new product lines. Indeed, the biotech-nology industry originated with the formation ofnew companies around technologies that hadbeen licensed from universities.

The university benefits by allowing its facul-ty members to supplement publicly-funded grantswith additional support. For many researchers thisadditional support contributes to maintainingthe productivity of their laboratories. In addi-tion, if a company converts technology licensedfrom the university into a successful product orservice, the university receives milestone pay-ments and royalties, which in the case of a block-buster product (such as a major pharmaceutical)can amount to millions of dollars per year. An addi-tional intangible benefit of a well-managed col-laboration can be the value of the intellectualcollaboration between the individual researchersin the industry and university laboratories.

Industry Benefits from Open Dissemination ofUniversity Research

The increasing significance of industry-uni-versity research partnerships does not diminishwhat continues to be the primary channel ofbenefit of university basic research to industryand the economy at large: the open dissemina-tion of new knowledge, without restriction on useor on the number of users. The primary benefitof university research to society stems from thefree and open dissemination of new knowledge,whether the research is funded by the govern-ment, foundations, or corporations, and whetheror not the research leads to patentable inventions.Indeed, most technology transfer agreementsat major universities are designed to permit onlya brief delay in public dissemination of indus-try-sponsored research for this reason.

As discussed in the previous chapter, recentanalyses of patent data suggest that the dis-seminated results of university research hasaccounted for a great deal of the new knowledgeembodied in commercial patent applications.There is also evidence that open disseminationof university research—through publications,public meetings and conferences, and throughinformal channels—has been much more impor-

tant to industrial innovation than more restric-tive relationships.23

Open dissemination of basic research doesnot occur solely in universities. The case of AIDSresearch at Merck (see Case Studies section)illustrates how and why many private compa-nies also subscribe to a standard of open dis-semination of knowledge derived from many oftheir basic research activities. As the Merck casedemonstrates, even when there are no legal bar-riers to establishing proprietary rights overresearch findings, companies like Merck havefound that a more open approach to disseminationnot only yields greater benefits to society atlarge, but in many cases to themselves. Also,industry researchers are, like university col-leagues, motivated by prestige to disseminate theirwork openly through peer-reviewed journalswhenever possible. The prestige generated by pub-lication benefits the company by enhancing itsrecruitment of top-notch scientists and increas-ing the perceived value of the company’s research,and ultimately its products.

Universities

Without question, the most important insti-tution in American basic research is the researchuniversity. The research university system hasbecome the nation’s largest basic research enter-prise as a result of large and sustained federal fund-ing throughout the post-World War II period(see Appendix 1 for an overview of resources foruniversity research). But the real success of thesystem is based upon:

1. the unique structure that supports the individualuniversity researcher, rather than the institu-tion itself;

2. the dual role that universities play in con-ducting research and training graduate students;

3. the competitive funding mechanisms for uni-versity research;

4. and the flexibility and diversity of research uni-versities.

As we describe later in this chapter and in theprevious section, the success of the universityresearch system has contributed to a great deal of

20

business interest in recent years, leading to newroles and challenges for universities and theirresearchers in the marketplace.

University Support for the Individual Investigator

The heart of the university research enter-prise is the highly-skilled independent investi-gator (see “The Critical Value of the IndividualResearcher,” page 27, for further discussion).Typically tenured or tenure-track faculty mem-bers,24 university scientists rely on a combinationof university support (in the form of salary, facil-ities, and research assistants) and external grantsto conduct their research. The university researchenvironment allows faculty scientists consider-able autonomy to define the nature of theirresearch projects, to pursue external support forthe projects, and to carry out the research in a waythat is suitable to the researcher and the fundingagency (or other entity). The quality of univer-sity research in general is guided by a compet-itive process through which thousands ofindividual university scientists nationwide seekexternal (mostly federal) grants. Of course,department chairs, deans, and university admin-istrative officials also play important roles inshaping and ensuring quality in universityresearch portfolios. Particularly important inthis regard is the extensive review undertakenin hiring faculty members.

As the trends in federal support for univer-sity research (outlined in Appendix 1) illustrate,university scientists often seek support from a num-ber of agencies, because the fundamental natureof their work can support a number of federal mis-sions.25 (For example, the same grant applicationmay be eligible for consideration by the NationalScience Foundation, the Department of Defense,and the National Institutes of Health). The plu-rality of routes to funding has increased grantopportunities for applicants and sustained theflow of new knowledge despite shifting agencypriorities and budget fluctuations (see Appendix1 for a description of these funding patterns). Theavailability of multiple funding sources has per-mitted a greater diversity of scientific investigationsand approaches than would be possible undera single-source funding structure.

The educational function of universities givesthem a particular advantage in basic research. Asan integral part of their training in science and engi-neering, graduate students are employed in uni-versity labs as assistants to the faculty researchers.Graduate students at research universities aretechnically proficient, intimately familiar with cut-ting-edge research, and highly motivated. TheirPhD theses and their future careers as researchersdepend on making a significant intellectual con-tribution by performing research that is worthyof publication in first-tier, peer-reviewed journals.Indeed, the laboratory experiments that lead tofundamental discoveries are typically performedby graduate students and post-doctoral fellows,whose tuition and stipends are funded by gov-ernment and foundation grants.

The Diversity of Research UniversitiesNearly all university research (96 percent)

conducted in the United States is concentratedin about 200 public and private institutions.26

Twenty-five universities (see Figure 1) accountfor about 35 percent of total research expendituresby the 3,600 higher education institutions in theUnited States. These top 25 recipients account for39 percent of federally-funded research. The top100 account for 78 percent of total federal fund-ing—representing only modest diffusion ofresources over four decades of expansion of theuniversity system in the United States.

Even though the diffusion of total resourceshas not changed dramatically, many more aca-demic research institutions receive federal researchfunds today than did so 25 years ago—from 567institutions in 1971 to 875 in 1993.27 Almost allof this increase has gone to non-doctorate grant-ing universities—i.e., liberal arts, two-year com-munity colleges, and other technical andprofessional schools.

Given the emphasis in the United States on inves-tigator-initiated grants, it follows that the repu-tation of a research university and its collectiveimpact as a research institution are the aggre-gate of the productivity, reputation, and grants-manship of its individual faculty members and thestudents they recruit to their laboratories. Asindividual scientists and engineers move from oneuniversity to another, the financial and educational

21

benefits of their research activities move withthem. Consequently, while there are differencesin quality and in economies of scale among insti-tutions which tend to sustain institutional patternsof research funding over time, the hierarchy ofresearch universities that receive the most federal

research support is not static. Competition forresearch grants by individual investigators has asimilar competitive effect on the academic insti-tutions that employ them. It is in universities’ inter-est as producers and disseminators of knowledgeto employ the highest quality scientists.28

Figure 1

Top 25 Research Universities, by Total Research Expenditures and Federally FinancedResearch in 1995

Total Research Expenditures Federally Financed Research(dollars in thousands)

Total, all institutions $21,654,220 Total, all institutions $12,884,158

1 University of Michigan $443,070 1 University of Washington $291,2842 U WI Madison $403,541 2 U CA San Diego $284,4453 University of Washington $289,160 3 University of Michigan $275,9564 MA Institute of Tech 1/ $370,800 4 MA Institute of Tech 1/ $273,5435 Texas A&M University $362,539 5 Stanford University $273,157

6 U CA San Diego $357,333 6 Johns Hopkins Univ 2/ $259,0497 Cornell University 1/ $343,786 7 U WI Madison $299,3818 University of Minnesota $336,524 8 U CA San Francisco $224,2719 Johns Hopkins Univ 2/ $331,387 9 Cornell University 1/ $207,39110 Pennsylvania State U $330,881 10 Columbia University $206,495Total, top 10 institutions $3,669,321 Total, top 10 institutions $2,524,972

11 U CA San Francisco $329,742 11 Harvard University $203,96512 Stanford University $318,871 12 U CA Los Angeles $201,77313 U CA Los Angeles $303,668 13 U of Pennsylvania $200,89514 University of Arizona $292,351 14 University of Minnesota $194,81915 U CA Berkeley 1/ $291,200 15 Pennsylvania State U $187,481

16 Harvard University $276,422 16 Yale University $174,86817 U of Pennsylvania $272,393 17 University of Colorado $168,67418 University of Colorado $249,695 18 University of Arizona $168,79119 Ohio State University $246,287 19 U of Southern California $163,60620 U of Illinois Urbana $246,174 20 U CA Berkeley 1/ $157,826

21 Columbia University $244,991 21 U of NC Chapel Hill $156,62622 U CA Davis $244,116 22 Duke University $148,52623 Yale University $231,819 23 Washington University $146,92124 U TX Austin $228,676 24 University of Pittsburgh $144,45725 U of Southern California &222,159 25 U TX Austin $143,939

Total, top 25 institutions $7,667,885 Total, top 25 institutions $5,089,169

1/ These data do not include research expenditures at university-associated federally-funded R&D centers.2/ Johns Hopkins University data do not include Applied Physics Lab, with $447 million in total R&D expenditures.NOTE: Because of rounding, data may not add to totals.SOURCE: National Science Foundation, Science Resource Studies Division, Survey of Scientific and Engineering Expenditures atUniversities and Colleges, FY 1995 (data available at www.nsf.gov).

22

Research Universities and the Marketplace

Two important trends are helping to shape theface of tomorrow’s research university. Bothpromise to push university research closer to themarketplace. The first is the ability of universi-ties and academic researchers to reap financialrewards for inventions that are to be commer-cialized by industry. The other related trend isthe growing number of industry-university col-laborations and the increasing emphasis ofresearch institutions on developing such col-laborations into a major long-term source offunding.

For nearly 20 years it has been the policy ofthe federal government to encourage universi-ties (and other performers of government-sup-ported research) to file patents and to transferuseful inventions to industry for commercialdevelopment. As described earlier, the key leg-

islation related to this policy was the Bayh-DoleAct of 1980, which ushered in a new era foruniversity research. Figures 2 and 3 illustrate themagnitude of these changes. A gauge of theimpact of Bayh-Dole is the number of patentsissued to academic institutions, which increasedfrom 249 in 1974 to 1,761 in 1994.29

It is important to recognize, however, thatinvolvement of universities in commercializationof technology is still a small portion of totaluniversity research, and, indeed of total indus-try R&D. For example, the patents awarded bythe U.S. Patent Office to universities in 1994amounted to less than one percent of all patentsissued that year. Despite the growing presenceof industry in funding sponsored research on cam-puses, companies provide only seven percent oftotal university research support.

The open nature of university research hasalways allowed for external funding from manysources, not just the federal government.Nonetheless, the nature of industry fundingand the effects of university licensing raisesissues and concerns not encountered with pub-lic funding. These concerns include the poten-tial for:

• restrictions on open dissemination of research;

• a reorientation of university research to accom-modate the interests and needs of industrialsponsors;

• conflicts of interest between the individual finan-cial interests of researchers and the broaderinterests of the academic department, univ-ersity, and society as a whole;

• application of criteria other than basic researchproductivity and teaching performance, suchas ability to obtain industry support andgenerate royalties, in consideration of hir-ing and promotion decisions for faculty.

Most research universities have developed poli-cies and guidelines that confront these issues, andhave promulgated them among faculty, staffand graduate students (see Appendix 2 for twoexamples of university patent policies).Nonetheless, universities deeply involved insponsored research must now serve a new con-stituency, their industrial partners. The acade-

Figure 2

Number of U.S. Patents Awarded toAcademic Institutions

SOURCE: National Science Board, Science & EngineeringIndicators—1996 (Washington, DC: U.S. Government PrintingOffice, 1996), Appendix Table 5-42.

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

Nu

mb

er o

f P

aten

ts

1974to

1980

1981to

1987

1988to

1994

23

mics must cope with a new bureaucracy—the uni-versity’s Office of Technology Transfer. And theuniversity’s administrators must wrestle with arange of potential conflicts of interest on thepart of faculty and staff.

Patenting and Licensing of University Research

Because university inventions are frequent-ly made with the use of public funds, universi-ties have a special responsibility to maximize thesocietal benefit from the use of such inventions.This responsibility goes beyond the need tofreely disseminate the descriptions of the inven-tions, and extends to the manner in which suchinventions are commercialized, since patents, bytheir nature, imply a restriction on use.

Clearly, there is no simple formula by whichsuch decisions can be made. Many leadingresearch universities have technology transferpolicies that specifically identify the public inter-

est as the primary criterion for making tech-nology licensing decisions (see Appendix 2). Ingeneral, once a university has filed for patent pro-tection on a discovery, it then must decidewhether to commercialize the technology onan exclusive or non-exclusive basis.

Research advances that can be translatedinto distinct products or services, such as a newtherapeutic compound, a new polymer, a newelectronic instrument, or a new medical device,are typically considered appropriate candidatesfor exclusive licensing. In these cases, the pro-tections of both patent protection and an exclu-sive license are essential incentives for a companyto invest time, money, and resources in creatinga new product. Society benefits from the will-ingness of industry to make those commitmentsand bring a useful new product to market.

On the other hand, new knowledge thatserves as a research tool potentially benefiting

Figure 3

Growth in U.S. University Licensing Activities Relative to U.S. Totals,1991-95 Cumulative % Change

*Adjusted gross royalties are gross royalties minus royalties paid to other institutions.SOURCE: Association of University Technology Managers, Licensing Survey FY 1991-95, (ATUM, 1997).

0

20

40

60

80

100

120

Per

cen

tage

Ch

ange

only

Patent Applications Filed

U.S. total Universities

Licenses and Options Executed Adjusted Gross Royalties*

24

the development of a range of new products inmultiple companies, is typically commercial-ized through a non-exclusive license made wide-ly available on commercially reasonable terms.The most prominent example is the Cohen-Boyer patents, assigned to Stanford Universityand the University of California, which cover thefundamental techniques of gene splicing. Sincethese patents were issued, Stanford and theUniversity of California at San Francisco (UCSF)have made a non-exclusive license available toall companies involved in genetic engineering fora minimal annual fee starting at $10,000 and a smallroyalty on sales. These modest requirementshave imposed no undue burdens on industry, andin 1997 generated $38.5 million for the two uni-versities.30 Moreover, basic research institutionshave been able to practice the methods withoutcharge.

The distinction between products and tools isneeded, since a particularly important tool basedon a major breakthrough such as gene splicingmay be essential for the unfettered pursuit of addi-tional basic and applied discoveries. If excessivebarriers are placed in front of others’ entry intoa field, important new avenues of research willbe impeded. The policies of Stanford and UCSFtoward the commercialization of the inventionby Professors Cohen and Boyer have become amodel for other universities that own a broad-ly useful tool or process: license non-exclusive-ly; make the terms reasonable; and allow basicresearchers to use the technology at no cost.This approach optimizes the benefit both to soci-ety and to the university.

The Impact of Industry Funding on University Research

Initial evidence on the growing university-industry research relationship suggests theremay be some cause for concern. According to anumber of recent studies on the impact of indus-try support for academic research, industryfunding is associated with greater restrictive-ness in disclosure of research results and withresearch that is less basic in nature.31 This is notto say that industry sways the direction of uni-versity research toward applied and developmentwork. Rather, companies seek types of univer-

sity research projects that suit their needs, andthese types of research typically may not be asclearly basic in nature as those supported throughfederal grants. In the aggregate, there is no strongevidence that university research has shiftedaway from basic science and engineering, evenas industry support has increased substantially.

Yet, given the characteristics of industry fund-ing just described (greater restrictiveness and pro-jects that are less basic in nature), it clearly ismisguided to view industry and government fund-ing for university research as interchangeable. Asignificant change in the balance between gov-ernment and industry funding for universityresearch as a whole would likely impact thecharacter of university research and its dissem-ination.

Among the conflicts of interest receiving themost intense scrutiny are those that may be gen-erated by the university researchers who bene-fit personally from affiliation with companies.Faculty find themselves able to improve their finan-cial status through:

• participation in royalties on their inventions,(an explicit directive of Bayh/Dole);

• paid consulting to companies, particularly tocompanies that are sponsoring research inthe faculty member’s laboratory; and

• equity ownership in start-up companies.

In many cases, these potential areas of conflicthave been resolved to the benefit of the univer-sities, faculty members, and the companies withwhich they interact. Nevertheless, these rela-tionships are evolving rapidly; new precedentsare set each year; and universities must be vig-ilant in defending their fundamental educa-tional and basic research missions.

As the Columbia University case illustrates (seeCase Studies section), the leading research uni-versities are well aware of the complex nature ofuniversity-industry research relationships. Inthis case, the real and potential benefits are clearfor all parties: Columbia receives funding fromthe private firm (VIMRx Pharmaceuticals),VIMRx has access to potential knowledge thatmay translate into new products, and societymay ultimately benefit from a faster innovation

25

process due to the direct relationship between uni-versity and firm. The potential downside is alsorecognized by Columbia officials: the possibili-ty that the traditional openness in disseminationof basic research findings will be threatened;the narrowness of single company relationships;and the influence on the overall balance ofColumbia’s research portfolio, potentially awayfrom basic research. As a case study in progress,Columbia’s relationship with VIMRx illustratesthe benefits, costs, and unknowns of the researchuniversity’s new role in the marketplace.

The Federal Laboratories

Although a vast majority of federal R&D iseither not basic research or is performed outsideof the government on a contract or grant-awardbasis, the government itself does perform basicresearch through the federal laboratory system.In 1995, for example, federal laboratories account-ed for $2.7 billion or about 20 percent of feder-al basic research dollars (see Appendix Figure 6in Appendix 1).32 Historically, the federal labs havebeen a very important venue for basic research,generating many important discoveries and sci-entific insights in fields as diverse as the healthsciences and nuclear physics.

The management structure of federal labsvaries. Some are operated directly by the gov-ernment (e.g., the National Institutes of Health,the National Institute for Standards andTechnology, and the U.S. Geological Survey),while others are operated for government byprivate or nonprofit entities, including univer-sities and other nonprofit organizations (e.g.,Los Alamos, Oak Ridge, and the Jet PropulsionLaboratory). Among the largest of the 700 labsfunded by the federal government are the so-called“national labs,” administered by the Departmentof Energy and historically charged with energyand defense missions. These labs perform a mixof basic, applied, and development research and command substantial R&D budgets (seeFigure 4).

Since the end of the Cold War, the missionsof the Department of Energy national labs havebeen in flux. These labs have moved to dealwith the problem of disappearing missions by

seeking new “missions,” often related to civil-ian technology needs and the international com-petitiveness of U.S. industries. For example,Intel, Motorola and Advanced Micro Devicesrecently announced a commercial research part-nership with three national labs to developcomputer chips using extreme ultraviolet technologies.33

Critics have charged that such efforts arewasteful of federal tax dollars, and that the num-ber and size of labs administered by theDepartment of Energy are no longer justifiedby the mission of the agency that supports them.In the case of the computer chip partnership, therehas been further criticism due to the potential forforeign firm participation in the project, whichwould amount to a federal subsidy of foreign com-panies.34 This criticism highlights the political problems that can arise as agencies attempt to justify initiatives under the “national competi-tiveness” rubric.

The problems of many federal labs are not lim-ited to their missions. Of all the institutions thatperform basic research, the DOE labs are said tobe the most lacking in flexibility and cost-efficiency,

Figure 4

Federal Obligations for Research andDevelopment at the “NationalLaboratories”

Fiscal Year 1995(Dollars in Thousands)

Total National Labs $3,012,548

Idaho National Engineering Lab $77,745

Oak Ridge National Laboratory $288,332

Sandia National Laboratories $654,472

Argonne National Laboratory $252,879

Brookhaven National Laboratory $216,094

Ernest Orlando Lawrence Berkeley National Laboratory $170,870

Lawrence Livermore National Lab $500,622

Los Alamos National Laboratory $102,278

Pacific Northwest National Laboratory $208,529

SOURCE: National Science Foundation, Survey of Federal Funds for R&D:FY 1995, 1996, 1997, (data available at www.nsf.gov).

26

resulting from an overly-centralized, micro-managed resource allocation system.35 Indeed,governance reform was a key area of concern inthe 1995 Galvin Task Force, charged with advis-ing the Secretary of Energy on alternative futuresfor the national laboratories.

Unlike the decentralized, competitive inves-tigator-driven model that characterizes univer-sity research, administration of research at thenational labs has been centralized, top-down(from within the Department of Energy and fromCongress) and largely lacking in outside peerreview. The Galvin Task Force reported that,while the DOE labs support a small amount of com-petitively-determined individual investigatorresearch projects, “the research culture at manyof the laboratories has been influenced by theirrelative physical and intellectual isolation and bya sense of entitlement to research funds.”36

Despite all of these problems, the labs have con-siderable resources in the form of physical andhuman capital. Unlike the individual investi-gator work that typically characterizes univer-sity research, basic research performed at thenational labs has often been performed on amuch larger scale, with considerable physical cap-ital investments made over many years. Criticsand supporters of the labs alike now struggle withthe question of how to put these large-scaleresources to work, short of abandoning themaltogether. This question has been the impetusfor a number of blue-ribbon commissions inrecent years.37 Unfortunately none of these ini-tiatives has produced a politically-viable consensuson meaningful restructuring.

Nonprofit Institutions

Basic Research at Nonprofit Research Institutions

Although they account for only seven percentof basic research in the United States, nonprof-it institutions (that are not universities) performa significant amount of basic research. In somecases, they provide a unique alternative to the dom-inant model of university-based basic research.

Many nonprofit research institutions are affil-iates of universities, maintaining formal tieswith the university but remaining independentof its non-research functions. The Whitehead

Institute for Biomedical Research, an affiliate ofthe Massachusetts Institute of Technology, isone notable example.

Nonprofit institutions sometimes offer theindividual researcher attractive alternatives tothe predominant university basic research venue.Two noteworthy examples are The HowardHughes Medical Institute (HHMI) and the ScrippsResearch Institute. HHMI offers much sought-afterappointments for scientists, who are attracted bygenerous and stable financial support and aresearch environment that does not require timeoutside the lab devoted to fundraising. HHMI lab-oratories are located at research universities acrossthe country; these universities serve as collaboratorswith the Institute. HHMI maintains the laboratoryfacilities at these universities and employs thescientist and all necessary support staff (includ-ing research associates and technicians). Scientistsemployed by HHMI are appointed to five orseven-year terms, with the possibility of renew-al under rigorous standards of review. Theseterms represent a longer and more stable periodof support than many university scientists areable to achieve through research grants from thefederal government or other funding sources.

Scripps maintains its own research campus inLa Jolla, California, also attracting eminentresearchers through generous and stable finan-cial support, affording researchers maximumtime spent in the labs and away from fund-rais-ing duties. Like the nation’s research universi-ties, Scripps relies on federal grants for a majorityof its research funding, though it has aggres-sively pursued funding from industry and phil-anthropic sources. Unlike the academic researchmodel though, Scripps does not isolate its facultyand labs into separate disciplines. Rather, itemphasizes cooperation and collaboration acrossdisciplines.

Institutions like Scripps also present an inter-esting alternative to one aspect of the universi-ty-based training model for graduate students.Scripps offers graduate-level training in variousresearch disciplines, but does not educate under-graduates. As the “teaching versus research”debate continues on university campuses, this alter-native model—which couples basic research withgraduate training, but excludes undergraduate

27

teaching—may become more attractive for fac-ulty scientists and graduate students.

Basic Research Funding from the Nonprofit SectorCertainly, HHMI is tremendously important

to basic medical research, not simply due to itsunique administrative and performance structure,but also for the considerable financial resourcesit brings to the research enterprise. In fact, the non-profit philanthropic sector has always been animportant source of funding for research. HHMIis a relative newcomer: the Rockefeller Foundationand the Carnegie Institution have been impor-tant sources of such philanthropic supportthroughout this century.

As a funding source, foundations are verysmall relative to the federal government; yet,they can offer distinct advantages, which allowthem to leverage their funds to greater effect. Forexample, foundations can move more quickly thanthe government to fund projects. In this way, theycan act as a funding bridge, stepping in to main-tain funding for important projects whose grantsare about to expire. Foundations may also be moreprogressive than the government in valuing andfunding cross-disciplinary research.

THE CRITICAL VALUE OF THE INDIVIDUAL RESEARCHER

As discussed in Appendix 1 (see Appendix Box 1, “Research Dollars Versus Other Inputs”),basic research in the United States is not adequatelydescribed by dollars spent. Nor can it be char-acterized only by the institutions that house it.In fact, the core strength of the research enterpriselies in the people who do the research. Indeed,the history of science and engineering and of thegreat discoveries is a history of individual scientists,whose names—James Watson, Jonas Salk, LinusPauling—are often more readily recognized thantheir discoveries. The success of the American basicresearch system has rested in its ability to fostersuch creative minds and ensure a robust flow ofnew scientists through the education, training,and employment pipeline.

As we described earlier, a great strength of ourbasic research system is the symbiotic relation-

ship between research and graduate educationin the research university. The university scien-tist has in the institution’s graduate students apool of highly skilled and motivated labor. At thesame time, the experience and training that thegraduate student receives in a university labensures a stream of talented future scientists.The education function of the university—thatis, the training of future scientists—is as impor-tant to the future of basic research as the researchfunction itself.

Young scientists trained in the research uni-versity go on to careers in the university, gov-ernment, and industry sectors. As Figure 5, page28, indicates, a growing proportion of scientistsand engineers are employed by industry, althoughin absolute terms employment in all sectors isgrowing (see Appendix Figure 3 in Appendix 1).Industry-employed scientists are an importantlink in the transfer of basic research to industry.In-house basic research expertise is often necessaryeven when little basic research is conductedwithin the company. Industry scientists play animportant role in identifying and interpreting basicresearch performed outside of the company foruse internally.38

It remains, though, that much of the successof our basic research system lies in the ability ofacademic researchers, and particularly youngacademic researchers, to exploit the individualautonomy and freedom that their institutions pro-vide them. The university environment fostersa spirit of independence and creativity among sci-entists that is hard to find in other organiza-tions. This environment is particularly importantfor young scientists, who are given opportuni-ties for advancement early in their academiccareers that are rare in more hierarchical orga-nizations. In short, the university research envi-ronment is able to recognize and exploit talentregardless of seniority. Consequently, the flow ofnew scientists into the university workforce hasbeen an important dynamic in the basic researchenterprise.

Yet, there are signs along the length of the“pipeline” of new scientists and engineers tosuggest that the future quality of the humaninfrastructure for basic research is not assured.In order to assess our human capacity to do

28

basic research in the future, we must consider thecondition and quality of today’s educationalpipeline at all of its stages:

1. The poor performance of our K-12 students inthe sciences is well-documented.39 A recent survey reported that 43 percent of high schoolseniors have a below-grade level knowledgeof science.40 Further, evidence from interna-tional studies suggest that even our best stu-dents perform poorly relative to students inother countries in the sciences and math.41 Poorperformance in the sciences and mathematics(and all disciplines) during these early yearsnot only effectively precludes many Americanstudents from later scientific training andemployment, it also places an increased bur-den on universities and colleges to provideremedial education, at the expense of moresophisticated study in the sciences.

2. At the undergraduate level, the share of sci-ence and engineering degrees has declined overthe past three decades.42 Increasingly, manyof our “best and brightest” are opting out offuture careers in the sciences during theirundergraduate years.

3. An important source of high quality scientistsand engineers—foreign students—may beleveling off (see Figure 8, page 31) and is farfrom assured in the future. As other coun-tries develop their own basic research capa-bilities, the supply of foreign students who trainin the United States and remain here forcareers in basic research should not be expect-ed to match earlier decades, when there wasa paucity of basic research performed in othercountries (particularly in Asia).

4. The current system of graduate training hascontributed to problems in the employment

Figure 5

Where Doctoral Scientists and Engineers Are WorkingDistribution of Employment of PhD Scientists and Engineers by Employment Sector

NOTE: Percentages do not total 100 percent due to non-response and omission of the non-university education sector.SOURCE: Reshaping Graduate Education of Scientists and Engineers, Committee on Science, Engineering and Public Policy, (Washington,DC: National Academy Press, 1995).

10%6% 9% 7%

0%

10%

20%

30%

40%

50%

60%

1973 1991

45%

57%

24%

36%

University Industry Government Non-profit

29

of new PhD’s and may have a discouragingeffect on the number and quality of studentsentering graduate programs in the future:

• The time to degree for PhD recipients in thesciences has been rising, from 5.4 years in1962 to 7.1 years in 1993.43

• The number of new PhD’s entering post-doc-toral fellowships and other temporary or part-time employment is also on the rise,44

creating an increasingly unstable workenvironment during the critical early yearsof employment in basic research. In somefields, recent graduates have gone from afirst to a second, and in some instances a thirdpost-doctoral research grant while contin-uing their search for an academic position.

• Without adequate flexibility in trainingand in career counseling, new PhD’s emergefrom graduate programs into a very nar-row, academic job market. While trainingfor a career in academic research is criticalto the quality of science and engineeringPhD’s, additional training to provide stu-dents with additional options upon grad-uation is also important.

5. Finally, the employment environment in uni-versities for scientists and engineers is strainedby the amount of time devoted to fundrais-ing-related activities and away from theresearch lab.

All of these issues point to potential troublespots in coming years. It is worth reaffirming,however, that these problems arise within asystem of education and training for basicresearch that continues to perform very well, per-haps better than any other in the world.Nonetheless, as we discuss in the next chapter,all of the concerns we identify here should andcan be addressed today, long before small prob-lems become big ones.

THE INTERNATIONAL CONTEXTFOR AMERICAN BASIC RESEARCH

The 20th century has seen the United Statesrise to a position of global preeminence in gen-

erating new knowledge. World leadership ininnovation has been viewed as both a causeand a consequence of the nation’s economicsuccess on the world stage. But it appearsinevitable that as other nations develop their inno-vative capacities, the United States will becomeless dominant in various scientific disciplines andeconomic sectors.

Recent trends suggest that while the UnitedStates continues to be the world’s leader in thegeneration of new knowledge, other countriesare contributing more than in the past. The U.S.share of all scientific and technical literatureworldwide continues to be much larger than anyother country’s, but it has declined slightly in recentyears from 36 percent in 1981 to just under 34 per-cent in 1993.45 Most of that decline can be attrib-uted to Japan’s increasing share.

It would be a mistake to view these trends as a threatto the United States. As basic research activities growworldwide and the global stock of new knowl-edge increases, all countries benefit. Further,other countries are increasingly our collaboratorsin basic research. Cross-national collaboration inresearch, facilitated by rapid advances in infor-mation technology, is a growing phenomenon andan increasingly important mechanism for pool-ing resources for large basic research projects.Figures 6 through 8, pages 30 and 31 illustratethree measures of this trend.

• As Figure 6 indicates, foreign R&D expendi-tures in the United States have grown dra-matically since 1980; R&D performed abroadby U.S. firms has also increased, though at aslower rate.

• As a share of total national science articles, inter-nationally co-authored articles have increasedfor all of the leading industrial countries (seeFigure 7, page 30).

• Finally, the number of foreign graduate stu-dents in academic research programs in theUnited States has increased 75 percent since1980, although as a share of all graduate stu-dents, their numbers have increased onlyslightly over this period and their share is nowdeclining (see Figure 8, page 31).

30

Figure 6Increasing Cross-National R&D Investments

SOURCE: National Science Board, Science & Engineering Indicators, 1996. Washington, DC: U.S. Government Printing Office, 1996).

SOURCE: National Science Board, Science & Engineering Indicators—1996, (Washington, DC: U.S. Government Printing Office, 1996),Appendix Table 5-34.

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

0

2

4

6

8

10

12

14

Bil

lion

s of

Con

stan

t 198

7 D

olla

rs

U.S. company R&Dperformed abroad

Foreign R&D

United Statesexpenditures in the

– – – – – – – – – – – – – –

Figure 7Internationally Coauthored Articles, Selected Countries

0

5

10

15

20

25

30

U.S. U.K. Germany France Japan Canada Former USSR

Per

cen

t of

Cou

ntr

y's

Art

icle

s 1981-87 1988-93

31

Some observers have questioned the abilityof the United States to capture the return on itsresearch investments in an increasingly globaleconomy and scientific community. For exam-ple, the growing number of foreign graduate students who leave the United States upongraduating has provoked concern in some quar-ters. But it is not at all clear that the U.S.-trainedforeigner who returns to his or her country represents a net loss to the U.S. economy. Afterall, much of the value of the U.S. research university is derived from the work performedby graduate students during their enrollment inthe university. Further, U.S. industry benefits fromresearch undertaken abroad, often performed byAmerican-educated scientists and engineers.

CONCLUSIONBasic research in the United States is unques-

tionably one of the great success stories of thepast 50 years. Indeed, the basic enterprisedescribed in this chapter is strong and effec-tive. But to maintain, and even build on, this enter-prise for the next 50 years, there are problems toaddress. From how federal policymakers allo-cate funds for basic research, to how individualresearchers balance their research and teachingresponsibilities with the demands of fundrais-ing and entrepreneurship—these issues set theagenda for policymakers and business leadersconcerned about basic research and will guidethe discussion of policy recommendations inthe next chapter.

Figure 8Full-time Foreign and U.S. Graduate Students in Science and Engineering Fields inU.S. Universities

SOURCE: National Science Foundation/SRS, Survey of Graduate Students and Postdoctorates in Science and Engineering (data available atwww.nsf.gov), Table B-5.

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

U.S. Citizen Foreign

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

32

The remarkable scientific progress that hashelped to define 20th century America is due inno small part to the unique characteristics of ourbasic research enterprise. Chapters 2 and 3 havedescribed these elements of success in Americanbasic research— public funding, merit-based allo-cation of resources, the central role of the researchuniversity, the primacy of the individual investi-gator, and a robust education pipeline for futurescientists and engineers— and how they havecontributed to the social and economic benefits thatwe enjoy today. But in each of these elementsthere are currently signs of stress that need to beaddressed if the promise of tomorrow’s basicresearch is to meet the expectations created by thesuccesses of the past. In this chapter we willdescribe the challenges to our system of basicresearch, offering policy recommendations thataddress:

• problems in the way resources are allocatedfor basic research

• the threats to ensuring adequate resourcesfor future basic research

• the challenges of maintaining a pipeline of high-quality scientists and engineers into the basicresearch system

• the potential conflicts and problems arising fromthe increased interaction between universitiesand the marketplace

• the challenges that American basic research facesin an increasingly global economy and researchenterprise.

IMPROVING THE ALLOCATION OFRESOURCES FOR BASIC RESEARCH

In addition to ensuring an adequate level ofresources for basic research, policymakers havea responsibility to maximize the potential ofthose resources through efficient allocationmechanisms. There is waste and inefficiency inthe current allocation systems for basic research.Shortcomings in allocation arise on two levels:1) in the allocation of the research grant from thefunding agency to the researcher, whether it isdone competitively, based on peer review, andwhether it is directed toward individual inves-tigators rather than institutions; 2) in the allo-cation of funds to agencies and missions fromCongress.

Mechanisms to Ensure Quality in FundingIf basic research were like any other produc-

tion process, efficient allocation of resourceswould be a relatively straight-forward matter.Resources would go toward efforts that demon-strated the highest productivity, as calculatedthrough a measure of output. But as describedin Chapter 2, measuring research outputs and theproductivity of basic research in general, letalone for individual basic research projects, is high-ly problematic.

Over the years, government agencies undervarious mandates have attempted to develop out-put measures and related productivity and qual-ity indicators for federally-sponsored research.The most recent initiative has been mandated by

Chapter 4

MAINTAINING U.S. LEADERSHIP IN BASICRESEARCH: THE CED PRESCRIPTION

33

Congress under the auspices of the GovernmentPerformance and Results Act (GPRA). GPRArequires all government agencies to submit per-formance plans for measuring and assessingprogram impacts as a means of increasing pub-lic accountability, an exercise that has provedunderstandably problematic for science andresearch-related programs. Efforts to increaseaccountability in government are laudable andgovernment-funded basic research should notbe exempt. Nonetheless, GPRA should notimpose one-size-fits-all criteria for measuringresults. Such an approach would, at best, proveunworkable for basic research programs; atworst, a rigid imposition of quantified perfor-mance standards would undermine basic researchby shortening the timeframe of projects andlimiting their scope to areas where the payoff ispredictable at the outset.

Measuring basic research output is no less atroublesome issue for business, which compa-nies have addressed with mixed success. The IBMcase (see Case Studies) highlights that compa-ny’s efforts to quantify its own research out-put, measuring for example, number of patentsfiled, number of external awards received by IBMresearchers, and self-assessments of key resultsfor the year.

Absent meaningful and practical outputmeasures, we must rely on other means ofefficiently allocating resources. The best alter-native, we believe, is the system of peer-reviewed competition for research grants.Peer-reviewed awards to individual research-ers— and increasingly to teams of researchers—for individual research projects helps to ensurequality on a project-by-project basis. Some crit-ics have charged that peer review encourages an“old boy” network because it favors the statusquo in grant awards. Yet, we know of no bettermechanism for emphasizing substance overreputation or political interests.

The worst abuse in the allocation of basicresearch dollars, leading to the least productive useof research funds, is Congressional earmarking.The most recent survey of these activities demon-strates dramatic growth in the number of ear-marks for academic research in recent years.46

Figure 9, page 34 shows the increase in number

of earmarks during the 1981-1992 period; over thecourse of a decade, academic earmarks increasedin value from the tens to the hundreds of millionsof dollars. Such earmarks frequently place narrowconstituent interests or even otherwise laudablegoals like regional economic development, overscientific merit.47 In bypassing the competitive,peer-review process for determining merit, ear-marking for basic research is a form of pork-bar-rel, like any other; it should be recognized as suchand acknowledged to have no place in our pub-licly-supported basic research enterprise.

Of course, allocation of funds through peer-reviewed grants to individual investigators doesnot meet all the needs of our basic researchenterprise. For example, there has long been, andcontinues to be, a need for public investmentsin large-scale projects, often referred to as “bigscience.”48 Historically, projects like the fusionreactor at Princeton University, the historicVoyager satellite missions, and NSF’s deep seadrilling project have proved very important tobasic research and could not have proceeded inthe context of relatively small individual inves-tigator grants. In cases such as these, whenthe nature and size of the research projectmakes individual investigator grants imprac-tical, direct funding of institutions rather thanindividuals is appropriate.

Unfortunately, because of the large institutionaland regional economic stakes, funding for bigscience often falls prey to political interests.Also, it is often threatened by budget cuts or otheroutside attempts to influence the character of theproject. It is the responsibility of policymak-ers to ensure that necessary investments inbig science and institutional grants proceed onthe basis of scientific merit (determined byexpert advisory committees) and in the largercontext of national needs and priorities.

Further, it is important to emphasize that allo-cation for big science through institutional grantssimply means investing in the infrastructure nec-essary to make certain fields of inquiry possibleby giving researchers the tools to do their work.The scientific work that proceeds from invest-ments in these tools should conform to thesame process of competitive peer review that indi-vidual investigator projects do.

34

In sum, the primary mechanisms for allocatingfederal basic research funds in all agenciesand to all institutions—whether universities,federal labs, or elsewhere—should be based onscientific merit determined through peer reviewand the support of individuals and projects.Political earmarks for basic research are anunproductive use of scarce funds and shouldbe halted. The needs of big science can be metwithout sacrificing scientific quality, but thisdepends critically on the motivations of poli-cymakers and the input of scientists.

Ensuring Adequate Funds for All BasicResearch Disciplines

Peer review goes a long way toward ensuringquality in the allocation of funds from federal agen-cies to individual research projects. But at the startof the funding stream—the allocation of basicresearch dollars from Congress to federal agen-cies—there is no such mechanism to ensureoverall quality. Setting broad priorities in basicresearch is the domain of policymakers inCongress and the Administration, and shouldbe the result of informed policy debate. Scientists

have an important role to play in ensuring thatthe debate is an informed one; but ultimately, thesedecisions are appropriately made by our elect-ed leadership.

Certainly, as the growth in earmarking illus-trates, there is room for improvement in how pri-orities are established in public funding forbasic research. CED does not believe an overhaulof the congressional appropriations process inorder to serve the needs of basic research ispractical or remotely likely. Just the same,Congress can act in a number of ways to ensurequality basic research in its appropriations. Webelieve it critical that Congress work to achievea balance among basic research missions, adiversity of missions, and that increasinglyimportant cross-disciplinary research receivesbetter treatment by congressional appropria-tors and agency administrators alike.

Balance in Basic Research Missions

Currently, the health mission dwarfs all othercategories of federal support for basic research(see “Life Sciences,” Appendix Figures 7 and 10).49

Although we recognize that missions must beprioritized to ultimately reflect the needs ofsociety at large, we also believe that neglect ofless popular areas of research is foolhardy.Health research has benefited immeasurablyfrom advances in the computer sciences, thebehavioral and social sciences, mathematics,and physics—yet these disciplines receive mostof their federal support from non-health missions.The defense mission, in particular, has been animportant, and in some cases a primary, sourceof funding for diverse scientific disciplines (seeAppendix Figure 12). But, the decline in fund-ing for this mission alone from its Cold Warlevels has had negative consequences for someareas of basic research.

While the health sciences have clearly benefitedfrom applications of advances in other disci-plines, various non-health fields of science have,in turn, been furthered by biomedical research.In addition, the needs of biomedical science havecreated demand for further research in otherfields such as chemistry, physics, materials andengineering, and information technology.

Figure 9Growth in Number of Congressionally-Earmarked Academic Projects

SOURCE: National Science Board, Science & EngineeringIndicators—1993, (Washington, DC: U.S. Government PrintingOffice, 1993), p.139.

0

200

400

600

800

1000

1200

NU

MBE

R O

F EA

RM

AR

KS

1981-1983 1984-1986 1987-1989 1990-1992

35

These observations underlie the concept thatbasic research and subsequent development is amutually reinforcing process occurring betweena variety of science and engineering fields. Allof these fields will require adequate support inan increasingly multi-disciplinary environment.

The agency missions themselves are depen-dent on many scientific disciplines. Ultimately,progress in any single field will not be sustainedif other fields whither on the vine. In order toachieve an appropriate balance in funding forscientific disciplines, appropriators need topay close attention to the impact of agencyand mission funding decisions on specific sci-entific disciplines. As some missions becomelower priorities in the federal budget (the post-Cold War defense mission), or as Congress pur-sues structural reforms in major programs tiedto basic research,50 the impact (direct or indirect)on scientific disciplines should be assessed andthose disciplines should be accommodated in away that is consistent with mission/agencygoals and the needs of basic research generally.Often, this may mean making basic researchfunding a higher priority in other missions andagencies (such as the NSF), as it becomes less ofa priority in declining missions (as in defense).

A Diversity of Missions

The National Science Foundation has longbeen a model of support for peer-reviewed basicresearch independent of agency missions at thefederal level, providing funds on the basis of meritto the entire spectrum of science and engineer-ing disciplines. Nevertheless, much of the scientificprogress of the past half-century resulted frommuch larger allocations to basic research fund-ed by the Department of Defense and the NationalInstitutes of Health. Consequently, any effort toabandon agency and mission-based support forbasic research in favor of the “secular” efforts ofthe NSF would ultimately reduce the amount ofpublic support for basic research in general andundermine the value of a diverse approach to fund-ing. Federal support for basic research shouldbe diverse in its funding sources, resistingefforts for central control or concentration in onemission area. The diverse model is most viable

politically and is best-suited for the unpre-dictable nature of basic research outcomes.Therefore, we do not support calls for a“Department of Science” or for an NSF thatwould envelop all other federal sources ofbasic research support.

Cross-Disciplinary Research within Missions

All sources of federal basic research fund-ing should recognize the cross-disciplinaryimperative of much of today’s basic research andencourage such approaches in their funding deci-sions. The case studies (and particularly thePfizer, Merck, and Harvard cases) provide com-pelling evidence that a cross-disciplinary approachto scientific investigation is a necessity in manyareas of research today and is an approachemployed successfully by businesses and uni-versities alike.

It is not enough to ensure balance among mis-sions; it is increasingly important to recog-nize research that must take place acrossfunding missions and across scientific disci-plines. Grant proposals that fall outside of thetraditional categories or scientific disciplinesshould not be punished by a traditionally dis-cipline-oriented funding structure and peerreview bias. This is an imperative that appliesto universities and peer review panels as muchas it does to agency administrators and con-gressional appropriators. Evaluating merit in thesecases may require more effort (requiring, forexample, a peer review panel that consists of sci-entists from more than one discipline). But theeffort should be made. Universities can pro-mote collaboration by recognizing its value intenure decisions and through university-level ini-tiatives that are independent of federal agencydirectives. Harvard’s Mind/Brain/Behaviorprogram (see Case Studies) illustrates such aninitiative.

Keeping Research Universities at the Core ofAmerican Basic Research

The Central Role of the Research University

CED believes the most productive use offederal basic research funds institutionally is

36

through the nation’s research universities. Theuniversities’ exceptional track record in per-forming high-quality basic research is not sur-prising given the central role of the individualinvestigator and the widespread use of com-petition in grant awards.

The research university deserves to be the pre-dominant basic research institution in Americafor yet another reason: federal support of uni-versity research ensures the future health of science and engineering by supporting the training of graduate students (see “Sustainingthe Education and Employment Pipeline,” page 40). Tomorrow’s scientists and engineersreceive the best possible training in today’s uni-versity research laboratories through direct par-ticipation in the leading research of the day.There simply is no substitute for this type of train-ing. A decline in support for research universi-ties would terribly weaken the foundation forthe basic research enterprise of the future.

Problems and Solutions in the Administration ofFederal Grants for University Research

Although federal funding of basic research inuniversities has been very successful, there areseveral aspects of the grants process that areproblematic—for the individual researcher andthe university—and should be reformed.

There are concerns among many universityresearchers that the competitive grant structurehas become overly-burdensome for the indi-vidual researcher and has deterred young scien-tists from pursuing careers in academic research.The administrative burden for the individualresearcher lies in: 1) short-duration grants, whichrequire repeated preparation and re-submis-sion of grant applications to sustain researchprojects; 2) the low rate of success in applications;3) the long time period between the submis-sion of applications and the notification ofawards; 4) the very high dependency of theindividual investigator on external support andthe lack of seed money. To alleviate these prob-lems we urge grant-making agencies (and uni-versities) to review their requirements andsystems of support with an eye toward reduc-ing administrative burden.

Grant-making agencies should explore waysto offer outstanding scientists longer-termgrant support, to provide young researcherswith sufficient resources to launch their careers,and to ensure that established researchers whotemporarily lose grant support are not forcedto abandon long-term, productive researchendeavors. Institutions like the Howard HughesMedical Institute and the Scripps ResearchInstitute (see Chapter 3) provide cues for grant-making institutions and universities in this effort.In particular, we urge agencies to consider extend-ing the funding period of their basic researchgrants. The continual process of applying and re-applying for grants has a very discouragingeffect on young scientists and engineers con-sidering careers in university research. Longerand more stable periods of grant support wouldalleviate this problem

Many funding agencies have already begunthe process of limiting administrative burden ingrant awards and have made some progress.For example, some agencies now permit electronicsubmissions of grants. However, a large andpersistent cause of administrative burden resultsfrom a cumbersome system of overhead, or indi-rect cost, reimbursement in the federal grantsprocess. Federal research grant awards to universityresearchers include funding to cover the directcosts of the research and overhead research costsincurred by the university, reimbursements forwhich vary considerably from grant to grant. Fromthe university’s perspective, there are large costsassociated with the investments in plant, equip-ment, and human resources necessary to pursueindividual research grants and sustain a researchenterprise at the university. The universities’tuition income and other funds received foreducation are designated strictly for that func-tion. Without overhead reimbursement, uni-versities would not have the necessary level ofresources or incentives to sustain a vibrant basicresearch enterprise.

But the current retrospective, cost-based pro-cedures for establishing federal reimbursementfor overhead have resulted in a fractious rela-tionship between universities and government.Present procedures are very costly and timeconsuming and, according to some experts, dis-

37

tort incentives for efficient expenditures.51 Toremove these distortions, and much of the unnec-essary regulatory burden, they recommend asystem in which overhead rates associated withall research grants are set by benchmarks, oraverage overhead rates for similar universities.The benchmarks would be determined on aperiodic basis by examining costs at a smallsample of universities within each class of insti-tution. Such a procedure would greatly reduceaccounting and auditing costs for universities.It would also minimize government leverage,which universities consider excessive because ofthe government’s power to demand lower reim-bursement levels. Indeed, universities haveargued forcefully that cost-shifting has escalat-ed in recent years, with federal agencies increas-ingly unwilling to adequately cover overhead costs.Finally, benchmarking would provide universi-ties with a strong incentive to trim overhead,because they would benefit from holding costsbelow the benchmark.

CED believes that the benchmarking ofoverhead reimbursement rates has merit inprinciple and encourages funding agenciesand universities to explore it on an experi-mental basis. The determination of “similaruniversities” is a difficult task, which shouldtake into account geographic variations amongother factors. In general, reform of indirectcost reimbursement should be guided by theprinciples of fairness (to both parties) and sim-plicity.

Too Many Research Universities?

Some observers are concerned about the num-ber of research universities vying for federalfunds today (see Chapter 3). As this numbergrows, basic research resources are spread morethinly and may be put to less effective uses. Werecognize this possibility, though we do notsupport attempts to set a fixed number ofresearch universities eligible for funding, orworse yet, a fixed list of specific universities eli-gible for funding. Historical experience indicatesthat it is very difficult to predict who will makediscoveries and where discoveries will takeplace. Further, such “fixes,” in our minds, vio-

late the principles of flexibility and competitionthat have defined the success of America’s uni-versity research system.

CED believes the appropriate number ofresearch universities will best be determinedby reinforcing existing mechanisms of peerreview, as well as by distributing funds at theproject and investigator levels rather than at aninstitutional level. Competition for grants basedon scientific merit will ultimately separate thewheat from the chaff among universities. Thosethat do not attract quality researchers will not beable to support a research program through fed-eral grants. And the openness of the competitionto researchers regardless of institutional homeensures accessibility to high talent wherever itis based.

Unfortunately, university research that isfunded outside of the peer review system is notsubject to this important competitive process.Again, political earmarks are not an appropriatefunding mechanism to support universities or anyother research institution. University research sup-ported by earmarks would not likely meet the peerreview standard, and a proliferation of researchuniversities due to earmarks would underminethe quality of our basic research enterprise.

Finally, although we do not support a “top-down” approach to limiting the number ofresearch universities, we also do not believethat all universities and colleges in the UnitedStates should feel an entitlement or an oblig-ation to pursue federal research dollars.Competition for research dollars is healthy forthe basic research enterprise; it is not always inthe best interest of individual universities and col-leges to enter this competition. In particular,schools should not neglect their education mis-sion in the process. For many institutions, thepursuit of a research function comes at the costof undergraduate education, which is critical notonly to the basic research enterprise but to oureconomy and society as a whole.

Existing Alternatives to Research Universities Are Weak

Although CED supports basic research fund-ing through a diversity of missions and agencies,

38

we question the degree of success that someagencies have had in supporting basic research.We believe that the further removed agencies arefrom a merit-based, competitive grant system ofallocation, the poorer the science will be whichthey support. The quality of research supportedby the Department of Agriculture has sufferedfor this reason.52 In fact, just 5.4 percent of theDepartment of Agriculture research budget goesto nationally competitive research grants. If pay-offs to our public investments in agriculturalresearch are going to improve in the future, thispercentage must increase substantially.

The Department of Energy’s national labsface an uncertain future due, in part, to inade-quate mechanisms for determining merit, but alsodue to shifting missions. It would be a mistaketo view the end of the Cold War as the end of themission justification for the national labs. Thereremain very strong defense and energy researchobjectives that are within the purview of manyof these labs. However, at the prodding of theAdministration and Congress, many labs areincreasingly chasing the technology du jour, withover half of their research dollars now devotedto commercial product development, oftenthrough industry partnerships.53 Although indus-try partnerships with the national labs mayserve the competitive interests of specific indus-tries in the development and commercializa-tion of technologies, they do not always serve theinterests of the nation and its taxpayers. As weargue later in this chapter (see “Choosing BasicResearch Among Competing Claims,” page 39),subsidizing civilian technology for nationalcompetitiveness purposes is not a justifiable fed-eral mission and it should not be allowed to dis-place federal investments in basic research.Further, the Intel case described in Chapter 3 illus-trates the political morass that such partner-ships can create when mission justification isweak, raising questions of favoritism amongcompanies, as well as concerns about indirect fed-eral subsidies of foreign companies.

At the same time, the national labs continueto represent a tremendous potential resourcefor the nation’s basic research enterprise, par-ticularly oriented toward large-scale scientificinquiry. We call on the Congress and the

Administration to make a clear determinationof the missions of the labs and assess whererealignments of missions and functions arenecessary. The recommendations of the GalvinCommission (see page 26) should serve as astarting point for this work. The labs themselvesshould be free to take actions to ensure the bestresearchers are attracted and retained to accom-plish lab missions. In particular, more of theresearch conducted in the national labs, includ-ing the large-scale science that they have thecapacity to perform, should be brought into a sys-tem of competitive, merit-based peer reviewthat characterizes the best of our basic researchenterprise.

Beyond scientific merit and mission justi-fication, cost-efficiency in these labs must beimproved; performing basic research in thenational labs should not cost more than simi-lar projects in other basic research institutions.To this end, research projects performed in thenational labs should be free of the multiple lay-ers of micromanagement emanating from theDepartment of Energy and Congress that havecreated gross inefficiencies and rigidity in thelabs.

In sum, it is clear to us that if the national labsare to continue to play a productive role inbasic research, that role must be justified on thebasis of strong missions, outside peer-revieweddeterminations of scientific merit, and effi-cient management and oversight structures.

PRESERVING OUR CAPACITY TO DOBASIC RESEARCH IN THE FUTURE

Future capacity to do basic research requireslong-term thinking today. Unfortunately, twothreats to tomorrow’s basic research enterprisecurrently command too little attention frompolitical leaders and policymakers.

One is funding for basic research. Today’sfederal budgetary environment, with rosy scenariosfor “budget surpluses as far as the eye can see,”has created a decidedly optimistic attitude with-in the science and engineering community (andamong its advocates in Washington). Speaking ofa threat to basic research funding in this envi-

39

ronment may strike many as odd if not down-right foolish. Yet, the likelihood of stable and ade-quate support, not just for the next few years butfor the next few decades, is far from ensured.

The second area that requires long-term think-ing today is in sustaining the human capacity todo basic research. Sustaining the education andemployment pipeline for basic research alsomeans raising the technical and scientific abil-ity of all students and society at large. Thiseffort is important both so our citizenry canbetter exploit the increasingly flow of newknowledge in an increasingly sophisticatedworkplace and also so the importance of basicresearch is not lost by a society that is less con-nected to progress of science.

Ensuring Adequate Resources for Basic Research

Throughout this chapter we recommendways that basic research resources can be usedmore productively. But none of these policy rec-ommendations should be taken as support forgetting along with less public support. As wedescribed in Chapter 2, the economic and socialpayoff to American investments in basic researchhas been tremendous in this century alone.Given this level of benefit, adequate and sus-tained funding for basic research must be a highand consistent national priority.

With this in mind, we are deeply concerned bytrends that have the potential to squeeze thelevel of resources for basic research in the future.Both the Administration and Congress now claima desire to increase basic research funding sub-stantially over the next few years. It remains tobe seen, however, how these increases will mate-rialize and if they will be sustained. Windfallsderived from a booming economy or a hypo-thetical tobacco settlement would certainly bewelcome additions to basic research funding.They do not, however, eliminate political andbudgetary imbalances that stand as threats tofuture funding.

Recognizing the Role of Private SectorSupport for Basic Research

Advocates of a smaller government role in basicresearch point to increasing private sector R&D

budgets as evidence that the federal govern-ment’s efforts are no longer as important in thisarea. Industry is doing more, the argument goes,therefore government can do less. However, pri-vate spending on total R&D should not be con-fused with spending on basic research, norshould industrial basic research be confusedwith government-supported basic research. Asindicated in Chapter 3 and as the Pfizer casestudy suggests, although the “Web of R&DInnovation” is a complex one, with complexinteractions between basic and applied research,it remains clear that the private sector role in basicresearch, and in R&D in general, is and willcontinue to be largely distinct from the govern-ment’s role (also see Box 2, page 12).

In sum, industry will continue to support andperform important areas of basic research; butthis work should not be viewed as a substitutefor the much larger role of the federal govern-ment in support of basic research. Given thesetrends and the misunderstanding of them thatis prevalent among some political leaders andthe public generally, we urge business to begina dialogue with political leaders and theAmerican public so that they might betterunderstand the critical importance of stead-fast government support of basic research.

Choosing Basic Research Among Competing ClaimsMeeting the Entitlement Threat

Despite robust political support for short-term funding increases, support for basic researchis likely to become less sure in years to come, dueto the budgetary costs of an aging population.Basic research is one of many competing claimson a shrinking discretionary portion of the fed-eral budget, necessitated by rapid expansion ofentitlement programs, particularly Social Security,Medicare, and Medicaid. The considerable rev-enue boost the federal budget has received fromthe current expansionary economic cycle does noterase the underlying structural deficiencies in thebudget. This is a problem that will become dra-matically worse the longer we wait to deal withit; indeed, we may ultimately face devastatingcuts in the basic research budget if entitlementspending is not brought under control before the

40

baby-boom generation retires. As CED hasargued frequently in recent years, our politicalleadership simply must deal with the growingburden of the federal entitlement programs.54

Otherwise, the federal budget will have no roomleft for government’s most important activities,including investment in basic research.

Federal Funding for Civilian Technology

One of the competing claims that basic researchfaces in the federal budget comes in the form ofspending on applied and development research,much of which is necessary to achieve objec-tives in various missions, such as the develop-ment of weapons systems technology for nationaldefense. A significant initiative of recent years,however, has been to seek to improve the inter-national competitiveness of U.S. industry byincreasing federal expenditures in certain areasof applied research and civilian technologydevelopment. The Advanced Technology Programin the Department of Commerce is a product ofthis initiative.

CED does not believe that national com-petitiveness programs have the same strongclaim to federal support that basic researchdoes. In Chapter 2, we described the very com-pelling case for government support of basicresearch. Although proponents of competitive-ness programs often use the same language tomake their case—that public subsidies are nec-essary to correct for insufficient levels of fund-ing by the private sector in key areas (i.e., “marketfailures”)—we find the case far from convincing.Indeed, too often government spending on pro-grams of this type amounts to little more than tax-payer subsidies for favored industries and firms,supporting research that the private sector wouldhave supported on its own or research that is notworthy of public or private funding.55

With few exceptions, we do not think gov-ernment should be in the business of directly fund-ing what we view to be a function of the privatesector—development and commercialization oftechnologies. The exceptions apply in caseswhere the funding can be viewed strictly as a pro-curement function, as in the national defenseexample, or to correct a clearly defined and wellsubstantiated market failure.

Finally, CED believes that the governmentshould be sensitive to the basic research activ-ities occurring in industry and should avoidduplicative initiatives. Federal funding is mostvulnerable to duplication when research initia-tives are overly-prescribed—as in research thatis targeted at specific diseases—and removed from a general peer-review process.

Sustaining the Education and Employment Pipeline

In addition to the funding of basic research,a second threat to our long-term basic researchcapacity lies in our ability to sustain a pipelineof future scientists and engineers, a work forceadequately skilled in the sciences to exploit newknowledge, and a citizenry with enough under-standing of basic science to recognize its impor-tance and support its progress.

The United States is not in danger of run-ning out of scientists and engineers to performbasic research any time soon. After all, theycomprise only a tiny fraction of the nation’swork force: in an American labor force of 132 mil-lion, just 542,000 are doctoral scientists and engi-neers.56 But there is a growing disconnect betweenthe need for a highly-skilled basic research laborforce on the one hand, and the quality of our K-12 math and science education and the inter-est in science at the undergraduate level on theother hand. Further, the desirability of basicresearch employment at universities is eroded bythe amount of time devoted to applying for andcomplying with federal grants, as well as thedemands of university technology transfer offices.Left unchecked, we are concerned that thesetrends will gradually erode the base of Americanstudents willing or able to enter basic researchcareers.

These trends also exacerbate the on-goingchallenge of attracting women and minorities totraining and careers in basic research. The severeunderrepresentation of these groups in the sci-ence and engineering disciplines has far-reach-ing effects on our basic research enterprise; in part,in contributes to concerns that the peer reviewprocess is biased against some researchers andresearch projects. A more diverse research enter-prise has been identified as a priority by uni-

41

versities, agencies, and science organizations;we support this goal and urge these institutionsto hasten its realization.

Finally, we are also concerned about a societythat is increasingly isolated from the world of sci-ence and discovery. To a large extent, this isola-tion is driven by science itself. Long gone are thedays when a well-educated citizen could beexpected to have even a cursory grounding in allareas of knowledge (scientific or otherwise). Butthe isolation we are particularly concerned aboutinvolves a lack of understanding and accep-tance of scientific methods and principles. Thisdeficiency is a detriment to the skill needs of today’s workplace. It also threatens to underminethe public support for basic research, an enter-prise that relies primarily on the public sector forsupport. If society becomes less enthusiasticabout science and more suspicious, or simply indifferent, the case for research support willbecome much more difficult to make in the hallsof Congress.

Based on all of these concerns, we offer the fol-lowing recommendations to strengthen the sci-ence education and employment pipeline andscience education in general, not only for futurescientists but also for a public that must ultimatelysupport them.

Better Math and Science Education at the K-12 Level

In a series of policy reports on educationalimprovement, CED’s trustees have done exten-sive research into strategies for raising academ-ic achievement in general and science and mathperformance in particular.57 The complexity ofthis task requires comprehensive and coordi-nated change in several interdependent areas:• establishing national mathematics and sci-

ence performance standards and assessingprogress based on these standards;

• increasing teacher knowledge and skill throughbetter training and more incentives;

• upgrading classroom curriculum and teach-ing methods, including but not limited toexpanded use of technology in the classroomand adequate investments in infrastructure,such as lab space.

We maintain our strong support for highachievement standards at the national level inall core academic subjects, with particularemphasis on mathematics and science. However,given the inherent difficulties in getting nation-al standards in these subjects developed andaccepted, we urge teachers and administratorsto actively and continuously pursue informationand research on new knowledge and effective,innovative classroom practices and to redirect themission and goals of their schools toward rais-ing standards for student performance.

Improved learning cannot happen withoutimproved instruction. Therefore, our schools needto both attract and continuously support bet-ter-qualified math and science teachers, par-ticularly at the middle and high school levels.Strategies to accomplish these ends include:

• improving the way teachers are educated in col-lege, which includes requiring prospectiveteachers to learn how to integrate technologyuse and project-based learning into the cur-riculum and requiring a math and/or sciencemajor for teachers who plan to teach at the mid-dle or high school level;

• raising certification standards to require mid-dle and secondary science and math teachersto have majored in their subjects and ele-mentary school teachers to have taken substantive coursework in these subjects;

• employing stronger incentives, such as dif-ferential pay, to attract more qualified scienceand math teachers;

• developing alternative certification practicesto allow experienced scientists and mathe-maticians from industry to enter teachingwithout having to spend unnecessary time informal pre-service training, as long as they candemonstrate their teaching skills;

• encouraging businesses and other organiza-tions that employ scientists and engineers tooffer summer internships to teachers, givingthem direct exposure to math, science, and tech-nology in the workplace. These organizationsshould also explore opportunities to maketheir scientists and engineers available for

The Carnegie Institution of Washington andthe New York Academy of Sciences are bothactive in addressing the shortcomings of K-12math and science education in the United States.Their strategies reflect the key areas of concern:that math and science instruction be adequatelysubstantive and that the instruction be engagingenough to spark students’ interest in the subjectmatter.

Under the leadership of Dr. Maxine Singer,the Carnegie Institution launched the five-yearCarnegie Academy for Science Education(CASE). Now in its fourth year, CASE providesteachers in the District of Columbia (350 to date)with intensive training in fundamental scientificand mathematical concepts. CASE attempts tore-train teachers whose standard training typi-cally does not include sufficient grounding inscience and math.

CASE recognizes that increasing the substan-tive knowledge of teachers is half the battle. Howteachers impart this knowledge to students isalso important. The Summer Institute of CASEemphasizes a pedagogy based on experimenta-tion and questions, rather than rote answers andmemorization. Teachers participating in theSummer Institute are engaged in the same exper-iments and exercises that they will later pass onto their students—for example, constructing theperfect recipe for chalk, testing the city’s drink-ing water, or building wind-powered land racers.

The New York Academy of Sciences coordi-nates a summer internship program in the sci-ences for New York City school students. Theprogram places students in research labs (uni-versity, government, and industry) alongsideprofessional scientists. The defining element ofthe internship is the mentor-student relation-ship, which gives the students access to theworld of scientific research through the eyes of ascientist. Students also participate in lectures,workshops, and discussion groups, all intendedto help them assimilate the knowledge derivedfrom the lab experience. The internship culmi-nates in a research paper containing all the ele-ments of a professional document.

Both of these initiatives make welcome con-tributions to science education. Both alsodemonstrate the importance of an organizational“broker” in developing ties between the researchcommunity and the local school system. One-on-one efforts by individual scientists to assistteachers or mentor students are laudable, butcan only hope to have a large-scale impact in thecontext of coordinated efforts by intermediaryorganizations. We believe the best candidates forthese efforts are organizations—like Carnegieand the New York Academy—with public mis-sions and considerable flexibility and autonomy.Of course, not every city has a science-focusedfoundation or professional society, which is whywe encourage universities to step into this role.

42

in-school and after-school sessions with teach-ers and students to help them keep up withchanges in knowledge and practice in tech-nology and science-related fields (see Box 4 fortwo examples in practice).

Finally, it is critical that students be activelyengaged in learning in the sciences and math. Webelieve that improved teaching methods, togeth-er with currently available interactive learningtechnologies, including computers and Internetconnections, can effectively motivate studentperformance and accelerate improvements in

learning. Substantial investment in infra-structure improvements are also critical in thisregard—many schools have little or no labfacilities to support science instruction.

In order to further engage students in learn-ing and to excite students about a possible careerin the sciences, they should have the opportunityto interact with members of the research com-munity. Businesses, universities, and schoolsshould work together to place more profes-sional scientists and engineers in the class-room as volunteers to assist in class lectures, labwork, and field trips. These volunteers raise

BOX 4

BUILDING TIES BETWEEN PROFESSIONAL SCIENCE AND THE SCHOOLS

43

the quality of instruction, demonstrate to the stu-dents the future job benefits of a science and matheducation, and communicate the role of scienceand math in the world today. Similar initiativesgive students and teachers the opportunity to ven-ture out of the classroom and explore the natureof basic research in a professional setting (see Box4, page 42).58 High-quality and engaging videosand films can support this objective, while alsoovercoming geographic, time, and resource con-straints that might not allow for in-person inter-action.

Improving Graduate Training

Graduate-level scientific training is perhapsthe most important segment of the education-al pipeline for basic research. A science educa-tion that is 16 years in the making can continueon to a career in basic research or can be divert-ed to some other field, depending on whetheror not the college graduate decides to completegraduate training in the sciences. For thosechoosing a future in basic research, their impactas scientists and engineers depends a great dealon the quality of their graduate training.

Good scientific training is grounded in hands-on exposure to basic research and scientificinquiry. It is essential that graduate studentshave access to hands-on research projects. Thus,training of graduate students should be a para-mount criterion for universities and facultiesin considering their research portfolios, par-ticularly concerning the nature of the research.

The federal government can help to makegraduate student training a higher priority inthe research university by increasing theamount of scholarships and training grantsavailable to students. Grant support that goesdirectly to the student, versus support thatcomes indirectly from grants for research pro-jects, clearly places the priority on the student’straining rather than the needs of any particularresearch project.

Prolonged time to degree in graduate train-ing is another concern, though admittedly onethat is not well understood. To the extent that moretime spent attaining the graduate degree reflectsa greater complexity of the field of study, there

is little to be done and the increased time like-ly produces a more knowledgeable and pro-ductive graduate. But it is undesirable forprolonged time to degree to be caused by high-er student-to-faculty ratios, excess time spent onassisting faculty research, or difficulty in secur-ing stable funding.59 The direct burden andopportunity costs of a longer time to degreeare born by today’s student, and will also like-ly have a deterrent effect on those consideringgraduate training in the future. We urge researchuniversities to undertake frank and self-criti-cal examinations of their graduate trainingprograms in an effort to expedite the time todegree and reduce the financial and time bur-den on their graduate students. Again, greaterinvestments in graduate education by scienceand engineering-related federal agencies canalso play an important role in reducing this burden.

Academic employment is at the core of basicresearch and will remain so. But an increasingnumber of PhD scientists and engineers arefinding roles, whether by choice or due to alack of academic alternatives, in the private andother non-academic sectors. Private sectoremployment of PhD scientists and engineersplays an important role in the dissemination ofscientific knowledge. In order to facilitate thetransition from academic training to private sec-tor employment, graduate schools should offermore training programs and mentorships in theircurricula that prepare and orient students foremployment outside of academe.

Team-oriented and cross-disciplinary work—an important component of these efforts—neednot detract from core disciplinary training. In fact,the Pfizer, Harvard University, and BBN case studies (see Case Studies) all suggest that multi-disciplinary, cross-functional project teamsare an increasingly important element of successin many areas of basic research, regardless ofwhether it is private sector or university research.At the same time, some university researchadministrators have questioned whether thereis adequate grant support for cross-disciplinaryand team-oriented research, given the disci-pline-oriented nature of current grant systems.CED urges grant-making agencies to support

44

more team and cross-disciplinary efforts inresearch projects.

CED notes that several recent reports addressthe graduate training issues discussed here, andthat reforms in graduate education in the sciencesand engineering are underway.60 Agencies suchas the National Science Foundation are respond-ing to the recommendations made by the NationalAcademies and others. Universities also are par-ticipating in the renewal of graduate educationin the sciences and engineering.

Academic Employment of Young Researchers

Although there is an important role for privatesector employment in sustaining the basic research“pipeline,” academic employment of scientistsand engineers remains central to the health of thebasic research enterprise. The trend away fromfull-time faculty positions at research universi-ties and the increased reliance on post-doctoral,part-time, and temporary employment are notpositive signs. They send the wrong signals totoday’s college and graduate students as theyembark on the long and rigorous training nec-essary for a career in basic research.61 Hearinghorror stories of post-doctoral training thatextends well into a young researcher’s career, some prospective scientists and engineers are likely to opt out of this career path. Gettingyoung researchers out of temporary positionsand into stable employment should be a priorityfor all research universities.

As we discuss earlier in this chapter (see“Problems in the Administration of FederalGrants for University Research,” page 36), aca-demic employment also carries with it a largeadministrative and grant-raising burden. Researchuniversities and the federal agencies that spon-sor universities should be exploring ways tomake the academic scientist’s work environmentmore stable and more productive, with lesstime spent on raising money.

PRINCIPLES FOR UNIVERSITYRESEARCH IN THE MARKETPLACE

American research universities have alwaysbeen oriented toward practical concerns, while

maintaining excellence in the most theoretical and exploratory corners of science. The transferof knowledge and new technologies from the uni-versity to industry—as embodied in graduatesentering the work force, through open dissem-ination of research results, or through private part-nerships, patents, and licensing—is of greateconomic benefit to the United States. But uni-versities must walk a fine line as they seek toincrease the economic value of their basic research,while maintaining a public mission defined byopenness and wide dissemination of research findings. This balancing act is made more diffi-cult by the problems researchers face in obtain-ing public grant support, as indicated earlier inthis chapter. CED believes universities, industry,and the federal government should adhere to thefollowing principles as they pursue relation-ships in basic research.

1. The primary channel of benefit from uni-versity research to industrial innovation, andto society at large, is through wide andopen dissemination of knowledge in researchjournals, at conferences, and by the educa-tion of graduate students. New knowl-edge generated from university researchshould continue to be openly dissemi-nated. The publication of research findings,upon which future research frequentlydepends, should be only minimally delayedby patent preparation and other require-ments of sponsored research agreements.

2. Pursuit of patent protection on universityinventions, initiation of university-industrypartnerships, and licensing of intellectualproperty for commercial development canstimulate important new research and facil-itate and expedite the transfer of new tech-nologies to industry, ultimately benefitingsociety at large. Therefore, such partnershipsand licensing activities, when structuredappropriately (see items 3-5), should beencouraged.

3. These relationships can create conflicts ofinterest among the universities, their faculty,and collaborating industrial concerns.Universities should be strongly encouraged

45

to develop, and continually improve, poli-cies and procedures for technology trans-fer and industry relations, such that the basiceducational and research mission of the uni-versity is neither diluted nor compro-mised (see Appendix 2 for two examples ofuniversity patent policies).

4. In addition to the need for universities to makeresearch results broadly available, theseinstitutions also have an obligation to deviselicensing agreements for technology devel-oped using public funds in a fashion that gen-erates the greatest benefit for society. Theguiding principle should be to prevent theuniversity technology transfer system fromimpeding further research advances, wher-ever they may occur. As a general principle,research advances that can be turned intodistinct proprietary products are appro-priate candidates for exclusive licensing anddevelopment. On the other hand, tech-nology that can serve as a tool for manyresearchers to create new knowledge shouldbe made widely and nonexclusively avail-able under commercially reasonable terms.Industry should recognize and respectthese distinctions.

5. The success of America’s basic researchenterprise will continue to depend on theuse of patented inventions in basic research.In general, basic research will benefit if apatent holder’s rights (whether the owneris a company or university) are not enforcedin a way that restricts further basic research.Access of all parties to tools for basic researchis particularly important in cases wherefederal funding supported the initial dis-coveries. When a research tool is also aresearch product (as may be the case in areasof biomedical research), the interestedparties should work out terms that, first andforemost, do not prevent broad future useof the research tool, and secondly, permituse of the product in the marketplace.Industry and universities should contin-ue to seek avenues which allow the fewestpossible restrictions on use of patentedinventions for new basic research.

6. The major research universities have devel-oped policies that honor the public respon-sibilities of their research mission. Thesepolicies are particularly important in main-taining the central purpose, value andmorale of the faculty who are the lifebloodof the university. In particular, it is impor-tant that the interests of industry not influ-ence critical decisions of the universityregarding hiring, compensation and pro-motion. The continued strength of basicresearch in America depends on the abil-ity of faculty members to devote theirenergies to, and be evaluated by, theirquality and contributions as basicresearchers and their success in educatingnew scientists. Commercial successes aresecondary to these missions.

7. Finally, private sector funding of univer-sity research or research supported throughlicensing fees should not be viewed by uni-versities or the federal government assubstitutes for federal funding. The objec-tives and goals of private firms are differ-ent from public missions and should not beallowed to define the character of univer-sity research in general. Ultimately, theonly way to maintain basic research in theuniversity as it has been characterized overthe past half-century is through predomi-nately public funding.

INTERNATIONAL CHALLENGES ANDOPPORTUNITIES FOR AMERICAN BASIC RESEARCH

Maintaining a Commitment to BasicResearch in a Global Economy

Basic research activities around the worldare increasing as other countries step up thepace of their research investments. At the sametime, the new knowledge derived from basicresearch moves more easily across national bor-ders than ever before, thanks to the dramaticadvances in information technology. Transferof information is also facilitated by the steady paceof economic globalization.

46

These trends have led some to question the via-bility of our national basic research enterprise inan increasingly international economy. In fact, someargue that because the products of our basicresearch now move freely and quickly to othercountries, or to foreign subsidiaries of U.S. firms,U.S. taxpayers should not pay for it. They arguethat the United States should take better advan-tage of other nations’ basic research, or shouldattempt to limit foreign access to new Americanknowledge. CED strongly believes that argumentsfor reducing or restricting access to the out-comes of U.S. efforts in basic research, basedon fears of foreign appropriation of Americandiscoveries, are misguided and could underminethe central role that the federal governmentplays in funding American basic research.

We readily acknowledge that other countriesbenefit from our basic research. In some cases,U.S. firms have been slow to capitalize on tech-nological breakthroughs. This reflects poorly onU.S. technological development and is not a fail-ing of the basic research enterprise.

The foreign benefit from American basicresearch need not be our loss. Performing initialbasic research should position us to exploit ben-efits faster than can the followers and the “free-loaders.” If we reduce our efforts in basic research,we will lose this “first mover” advantage. Also,by promoting an open two-way environment fornew knowledge, we are able to share in other coun-tries’ basic research findings, even to the pointof exploiting those findings more quickly thanfirms in those countries.

Finally, a robust basic research enterprise inAmerican universities keeps our high-tech firmson-shore, while also attracting foreign firms toinvest within our borders. The scientists andengineers at our research universities are anenormous international strategic advantage.

In sum, more basic research activity aroundthe globe will present tremendous opportunitiesfor scientific progress, with large payoffs for allnations. Nationalistic attitudes and policies arecounterproductive to this progress and have noplace in a productive global basic research enter-prise. It is critical, then, that the United Statestake a leadership position in ensuring the free

flow of fundamental knowledge and basicresearch findings globally.

Intellectual Property Worldwide

Although CED does not view the interna-tional spillover of non-proprietary knowledgederived from basic research as a negative, we areconcerned about the impact of weak intellectu-al property laws in other countries on innovationin the United States and globally. U.S. patentlaws play a very large role in stimulating inno-vative activities in this country by protecting acompany’s rights to discoveries and innova-tions that are proprietary in nature; in this way,the company is able to capture the returns on itsinvestment. In an increasingly global environmentand one in which a growing number of countriesare investing in research, it is in the interest of allcountries to play by the same rules regarding intel-lectual property. The pirating of intellectualproperty that occurs in other countries ulti-mately hurts all levels of innovative activity,from basic research to commercial development.

Pursuing International Collaboration

Some areas of science have become too expen-sive and too risky to be supported solely by asingle country. Because of their ultimate potentialbenefit, such projects deserve international coop-eration— in funding, institutional collaboration,and sharing of scientists and results. Mistrustbetween countries in such projects ignores thereality of global science today. A great deal of sci-entific information already flows freely acrossnational borders through publications, research con-ferences, and through the Internet and other formsof information technology. By cooperating in large-scale projects, the United States and other coun-tries extend globalization to projects that are notsustainable at a national level. CED believes theUnited States must pursue international coop-eration in “big science,” in order to optimizeadvances in science and technology.

Supporting Foreign Scientists and Engineersin the United States

In an increasingly global scientific community,American training of foreign scientists is anoth-

47

er welcome form of international research co-operation. The participation of foreigners inAmerican universities and in American scienceand engineering programs benefits the U.S.research enterprise and the economy in gener-al.62 Foreign students contribute to our basicresearch through participation in universityresearch. Given the capacity of the Americanresearch university system, it is unlikely thatthese students are significantly “crowding out”high-potential American students. Many for-eign graduates will choose to remain in theUnited States as professional scientists and engi-neers (about 40 percent in recent years).63

The direct participation of foreign students andprofessionals in American basic research willlikely decline in the future, as foreign economiesbecome more sophisticated. How the UnitedStates reacts to these trends is critical to thefuture of our own basic research enterprise. CEDbelieves the United States should continue to makereasonable efforts to attract foreign scientists.Our immigration policies should be furtherliberalized to allow foreign scientists and engi-neers to live and work in the United Statesthrough permanent visas. Also, immigration poli-cies should encourage foreign scientists and engi-neers to visit the United States on a temporarybasis as consultants or participants in researchcollaborations.

CONCLUSION

We are in the midst of a remarkable period ofdiscovery and innovation. Ask those engagedin the discovery process what they think about

the future of their particular discipline and theanswer is likely to be strikingly positive, with pos-tulations about breakthroughs that are unfath-omable to us today. Add to this optimism the impactof the current economic boom, which is makingthe most of product innovations and bringing newtechnologies to market at a dizzying pace. Together,these trends create an atmosphere of almostboundless enthusiasm about the future.

Raising a cautionary note about the future ofthe basic research enterprise in this environ-ment, then, is no easy task. Certainly, we do notforesee calamity on the horizon. The UnitedStates will continue to see pay-offs from its basicresearch investments, as it has throughout thiscentury.

Yet, today’s emerging problems, left unchecked,become a potent threat to tomorrow’s research enter-prise. In this sense, the reforms we have recom-mended in this report—shoring up the system ofcompetitive peer review, maintaining adequateresources, and correcting deficiencies in the edu-cational and employment pipeline, to name afew—are more than just fine-tuning. Problems ineach of these areas have the potential to erode thequality and quantity of output we have come toexpect from American basic research.

Dealing with these challenges requires resolve.Certainly, if reforming the national laboratoriesis politically difficult, K-12 educational reform iseven more so. The key for policymakers, and forthe citizens they represent, is to keep their focuson the potential for the future. The pay-offs froma basic research enterprise that is working at itsfull capacity are tremendous, and far too greatto forgo for lack of political will.

48

An important constant in the success ofAmerican basic research has been the substan-tial financial support it has received throughoutmost of the post-World War II era. In terms ofdollars spent and the number of scientistsemployed in research activity (see AppendixBox 1, “Research Dollars Versus Other Inputs”),both total R&D and basic research have expand-ed over the past fifty years.

However, in recent years, as post-Cold Warpriorities and missions are being sorted out,growth in basic research spending relative to GDPhas been weaker. As Appendix Figure 1, page 49indicates, federal basic research is about thesame today as a proportion of GDP as it was in

1970. While industrial support for basic researchhas risen slightly in the last decade, it remainswell below federal support. Moreover, the rateof increase in industrial support for basic researchhas slowed considerably during the 1990’s.

Basic research is not a large share of totalR&D for industry or government (see “WhyDoes Government Support Basic Research?,”page 12). Industry spends considerably moreon R&D overall but much less than the federalgovernment on basic research (see AppendixFigure 2, page 50). Indeed, the large amount ofindustry development spending overshadows allother categories of research spending both inindustry and government. Of the nearly $63 bil-

APPENDIX 1

OVERVIEW OF RESOURCES FOR BASIC RESEARCH

Focusing solely on the financial resourcesdevoted to basic research overlooks otherimportant factors in measuring researchactivity, namely the amount of labor andcapital inputs devoted to science. Whileresearch dollars are frequently used as aproxy for these inputs and for research activ-ity in general, it is important to keep in mindthat these dollars purchase the services ofscientists and engineers, as well as the equip-ment they employ. Does an increase ordecrease in dollars spent on research activi-ties represent a one-for-one movement in thenumber of researchers or amount of researchequipment employed? Or has the cost ofthese inputs—that is, the real cost of doingresearch—changed in a way that our stan-dard cost measures fail to capture?

In fact, the amount of research inputsemployed appears to have kept pace with

dollars spent on R&D. Over the same period(1981-1993) that total R&D spendingincreased about 40 percent in inflation-adjusted dollars, the number of scientistsand engineers engaged in research activitiesalso increased about 40 percent (seeAppendix Figure 3, page 51).

At the same time, attempts to account forthe cost of research at a macroeconomic levelsuggest that some types of research havebecome more expensive relative to otherforms of economic activity.1 Specifically, thecost of university research (reflecting person-nel, equipment, and facilities) rose 10 percentfaster than prices in the economy in general in1992. However, the cost of industrial researchgrew at a slightly slower rate than economy-wide prices, and non-manufacturing R&Dcosts grew at an even slower rate.

APPENDIX BOX 1

RESEARCH DOLLARS VERSUS OTHER INPUTS

49

lion that government spends on R&D annually,$17.7 billion goes to basic research. And of the $133 billion that industry spends on R&D, amuch smaller share, just $8 billion, goes to basicresearch.

The Central Role of Federal Government inFunding Basic Research

The most important source of funds for basicresearch since World War II has been the feder-al government. Through most of the post-WorldWar II period, growth in federal support forbasic research was substantial and quite stable.However, relative to growth in the economy,federal support for basic research has remainedvirtually flat in recent years (see AppendixFigure 1) and long-term federal budgetary anddemographic trends suggest an even less favor-able outlook for the future.

As a relative priority in the federal budget, basicresearch expenditures are well below the peak

achieved during the height of the space pro-gram in the mid-1960’s (see Appendix Figure 4,page 52). However, basic research spending hasmaintained a level of just over one percent of fed-eral outlays in the past decade, a period whentotal federal R&D slipped as a budget priority.The relative importance of basic research, andof R&D in general, has been greatly overshad-owed by rapid growth in outlays for entitle-ment programs and interest on the debt, whichhave crowded out discretionary spending ingeneral, including research spending (seeAppendix Figure 5, page 52).

It is the long-term funding picture, however,that is most striking. The retirement of the babyboomers will place unprecedented demand onthe federal budget, primarily for Social Securityand Medicare.2 Under current policies, entitle-ment and interest outlays are projected to account for70 percent of the federal budget by 2002, comparedto just one-third of the budget during the peak years

Appendix Figure 1

Federal, Industry, and Other Funds for Basic Research as a Percentage of GDP

*Industry data prior to 1986 are not comparable to post-1989 data.**Includes universities’ and colleges’ own funds, state/local government funds to universities and colleges, and funds from other non-profits.SOURCE: National Science Board, Science & Engineering Indicators—1996, (Washington, DC: U.S. Government Printing Office, 1996),Appendix Tables 4-1 and 4-5.

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

1970

Federal

Other**

*Industry

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

50

for research spending (the mid-1960’s). Under sucha scenario, basic research would almost cer-tainly see a substantial decline in funding.

Again, a majority of federal R&D funds isdevoted to development work, which is pri-marily carried out by private companies (seeAppendix Figure 6, page 53). In contrast, the smaller portion of federal R&D devoted to basicresearch is directed primarily to universities,with a very small amount directed to industryand non-profit institutions. As Appendix Figure6 illustrates, the amount of federal basic researchperformed directly by federal agencies (in labslike the National Institutes of Health or theDepartment of Energy’s “national labs”) is smallrelative to the total federal R&D budget.

Most federally-sponsored basic research sup-ports agency missions. In this sense, basicresearch does not generally conform to its stereo-type, which is that it is untargeted and devoid

of any defineable goal beyond the advance-ment of knowledge. In fact, basic research sup-ported by the Departments of Defense andEnergy, and through the National Institutes ofHealth, has very strong goals in specific missionareas. Only 15 percent is funded through theNational Science Foundation, which provides sup-port for basic research outside of the agencymissions and is based primarily on scientificand engineering disciplines.

Federal support for basic research is by nomeans uniform across scientific disciplines. AsAppendix Figure 7, page 53 illustrates, feder-al basic research outlays in the life sciences farsurpass those in other disciplines. In fact, fund-ing for the life sciences is on a par with all otherdisciplines combined. This disparity in fundinglevels appears to reflect a strong public priori-ty placed on health-related research, particu-larly in high profile areas like cancer and AIDSresearch. However, as many prominent scientistswithin and outside of health research havenoted, breakthroughs in medical science oftenrely on research conducted in disciplines farafield from the life sciences, including mathema-tics, physics, computer science, and the social sciences.

The importance of multi-disciplinary researchis illustrated by the experience of the private sec-tor. Several of the case studies presented in the nextsection demonstrate that successful basic researchoften does not mean narrow basic research. Merck’sefforts in AIDS research have been multi-dimen-sional by necessity, given the uncertainties of thebasic research endeavor. The Procter & Gamble casestudy also illustrates the necessarily wide, cross-disciplinary net that must be cast to achieveprogress in particular areas of innovation: forProcter & Gamble, this meant supporting bio-logical research, far afield from the company’s corebasic research activities in chemistry.

Overview of Resources for UniversityResearch

Public and private funding for universityresearch has increased dramatically over thepast four decades, from just under $3 billion in1959 to over $22 billion in 1996 (expressed in con-

Appendix Figure 2

Basic, Applied, and DevelopmentResearch Spending by Government andIndustry for 1997

NOTE: Data are preliminary for this year.SOURCE: NSF, Science Resources Studies Division (data avail-able at www.nsf.gov).

0

20

40

60

80

100

120

140

Bil

lion

s of

Dol

lars

Development

Applied

Basic

Development

Applied

Basic

Government Industry

51

stant 1996 dollars; see Appendix Figure 8, page54). As Appendix Figure 8 indicates, this growthhas been driven largely by increasing federal support, although industry and institutionalsupport have also increased significantly inrecent years. Industry’s share of support hasgrown from 3 percent in 1965 to 7 percent in 1996(see Appendix Figure 9, page 54).

Reflecting the priorities of the federal gov-ernment, and society at large, university healthresearch (“life sciences”) has been by far thelargest recipient of federal and non-federalresources (see Appendix Figure 10, page 55).All fields have seen some increase in fundingbetween 1989 and 1996. At the low end, math-ematical sciences received average annual increas-es of 1.5 percent (after inflation) during thisperiod. At the high end, the social sciencesreceived annual increases averaging 7 percent.The growth rate for life sciences research was 3 percent; this increase in resources greatlyexceeds that for other disciplines because itbegan from a much higher base level.

Although National Institutes of Health fund-ing dominates university research (see “HHS”category in Appendix Figure 11, page 56), otheragencies are more important to disciplines out-side of the life sciences (see Appendix Figure 12,page 56). For example, research in the comput-er sciences relies on Department of Defensedollars for 58 percent of its federal funding.The National Science Foundation is the largestfunding agency for the physical, mathemati-cal, and environmental sciences.

Another important, though indirect, sourceof federal funding for university research comesfrom medical services revenues.3 Universityhospitals have become important providers ofmedical care, particularly to Medicare andMedicaid recipients. As a result, these two pro-grams have indirectly funded both health researchin the university hospital and research else-where in the university through cross-subsi-dization. With the likelihood of more stringentcost controls in Medicare and Medicaid, thisindirect source of funding for university researchis expected to decline.

Appendix Figure 3

Academic Employment and TotalEmployment of Scientists and EngineersEngaged in R&D in the United States

NOTE: Data for academic employment represent doctoral sci-entists and engineers. Other data represent all scientists andengineers.SOURCE: National Science Board, Science & EngineeringIndicators—1996 (Washington, DC: U.S. Government PrintingOffice, 1996), Appendix Tables 3-19 and 5-19.

1981

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

Nu

mb

er o

f S

cien

tist

s an

d E

ngi

nee

rs

Academic Total

1993

52

Appendix Figure 4

Federal Basic Research Spending as a Percent of Total Federal Outlays

SOURCE: Science Resource Studies Division, NSF (data available at www.nsf.gov); Budget of the United States Government, FY 1998.

0.00%

0.20%

0.40%

0.60%

0.80%

1.00%

1.20%

1.40%

1.60%

1.80%

1953

1955

1957

1959

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

– – – – – – – – – – – – – – – – – – – – – –

Appendix Figure 5

Entitlements and Net Interest on the Debt as a Percentage of Total Federal Outlays

SOURCE: Budget of the U.S. Government, FY 1999, historical tables, (Washington, D.C.: U.S. Government Printing Office).

0

10

20

30

40

50

60

70

80

90

100

Per

cen

tage

– – – – – – – – – – – – – – – – – – – – –

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

est.

2000

est.

2002

est.

53

Appendix Figure 6

Federal R&D by Character of Work and Performer 1995

SOURCE: National Science Board, Science & Engineering Indicators—1996, (Washington, DC: U.S. Government Printing Office, 1996),Appendix Table 4-18.

$-

$5

$10

$15

$20

$25

$30

$35

$40

$45

Bil

lion

s of

Dol

lars

Basic Research

Federal

Nonprofit

IndustryUniversity

Applied Research Development Research

Appendix Figure 7

Federal Obligations for Basic Research by Field

SOURCES: Science Resources Division, National Science Foundation, Federal Funds for Research and Development, Detailed HistoricalTables: Fiscal Years 1956-95, NSF 95-319 (Bethesda, Md.: Quantum Research Corp., 1995), and SRS, Federal Funds for Research andDevelopment: Fiscal Years 1993, 1994, and 1995 (Arlington, Va.: NSF, forthcoming).

1985 1995

Envi

ronm

enta

l Sci

ence

s

0

1

2

3

4

5

6

Bil

lion

s of

Con

stan

t 198

7 D

olla

rs

Life

Sci

ence

s

Psyc

holo

gy

Phys

ical

Sci

ence

s

Mat

h an

d Co

mpu

ter

Scie

nces

Soci

al S

cien

ces

Oth

er S

cien

ces

Engi

neer

ing

54

Appendix Figure 8Research Expenditures at Universities and Colleges, by Souces of Funds

SOURCE: National Science Foundation, Science Resources Studies Division, Survey of Research and Development Expenditures atUniversities and Colleges, Fiscal Year 1996.

Federal Government

State & LocalGovernment

IndustryInstitutional

All Other Sources

0

5

10

15

20

25

195

9

196

1

196

3

196

5

196

7

196

9

197

1

197

3

197

5

197

7

197

9

1981

198

3

198

5

198

7

198

9

199

1

199

3

199

5

Bil

lion

s of

Con

stan

t 199

6 D

olla

rs

Appendix Figure 9

Sources of Funding for University Research


1953 1965

1996

All OtherSources

10%Institutional

14%Industry

7%


15% FederalGovernment

54%


10%

Industry3%

All OtherSources

6%

FederalGovernment

73%

Institutional8%

All OtherSources

7%Institutional18%

Industry7%


8%Federal

Government60%

55

Appendix Figure 10

Federal and Non-Federal Funding by Research Field at Universities & Colleges


1989 1996

0

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

Mil

lion

s of

Con

stan

t 19

96 D

olla

rs

1,000

2,000

3,000

4,000

5,000

6,000

Mil

lion

s of

Con

stan

t 199

6 D

olla

rs

Life

Sci

ence

s

Engi

neer

ing

Phys

ical

Sci

ence

s

Envi

ronm

enta

l

Scie

nces

Mat

hem

atica

lSc

ienc

esCom

pute

r Sci

ence

s

Psyc

holo

gy

Soci

al S

cien

ces

Oth

er S

cien

ces,

n.e.

c.

Federal

1989 1996Non-Federal

Life

Sci

ence

s

Engi

neer

ing

Phys

ical

Sci

ence

sEn

viro

nmen

tal S

cien

ces

Mat

hem

atic

al S

cien

ces

Compu

ter S

cien

ces

Psyc

holo

gy

Soci

al S

cien

ces

Oth

er S

cien

ces,

n.e.

c.

56

Appendix Figure 11

Agency Support for Conduct of Research at Universities and Colleges, FY 1996

SOURCE: AAAS Report XXII: Research and Development FY 1998, Intersociety Working Group, (AAAS: Washington, DC), Table I-9.

HHS NSF Defense NASA Energy Agric. EPA Other0

1,000

3,000

4,000

5,000

6,000

7,000

8,000

Mil

lion

s of

199

6 D

olla

rs

2,000

Appendix Figure 12

Federal Academic Research Obligations, by Major Federal Agencies, for VariousSciences. 1993-95 Average

SOURCE: National Science Board, Science & Engineering Indicators—1996, (Washington, DC: U.S. Government Printing Office, 1996),Table 5-11.

Physical SciencesDept. ofAgric.

2% NSF33%

NASA19%

DoD11%

DoE26%

HHS9%

Mathematical Sciences

Dept. ofAgric.0.4% NSF

59%

NASA1%

DoD30%

DoE5%

HHS5%

Computer SciencesDept. ofAgric.

1% NSF35%

NASA5%

DoD58%

DoE1%HHS

0.2%

Life SciencesDept. ofAgric.

6%NSF5%

NASA1% DoD

2% DoE2%

HHS84%

Environmental SciencesDept. ofAgric.

1%NSF48%

NASA23%

DoD19%

DoE9%

EngineeringDept. ofAgric.

2% NSF25%

NASA13%DoD

45%

DoE10%

HHS5%

Social Sciences

Dept. ofAgric.23.1%

NSF23.9%

DoD6.22%

DoE0.22%

HHS46.5%

57

In this appendix, we present two examples ofuniversity patent policies, from the Universityof California system and Cornell University.Both affirm the primary intent of research attheir universities — the pursuit of knowledge —while recognizing the potential for by-productsthat have commercial potential and are suit-able for patenting.

These two policies differ, as do others, on thedistribution formulas for licensing revenues.Although the specific percentages for distribu-tion vary, formulas are generally structured toreward the researcher and the researcher’s depart-ment or laboratory with some percentage ofthese revenues. In this way, some of the rev-enues resulting from research are directed backto research activities.

A common feature of these policies is thestated requirement of researchers to disclosepotentially-patentable inventions to the uni-versity. The university then decides whether ornot to pursue a patent on the invention. Somepolicies state the perogative of the researcher toplace research findings directly in the publicdomain, effectively forfeiting the university’s abil-ity to assert intellectual property rights. However,there is tremendous pressure as well as a finan-cial incentive for researchers to first disclosetheir findings to the university. Universitiesemphasize that this disclosure and the patent appli-cation process need not interfere with the pub-lication of research findings, but filing for the patentmust occur before publication or other forms ofpublic disclosure.

APPENDIX 2

UNIVERSITY PATENTING GUIDELINES

UNIVERSITY OF CALIFORNIA PATENT POLICY(Reprinted with the permission of the University of California)

I. PREAMBLE

It is the intent of the President of the Universityof California, in administering intellectual prop-erty rights for the public benefit, to encourageand assist members of the faculty, staff, andothers associated with the University in the useof the patent system with respect to their dis-coveries and inventions in a manner that isequitable to all parties involved.

The University recognizes the need for and de-sirability of encouraging the broad utilization of the results of University research, not onlyby scholars but also in practical application forthe general public benefit, and acknowledgesthe importance of the patent system in bringinginnovative research findings to practical application.

Within the University, innovative researchfindings often give rise to patentable inven-tions as fortuitous by-products, even though

the research was conducted for the primarypurpose of gaining new knowledge.

The following University of California PatentPolicy is adopted to encourage the practicalapplication of University research for the broadpublic benefit; to appraise and determine rela-tive rights and equities of all parties concerned;to facilitate patent applications, licensing, andthe equitable distribution of royalties, if any;to assist in obtaining funds for research; to pro-vide for the use of invention-related income forthe further support of research and education;and to provide a uniform procedure in patent mat-ters when the University has a right or equity.

II. STATEMENT OF POLICY

A. An agreement to assign inventions andpatents to the University, except those resultingfrom permissible consulting activities without

58

use of University facilities, shall be mandatoryfor all employees, for persons not employed bythe University but who use University researchfacilities, and for those who receive gift, grant,or contract funds through the University. Suchan agreement may be in the form of an acknowl-edgment of obligation to assign. Exemptionsfrom such agreements to assign may be autho-rized in those circumstances when the missionof the University is better served by such action,provided that overriding obligations to other par-ties are met and such exemptions are not incon-sistent with other University policies.

B. Those individuals who have so agreed toassign inventions and patents shall promptlyreport and fully disclose the conception and/orreduction to practice of potentially patentableinventions to the Office of Technology Transferor authorized licensing office. They shall executesuch declarations, assignments, or other docu-ments as may be necessary in the course ofinvention evaluation, patent prosecution, orprotection of patent or analogous propertyrights, to assure that title in such inventionsshall be held by the University or by such otherparties designated by the University as may beappropriate under the circumstances. Such cir-cumstances would include, but not be limited to,those situations when there are overriding patentobligations of the University arising from gifts,grants, contracts, or other agreements with out-side organizations. In the absence of overridingobligations to outside sponsors of research, theUniversity may release patent rights to theinventor in those circumstances when:

(1) the University elects not to file a patentapplication and the inventor is preparedto do so, or

(2) the equity of the situation clearly indicatessuch release should be given, provided ineither case that no further research ordevelopment to develop that invention willbe conducted involving University sup-port or facilities, and provided further thata shop right is granted to the University.

C. Subject to restrictions arising from over-riding obligations of the University pursuant to

gifts, grants, contracts, or other agreements withoutside organizations, the University agrees,following said assignment of inventions andpatent rights, to pay annually to the namedinventor(s), or to the inventor(s)' heirs, succes-sors, or assigns, 35% of the net royalties andfees per invention received by the University. Anadditional 15% of net royalties and fees perinvention shall be allocated for research-relatedpurposes on the inventor's campus or Laboratory.Net royalties are defined as gross royalties andfees, less the costs of patenting, protecting, andpreserving patent and related property rights,maintaining patents, the licensing of patent andrelated property rights, and such other costs, taxes,or reimbursements as may be necessary orrequired by law. Inventor shares paid to Universityemployees pursuant to this paragraph representan employee benefit.

When there are two or more inventors, eachinventor shall share equally in the inventor'sshare of royalties, unless all inventors previ-ously have agreed in writing to a different dis-tribution of such share.

Distribution of the inventor's share of royal-ties shall be made annually in November fromthe amount received during the previous fiscalyear ending June 30th, except as provided for inSection II.D. below. In the event of any litigation,actual or imminent, or any other action to pro-tect patent rights, the University may withholddistribution and impound royalties until reso-lution of the matter.

D. The DOE Laboratories may establish sep-arate royalty distribution formulas, subject toapproval by the President. Distribution of theinventor's share of DOE Laboratory royaltiesshall be made annually in February from theamount received during the previous fiscal yearending September 30th. All other elements of thispolicy shall continue to apply.

E. Equity received by the University in licens-ing transactions, whether in the form of stock orany other instrument conveying ownershipinterest in a corporation, shall be distributed inaccordance with the Policy on Accepting EquityWhen Licensing University Technology.

59

1. Evaluating inventions and discoveriesfor patentability, as well as scientificmerit and practical application, andrequesting the filing and prosecution ofpatent applications.

2. Evaluating the patent or analogousproperty rights or equities held by theUniversity in an invention, and nego-tiating agreements with cooperatingorganizations, if any, with respect tosuch rights or equities.

3. Negotiating licenses and license optionagreements with other parties con-cerning patent and or analogous prop-erty rights held by the University.

4. Directing and arranging for the collec-tion and appropriate distribution ofroyalties and fees.

5. Assisting University officers in negoti-ating agreements with cooperating orga-nizations concerning prospective rightsto patentable inventions or discover-ies made as a result of research carriedout under gifts, grants, contracts, orother agreements to be funded in wholeor in part by such cooperating organi-zations, and negotiating with Federalagencies regarding the disposition ofpatent rights.

6. Approving exemptions from the agree-ment to assign inventions and patentsto the University as required by SectionII.A. above.

7. Approving exceptions to Universitypolicy on intellectual property mattersincluding patents, copyrights, trade-marks, and tangible research products.

F. In the disposition of any net income accru-ing to the University from patents, first consid-eration shall be given to the support of research.

III. PATENT RESPONSIBILITIES ANDADMINISTRATION

A. Pursuant to Regents’ Standing Order100.4(mm), the President has responsibility for allmatters relating to patents in which the Universityof California is in any way concerned. This pol-icy is an exercise of that responsibility, and thePresident may make changes to any part of thispolicy from time to time, including the percent-age of net royalties paid to inventors.

B. The President is advised on such mattersby the Technology Transfer Advisory Committee(TTAC), which is chaired by the Senior VicePresident—Business and Finance. The mem-bership of TTAC includes the Provost and SeniorVice President—Academic Affairs, the Directorof the Office of Technology Transfer, and rep-resentatives from the campuses, DOE Laboratories,Academic Senate, the Division of Agriculture andNatural Resources and the Office of the GeneralCounsel. TTAC is responsible for:

1. reviewing and proposing Universitypolicy on intellectual property mattersincluding patents, copyrights, trade-marks, and tangible research products;

2. reviewing the administration of intel-lectual property operations to ensure con-sistent application of policy and effectiveprogress toward program objectives;and

3. advising the President on related mat-ters as requested.

C. The Senior Vice President—Business andFinance is responsible for implementation ofthis Policy, including the following:

60

The following policy was approved by theExecutive Committee of the Cornell UniversityBoard of Trustees on May 26, 1995 to be effec-tive July 1, 1995.

A. General Statement

The Board of Trustees of Cornell University,recognizing that inventions and discoveries ofcommercial importance may be the natural out-growth of research conducted by faculty, staff andstudents, and desiring to secure both publicbenefit from the applications of such research andenhancement of the University's capacity forsuch research, has established the followingPatent Policy.

1. Cornell University’s primary obliga-tion in conducting research is the pur-suit of knowledge for the benefit and useof society.

2. The University depends upon finan-cial support from governmental agen-cies, private foundations, corporations,operated for profit and others for the basicand applied research endeavors of thefaculty and staff. As University Researchenjoys substantial public support it isincumbent upon the Univer-sity to seekassurance that any resultant patent rightbe administered consistent with thepublic interest.

3. Inasmuch as new ideas and discoveriesof commercial interest are often a con-sequence of University Research, andinasmuch as patent protection can oftenenhance the reduction to a public use-fulness of inventions which result fromUniversity Research, Cornell, as a gen-eral policy, will seek patent protectionfor those ideas and discoveries whicharise out of the research activities of itsfaculty and staff where it appears nec-essary or desirable to do so.

4. It is the judgment of the University thatthe reduction to a public usefulness ofinventions and discoveries resultingfrom University Research, the publi-cation and availability for educationalpurposes of the fruits of such research,and the achievement of a fair and equitable distribution of royalties which acknowledges both the con-tribution of the inventor, and theUniversity can best be assured by oper-ation of a uniform Patent Policy whichprovides for University ownership ofinventions.

B. University Research

University Research shall be defined, for thepurpose of this Patent Policy, to include allresearch conducted in the course of an inventor'semployment with the University (including butnot limited to the performance of a grant con-tract or award made to the University by anextramural agency) or with the use of UniversityResources.

C. Disclosure of Inventions

Inventions conceived or first reduced to prac-tice in furtherance of the University Research offaculty or staff shall be promptly disclosed in writ-ing to the Cornell Research Foundation.

D. Ownership of Inventions

1. All patentable inventions conceived orfirst reduced to practice by faculty andstaff of Cornell University in the conductof University Research shall belong tothe University. The inventor shall coop-erate and assist the University in allphases of the patent application processand shall assign such applications or anypatents resulting therefrom to CornellResearch Foundation, Inc.

2. Patentable inventions made by indi-viduals on their own time and with-

CORNELL UNIVERSITY PATENT POLICY(Reprinted with the permission of Cornell University)

61

out the use of University resources shallbelong to the individual inventor.

3. In cases in which the University has anownership interest in an invention pur-suant to paragraph D(1) and either doesnot file a patent application within oneyear, or fails to make a positive deter-mination regarding pursuit of a patentwithin a period of six months from thedate of disclosure, all of the University'srights shall be reassigned to the inven-tor upon request, subject only to suchexternal sponsor restrictions as mayapply.

E. Royalty Distribution

1. In the case of a patent owned by theUniversity pursuant to paragraph D(1)above, and in recognition of the effortsand contributions of the inventor, totalnet royalty income shall be distributedas follows:

50% of the first $100,000.0025% of net royalty income in excess of$100,000.00

Joint inventors shall share the percentage ofnet royalty income allocated to the Inventor.Any person hired or retained for the purpose ofproducing an invention shall not be entitled toa distribution of net royalty income with respectto that invention.

2. Cornell Research Foundation shallreceive 35% of the net royalty income toprovide operating funds to cover the costof service provided to the University withregard to intellectual property mattersand particularly to cover the costs asso-ciated with patenting and marketinginventions where royalty income orother cost recovery has not been achieved.Cornell Research Foundation's priordeficits shall be retired using this por-tion. The percentage of net royaltyincome to Cornell Research Foundationshall be evaluated annually by the Boardof Directors of the Foundation andreduced when deficits have been elim-

inated. The Cornell Research FoundationBoard of Directors shall be responsibleto adjust the percentage received byCornell Research Foundation with atwo year lead time following the elim-ination of the Foundation’s deficits.

3. Net royalty income shall mean grossroyalties received by the University lessdirectly assignable expenses resultingfrom patenting and licensing the par-ticular invention .

4. The remainder of the net royalty incomeshall be divided (a) 60% to the inventor'sresearch budget, subunit and Universityunit in a manner to be mutually agreedupon and (b) 40% to the University forgeneral research support.

5. For any year in which the net royaltyincome distributed to a unit of theUniversity for a particular inventionemanating from that unit shall exceed20% of the annual sponsored researchas determined by the Office of SponsoredPrograms for that unit in that year, theexcess received from Cornell ResearchFoundation shall be retained as endow-ment for the unit. The Dean or Directorof the unit may similarly require that cor-responding royalty income to a sub-unit exceeding 20% of the total sponsoredresearch of the inventor's appropriate sub-unit be retained as endowment for thebenefit of the subunit. In the event thata lump sum royalty payment contributesto the generation of excess royaltyincome in a given year as defined above,Cornell Research Foundation may dis-tribute such lump sum payment to theunit or subunit over a three year peri-od together with accumulated interest.In such case, the provisions of this para-graph shall apply to the resulting annu-al distributions.

6. In the case of an invention unresolvabledispute over distributions in 4 (a), netroyalty income distributed under 4 (a)shall be allocated and made on an equi-

62

table basis as determined by the VicePresident for Research.

F. Licensing Policy

It is the general policy of the University toencourage the development and marketing ofinventions resulting from University research soas to reach a public usefulness and benefit. It isrecognized that furtherance of such a policymay require various forms of agreements includ-ing the granting of exclusive licenses.

Cornell Research Foundation may, in appro-priate circumstances with due consideration tothe prospective licensee and when consistent withlaw applicable to federally supported research,license an existing patent or invention on anexclusive basis for a reasonable period up tothe full term of the patent, provided that suchan exclusive license shall contain provisions topromote the likelihood that the invention pro-vides a public benefit, including but not limit-ed to a requirement of diligence and march-inrights where the licensee does not adequately per-form.

G. Waiver Requests

Waiver of any provisions of the Patent Policyshall be granted only in extraordinary and com-pelling circumstances and pursuant to the pro-cedure described below.

A request for waiver of any of the provisionsof this Patent Policy shall be submitted to thePresident of Cornell Research Foundation &the Vice President for Research for transmittalto the Patent Advisory Committee. Such requestshall include an identification of the provisionor provisions of the Policy requested to bewaived, and a full explanation of the reasons forthe waiver including, but not limited to, themanner in which the waiver is consistent withthe educational purposes of the University andthe public interest.

The University recognized that certain spon-sors may wish to impose as a condition of theaward of contract or grant funds special provi-sions which are at variance with this PatentPolicy. Under such circumstances, the Universitymay entertain such proposals as requests forwaiver under this paragraph subject to the addi-

tional condition that all faculty or staff membersengaged in research to be supported by the pro-posed grant or contact containing such provisionsshall acknowledge and accept those specificprovisions.

The Patent Advisory Committee shall revieweach request for waiver and submit a report ofits findings and recommendation to the VicePresident for Research whose decision shall befinal. Each action under this section shall con-sidered on its own merits in light of all of the factssurrounding the particular request and shallhave no implication for consideration of sub-sequent requests. Waiver of provisions relatingto distribution of net royalty income shall, in addi-tion, require the approval of the Dean or Directorof the unit from which the invention emanated.

H. Deferral

This statement of Patent Policy shall not pre-vent participation under research agreements with,or the conduct of research for, governmentalagencies (local, state or federal) subject to lawsor regulations which require a different dispo-sition of patent rights than herein provided, orimpose other provisions which are in additionto, or inconsistent with, its provisions. Suchprovisions of this Patent Policy as are inconsis-tent therewith shall be deem superseded and theprovisions of such laws and regulation shallapply.

I. Patent Management Agencies

Cornell Research Foundation may make suit-able arrangements not inconsistent with theprovisions of this Patent Policy with patentmanagement agencies or firms for the purposeof obtaining services and advice with respect tothe patentability of inventions, the obtainingof patents thereon and the management andlicensing of any such patents.

J. Patent Agreements

In order to facilitate a distribution of patent rights and benefits consistent with the provisionof this Patent Policy, each participant in UniversityResearch shall execute a Patent Agreement.Pursuant to such Agreement each participant shall acknowledge that all such research is sub-

63

The Vice President for Research shall reportto the Board of Directors of Cornell ResearchFoundation and the President of Cornell Universityupon matters of significance relating to theadministration of this Policy.

Notes:

1. Use of University office space or libraryfacilities shall not constitute a use ofUniversity resources for this purpose.

2. For the limited purpose of this policy, staffmembers shall also include all researchassistants, graduate research assistants,teaching assistants, fellows, studentswho provide services under sponsoragreements which require Universityownership, and others who utilizeUniversity resources in the furtheranceof their research.

3. The distribution provisions containedherein shall apply to all existing andfuture inventions. The distribution tablecontained at paragraph E(1) shall beapplied on a cumulative basis to all netroyalty income earned during the life ofan invention, and not annually.

4. Direct expenses include the costs ofobtaining patent protection for the par-ticular invention and all marketing,promotion and licensing costs related tothe particular invention.

5. Typically the inventor's Department,School, Section or Center.

6. Typically the inventor's College.

ject to the terms of this Patent Policy, and shallagree to cooperate with the University or itsdesignee in the assignment to the University ofpatent rights in inventions or discoveries con-ceived or first reduced to practice during suchresearch and the preparation and prosecutionof patent applications, as may be required in orderto implement its provision.

This requirement may be waived by the VicePresident for Research only in those limitedcases where University Research occurs withina discipline in which the prospect of a patentableinvention is, in his or her judgment, extremelyremote.

K. Patent Advisory Committee

The Vice President for Research shall, after con-sultation with the Research Policy Committee ofthe Faculty Council of Representatives, establishand appoint, subject to the approval of the Boardof Directors of the Cornell Research Foundation,a Patent Advisory Committee which shall serveat his or her pleasure. It shall be the function ofthe Committee to advise and recommend tothe Vice President for Research with respect to:

1. guidelines and procedures for imple-mentation of this Patent Policy,

2. proposed amendments to the PatentPolicy,

3. the granting of individual exceptions tothis Policy,

4. the University's ownership of particu-lar inventions,

5. such other matters as the Vice Presidentfor Research may deem appropriate.

64

In order to deepen our own understandingof the role of basic research in the innovationprocess and the ways in which companies andinstitutions employ and conduct basic research,CED asked its Trustee companies and universitiesto offer representative examples of basic researchin their organizations. This section of the report

contains a collection of these responses, assem-bled as a series of case studies. We believe theyare enormously illuminating in their own right,and can be read with great interest indepen-dent of the report. But they also reinforce the find-ings contained in the Policy Statement.

CASE STUDIES

AIDS RESEARCH AT MERCK & CO., INC.(Prepared for CED by Merck & Co., Inc.)

America’s pharmaceutical industry has achieved global leadership by discovering morelife-saving medicines by far than any pharma-ceutical industry of any other country. In thepast 20 years, of all important drugs discoveredand introduced globally, nearly half have comefrom U.S. companies. In the American wing ofthis important industry, Merck has long been atthe forefront of innovation.

With this record of accomplishment in mind,it would be helpful to look at one of Merck’s suc-cessful research programs for guidelines as to howthe United States might sustain and strengthenits science-technology base. Particularly usefulin this regard should be the recent experiencesof the Merck Research Laboratories in discoveringand developing Crixivan (indinavir sulfate), thecompany’s new protease inhibitor for the treat-ment of HIV infection and AIDS.

Merck started this work in 1986, only three yearsafter the virus that causes the disease had beenpositively identified. Scientists at NIH and else-where were making important progress, butmuch research remained to be done on a virusand disease that were still poorly understood.During the next ten years, Merck scientists helpedimprove the knowledge of both. They conduct-ed studies that any university scientist would identify as basic or fundamental. These studieswere tightly intertwined with research that wasclearly developmental. All of Merck’s research,

basic or otherwise, was “applied” because itwas directed toward the development of usefultherapies for this deadly disease.

By way of contrast, when Merck launchedits research in the mid-1970s aimed at develop-ing inhibitors that would lower blood cholesterol,the laboratories were building on over 30 yearsof basic research into lipid biosynthesis. Most ofthat research had been conducted at the NIH orin research universities in the United States andabroad and very little of it would be classified as“applied.” Its immediate goal was to develop abetter understanding of fundamental biochem-ical processes, not to develop practical, effec-tive therapies (such as Mevacor and Zocor, forpersons suffering from hypercholesterolemia). That particular program and its outcome providesan excellent example of the kind of cooperativeinteraction among public, nonprofit, and prof-it-making institutions that has made the UnitedStates a leader in pharmaceutical innovation.

But in the case of AIDS, the pandemic was soserious, the public and public health authoritiesso correctly and deeply concerned, that Merckstarted a major, multi-pronged research pro-gram long before a solid foundation of basicresearch existed. Merck attacked the problemon several fronts because neither the Merck lab-oratories nor the scientists at NIH and the uni-versities could predict with any degree of certaintywhich approach would be successful.

65

Merck tried for several years to develop avaccine against HIV infection. The company wasencouraged by the fact that its Virus and CellBiology Research had successfully produced theworld’s first recombinant DNA vaccine againsthepatitis B (Recombivax HB), as well as by thefact that Merck would be able to partner withbiotechnology firms which could bring to bear theirspecial expertise.4 One of these firms, Repligen,discovered the V3 loop on the surface of thevirus, and Merck’s scientists tried to use thatportion of the virus to make a vaccine. The com-pany also probed the use of monoclonal antibodies(working with MedImmune) as a means of com-bating HIV infection. Both of these approachesfailed. As Merck’s vaccine researchers observedin 1990, “the mechanism by which initial virusinfection occurs is not well understood....”5 In layterms, they were groping in the dark.

Today Merck is building upon fundamentalresearch in the HIV vaccine field conducted bythe Merck Research Laboratories and by othersin the last five to six years. Merck’s scientists haveseveral new directions and technologies toexplore. As their knowledge of the virus andthe process of infection improved, they could seemuch more clearly what was likely to work andwhat was likely to fail. Virus and Cell BiologyResearch at Merck was able to refocus its vaccineprogram as a result. But due to the inadequatebase of fundamental research on the immunol-ogy of HIV infection and HIV vaccinology, sev-eral years had been lost in the effort to developa safe and effective vaccine.

Fortunately, during these same years, Merckand other companies were exploring severaldifferent paths to a successful HIV therapy. Themajor targets were the enzymes involved in theprocess by which the virus transcribes its RNAinto proviral DNA, which is then integratedinto the host cell’s genetic material. One of thesewas reverse transcriptase, which controls thefirst step in viral replication. AZT (zidovudine)is one of the compounds that interrupts the HIVlife cycle by inhibiting the natural nucleosides.But AZT’s adverse side effects, relatively low antiviral potency, and the ability of the virus todevelop mutations that conferred resistance tothe drug prompted Merck’s scientists to look

for more effective, non-nucleoside analogueinhibitors of reverse transcriptase.

After screening about 23,000 compounds overa two-year period, Merck discovered severalnon-nucleoside inhibitors (RTIs). But as the earlyclinical trials indicated, the virus successfullymutated around these drugs, as it had aroundAZT.6

That put all of the pressure on the third branchof HIV research, which was focused on the pro-tease enzyme. Some years earlier, federally sup-ported cancer research had resulted in thepublication of a paper analyzing the role a pro-tease enzyme played in the spread of a virusassociated with tumors in chickens. Merck scientistsfamiliar with that research suspected a proteaseenzyme also played an important role in HIVreplication. But they were not absolutely certainthe protease was a good target until Dr. Nancy Kohland her colleagues at the Merck ResearchLaboratories completed an elegant experiment thatproved the protease was crucial to the replicationof HIV. In the tradition of basic science, they pub-lished their findings in the Proceedings of theNational Academy of Science.7 Researchers at Merckand NIH then established the crystal structure ofthe protease. This research could have been con-sidered proprietary information of value to com-petitors, but the firm’s scientists also quicklymade these findings available to others.8

As this work progressed, MRL researchersmade other vital contributions to the under-standing of the disease and its treatment. Theyconducted analyses of HIV turn-over and pre-dicted on the basis of their study of one ofMerck’s non-nucleoside RTIs that HIV infectionis a highly active, continuous process. As the sub-sequent work of David Ho at the Aaron DiamondAIDS Research Center, Tony Fauci at NIH, andXiping Wei at the University of Alabama-Birmingham demonstrated, the virus replicatesat a high rate and over years erodes the capaci-ty of the immune system to keep it in check.Eventually, patients progress toward AIDS.

Prior to Crixivan, all anti-retroviral therapieshad limited effectiveness insofar as they medi-ated only transient declines of plasma viral lev-els. We now know—due to the work of scientistsat Merck, other pharmaceutical companies, NIH,

66

and academic institutions—that these therapiesfailed because of the eventual selection of resis-tant viral variants. The research of Merck’s JonCondra and his colleagues pointed other scien-tists toward a new concept of the progression ofthe infection and the development of resistancein the presence of suboptimal antiviral therapy.Merck studies of viral resistance to Crixivanestablished that resistance requires the continuedaccumulation over time of multiple mutations:there is a high “genetic barrier” to resistance.Suppression of the ongoing replication made itless likely that the virus would develop themutations it needed to produce resistance.Successful antiviral therapy, they established,had to reduce circulating viral RNA levels belowthe limit of detection in the available assays.9 Thisperspective is now generally accepted and thedesired decline has been achieved by patients usingCrixivan in triple combination therapy (withAZT and 3TC).10

Were there other potential targets for HIV/AIDSresearch? Yes, scientists might have developeda fourth line of attack by targeting the integraseenzyme, which performs the function of incor-porating the HIV genome in the host cells. Thiswould have been another means of preventingHIV from infecting new cells. Until recently,however, little was know about the integraseenzyme. Additional resources funneled intobasic research early on could have accelerated thisaspect of the search for effective HIV therapies.

So too in basic vaccine research. Companieslike Merck could have avoided some blind alleysand focused their resources more tightly onother targets if the basic research foundationhad been broadened and deepened in the 1980s.

Given the initial lack of fundamental knowl-edge about the virus or disease, however, Merckcan take considerable satisfaction in what has beenlearned and accomplished in the past decade.Throughout Merck there is great pride in hav-ing developed Crixivan. Without some elegantchemistry, a good measure of brilliant processresearch and engineering, and an unusuallydetermined effort in the manufacturing divi-sion, Merck could never have produced enoughof this complex molecule to complete clinical tri-als and then to treat the many patients whowere desperately in need of therapy.11 Nor couldMerck have organized a compassionate use pro-gram for patients who had exhausted all otherHIV therapies. This was done a full year beforethe FDA and other regulatory agencies world-wide cleared Crixivan for marketing.

For the present, it would suffice to say thatthroughout the years that it took Merck to devel-op this new drug, there was no visible seams andcertainly no quality distinctions between researchthat others might categorize as basic or applied,fundamental or developmental. From theCompany’s point-of-view, it was all good med-ical science serving society, which has beenMerck’s central mission for over a century.

COLUMBIA UNIVERSITY AND VIMRx PHARMACEUTICALS(Prepared for CED by Columbia University)

In 1996, Columbia University’s research expen-ditures amounted to $232 million, the vast major-ity of which ($194 million) derived from federalgrants and contracts. In recent years, however,a small, but growing, portion of the researcheffort at Columbia has involved collaboration withindustry that has taken various forms (such asdirect funding of research by industry, jointresearch projects, and partnerships in existingresearch centers).

Through the active efforts of the ColumbiaInnovation Enterprise (CIE)—a unit of the

Provost’s Office devoted to identifying, evaluating,protecting, and licensing intellectual property;encouraging technology transfer; and increasingprivate sector funding for research and devel-opment—Columbia has been exploring newforms of collaboration with industry. In thespring of 1997, Columbia and VIMRxPharmaceuticals, Inc. reached agreement on anambitious collaboration that could, if successful,form a model for similar partnerships at Columbiaand elsewhere.

Under the terms of the agreement, a sub-

67

sidiary of VIMRx (in which Columbia holds a 10percent ownership stake) will have an option forexclusive licensing of technology developed atthe Columbia Genome Center. In return, theVIMRx subsidiary will provide $30 million forresearch at the Genome Center over a five-yearperiod. The Columbia Genome Center is aresearch unit devoted to mapping, sequencing,gene discovery, and technology developmenton the genomes of human and selected modelorganisms. The explosion of new technologyand information on the structure of genes has per-mitted the localization and identification ofnovel human genes associated with many genet-ically based diseases; at the Columbia GenomeCenter, investigators have been at the forefrontof the race to identify specific genes associatedwith such diseases as cancer, late-onset Alzhei-mer’s Disease, epilepsy, manic-depressive disorder,and glaucoma. VIMRx is a development-stagecompany involved in the development of tech-nology to treat and prevent disease by control-ling disease-triggering flaws in the geneticchemistry of individuals.

Both parties to the agreement see potential ben-efits of the new arrangement which will permitthe translation of forefront research results intouseful, commercially viable medical products.VIMRx will have access to world-class researchthat it could not expect to carry out on its own;for its part Columbia will receive a significant infu-sion of research funding.

The collaboration will extend for a period offive years, after which Columbia and VIMRxwill have an option to continue the arrangementindefinitely upon mutual consent.

As a case study in progress, the collaborationraises a number of questions and issues about uni-versity-industry interactions:

• Does participation in the broad research pro-gram of a university research center lead to moreor different commercial benefits for a companythan would more targeted funding of partic-ular, individual research projects?

• Is the nature of the research undertaken by auniversity research center changed signifi-cantly by the need, real or perceived, to gen-erate commercially viable research results?Is there a tendency to focus on more appliedresearch? Is very basic research threatened?

• Are there effects on the traditional opennessof university research that result from theproprietary demands of the commercializationprocess?

• Do arrangements such as this encourage otherindustries to participate in similar ways, or doesan exclusive arrangement between a univer-sity and a particular company make it moredifficult or less desirable for other compa-nies to become involved?

PFIZER INNOVATION(Prepared for CED by Pfizer Inc.)

Pfizer is a health care company committed toinnovation as the key to improving the health andlives of people. Pfizer’s products, such as Norvasc(for hypertension), Zoloft (for depression),Zithromax (for bacterial infections), and Diflucan(for fungal infections) have helped millions ofpatients worldwide. These products play a cen-tral role in treating their corresponding diseasesby delivering quality outcomes which are muchimproved over those offered by earlier therapies.Unfortunately, many other conditions await

safer, more efficacious, and more convenienttreatments. Lack of such treatments results in dis-abilities, early deaths, decreased quality of life,and enormous economic costs. Aging of thepopulation and rising cost pressures will createfurther demand of new, cost-effective healthcare solutions. These demands will be met by excit-ing new technologies for diagnosing, treating,and preventing illness which are now emergingfrom rapid advances in the basic biomedicalsciences.

68

The Challenge of Basic Biomedical ResearchThese recent and ongoing advances have

grown out of a long-term commitment to basicbiomedical research conducted in universities,medical centers, research institutes, and gov-ernment laboratories. Academic research in thebiomedical sciences has focused on improvingunderstanding of the human body, other organ-isms, and disease processes. In the course ofaddressing essential questions about how and whythings occur, such research provides funda-mental scientific insights into biological function.An example is the concept of apoptosis, or pre-programmed cell death, an apparently importantcomponent of many disease processes. Basicresearch also results in essential tools and tech-niques for conducting further research. Coretechnologies such as recombinant DNA/genet-ic engineering, nucleic acid amplification andprobes, gene transfer, transgenic animals, cell cul-ture, immunoassays, and monoclonal antibod-ies have both come out of and supported furtherbasic research infields including molecular andcellular biology, microbiology/virology, immunol-ogy, neuroscience, physiology, pharmacology,and genomics. These technologies have subse-quently been combined and applied in waysunforeseen by original researchers, opening upentirely new avenues of scientific inquiry and tech-nical development.

A case in point is the Human Genome Project.Begun by the National Institutes of Health in1990 with the goal of mapping and sequencingthe entire human genome, this effort has beenmade possible by earlier, basic advances infields as disparate as molecular biology, genet-ics, microbiology, computer science, and robot-ics. Harnessing the efforts of multiple academiccenters, this effort will ultimately provide abasis for better understanding of underlying bodyfunction and disease processes by delineatingthe structure, function, and interaction of thegenes which control the production of pro-teins, the key actors of biology. It has alreadyprovided intriguing insights into disease occur-rence and process, such as breast and coloncancers and obesity, while spawning new tech-nologies such as bioinformatics. Genomics is rep-resentative of how biomedical research continues

to provide new methods and topics for inves-tigation, while it also highlights the growingunderlying complexity of health and diseaseknowledge.

Apart from providing basic laboratory tools,how does such academic research help lead to newdisease therapies? From the perspective of drugdiscovery, understanding disease processes andstructure in detail permits identifying points ofintervention (mechanisms and targets). Similarly,work in pharmacology and organic chemistry maysuggest particular or better approaches to accom-plishing this goal (lead concepts). Finally, meth-ods of assessing the validity and correspondenceof particular targets and leads (assays) may uti-lize some of the core technologies mentioned. Inaddition to improving conventional small mol-ecule drug development, genomic information willbe crucial to developing new diagnostic andtherapeutic modalities, including gene testing, pep-tide molecules, antisense (DNA blocker drugs),vaccines, and gene therapy. Academic medicalresearch centers also provide the sites and patientsfor testing new therapies in clinical trials, whichproduce the ultimate assessment of their utilityin patient care.

Pfizer’s ResponsePfizer has responded to advances in basic

science by seeking to harness and integrate thesenew technologies, while at the same time fosteringinteraction with academic researchers. In contrastto the explanatory focus of basic academicresearch, Pfizer’s role centers on how to meet spe-cific clinical needs through the applied anddirected development of science and technolo-gy. By spending approximately $2 billion in 1997on discovery and development of new drugs, afigure which has risen dramatically over thepast decade, Pfizer seeks to integrate basic sci-entific findings and new technologies to dis-cover and develop novel therapeutic agentsmeeting important patient needs. This repre-sents an intrinsically prolonged, risky, and veryexpensive innovation process which must beactively managed. And this process is becomingeven more expensive and risky with the risingcomplexity of underlying science and new dis-ease states targeted.

69

How does basic biomedical research factorinto Pfizer’s discovery and development efforts?First and foremost is the central role that publishedscientific literature and meetings play in stimu-lating thought and generating ideas which pro-mote the drug discovery and developmentprocess. Whether it is the structure of a receptortarget or pharmacologically active compound, orsome basic insight into pathophysiology, such workprovides valuable input into Pfizer’s directedefforts. Also, all of the new laboratory tools andtechnologies derived from earlier basic researchhave been incorporated by Pfizer into its drug dis-covery efforts. Finally, Pfizer scientists, who rep-resent the most critical resource in this process,trained at universities by conducting basic research.And in a way which parallels the integration ofcore technologies in drug discovery and devel-opment, these scientists are now working inmulti-disciplinary, cross-functional project teamswhich are able to leverage and pursue new devel-opments more rapidly and efficiently.

Pfizer supports academic research through avariety of means, including research collabora-tions with financial support, fellowships, and unre-stricted grants for basic research in syntheticorganic chemistry. Pfizer also conducts large-scale clinical trials of drugs in development in col-laboration with academic health centers.

Pfizer’s methods and strategy for innovationhave integrated basic research findings and tech-nological tools to create a fundamentally new drugdiscovery and development process. Methods havefocused on increasing productivity by increasingthe overall volume of compounds assessed, whileat the same time optimizing target selection.Pfizer has been a pioneer in incorporating genom-ic information into drug discovery. By providinginsight into disease processes and targets, this infor-mation has been combined with high speed com-binational chemistry (which creates vast numbersof potential drug-like compounds) and highthroughput screening techniques (which teststhem for possible drug properties by checking fitand correlation with targets), technologies derivedin part from basic science research. Technologyaccess strategy in this increasingly rich, com-plex, and fast-moving environment has focusedon early stage collaborations with a variety of small-

er, entrepreneurial firms and academic researchers.The total number of collaborations has reached270, placing Pfizer in a position to understand andfoster basic research. This approach gives Pfizerthe flexibility to follow the directions and oppor-tunities opened up by science, taking advantageof unforeseen spill-overs into new areas rather thanbeing limited to specific, pre-set fields of devel-opment.

Pfizer has referred to these collaborativeefforts as a “Web of R&D Innovation” whichhas produced tangible results. For each ongoingproject, 1000 compounds a week can now besynthesized (versus 6 five years ago), whilethousands of compounds can be screened eachweek. In 1996, 20 million such tests were performedon 100,000 compounds, yielding 18 new drug can-didates. This represents 70 staff-years to gener-ate a new drug candidate, compared to 100-250staff-years for other firms. Use of these newtechnologies has effectively reduced time frominitial idea generation (project start) to candidatenomination (preclinical development) from justover 4 to 2.3. staff-years, with an additional 12months of study required prior to first human tri-als. Pfizer now has over 100 research projectsunderway - more than at any time in its histo-ry. Many of these gains have resulted in part frombasic biomedical research. The final result willbe more, innovative new drugs available topatients sooner.

Key Final PointsSome key lessons have emerged from this

review of basic biomedical science research andPfizer’s corresponding role in medical technol-ogy development:

• Innovation is central to improving health andquality of life, resulting in important productswhich meet significant patient needs.

• Basic biomedical research plays a crucial rolein providing a healthy environment for inno-vation by providing fundamental insightsinto normal body function and disease, devel-oping new technological tool and techniques,and expanding human knowledge capitaland skill base.

70

• Results of basic biomedical research are oftencreatively applied and combined in unforeseenand unpredictable ways, making short-termand directed planning approaches risky andlikely detrimental to long-term innovation.

• Industry and academia each play vital yetdifferent roles in this process, with industryfocusing on integrating basic science find-ings in a directed, applied process of manageddrug development, which has been trans-

formed by new technologies and multi-dis-ciplinary team approaches.

• This development enterprise is considerablydifferent from the inquisitive, open-endednature of basic academic research.

• Basic biomedical research needs continuedsupport from all sources in order both realizeits exciting current potential and chart new pathswhich will ultimately form the basis for futureinnovation.

PROCTER & GAMBLE BIO-ENGINEERED ENZYMES(Prepared for CED by Procter & Gamble)

The use of enzymes in synthetic detergents goesback over 30 years to their first application in bothEurope and the United States as a cleaningenhancement. Many difficult-to-remove stains,such as grass, blood, and fats consist of verylarge molecules that are more easily removed whentheir size is reduced. Enzymes such as proteas-es and lipases are extremely effective in selectivelysolublizing these stains and removing them.

Enzymes are well-established biological prod-ucts derived from fermentation reactions. Theirdiscovery, production and usage dates back tothe original work of Louis Pasteur over 100years ago. But enzymes are also difficult mate-rials to make effective given their sensitivity tothe reaction conditions created in a typical wash-ing machine cycle. They are only active in a fair-ly narrow temperature range, are highly sensitiveto the solution pH, and their catalytic performanceis affected by trace contaminants.

Despite these limitations, enzymes were suc-cessfully introduced into detergents in 1960.While their performance enhanced the overallcleaning of the washing products, there wereimportant concerns about allergic reactions dueto inhalation. Inadequate industrial hygiene insome manufacturing facilities resulted in a lowlevel of employee sensitization, and a concern aboutlarger scale problems in the general population.These concerns resulted in a highly cautiousapproach to greater usage, higher performancelevels or a broader range of enzyme materials.In 1974, the detergent industry voluntarily with-

drew all enzymes from the market until theindustry could assure their safety. They werenot reintroduced until enzyme-containing par-ticles were developed that essentially avoided theinhalation issue.

In the late 1970’s the emergence of new sciencein biotechnology created the possibilities forbio-engineered materials tailored for specificapplications, as well as the dream of much lowerusage levels. The fermentation products suc-cessfully used up until this time were not wellcharacterized. But advances in analytical chem-istry—a P&G core competency—were now allow-ing a better understanding of the protein structureand composition of the most desirable materials.Nevertheless, the field of biotechnology wasnew to Procter & Gamble and it had to make astrategic decision on how to enter this new andexciting technical field.

A few highly committed technology man-agers and research scientists became convincedthat Procter & Gamble could develop the exper-tise and exploit biotechnology effectively and safe-ly in our consumer products. They set up a smallbiotechnology group by recruiting talent pri-marily from key departments in leading uni-versities to augment experienced chemists on ourstaff. This became the catalyst for technical dis-coveries.

Early on, Procter & Gamble recognized thatcommercialization of basic scientific discoverieswould require experienced industrial partners.Procter & Gamble approached the two leading

71

companies, Novo in Denmark and Genencor inthe United States, to propose joint developmentrelationships. Genencor was at the leading edgein molecular modeling and gene transplanting.Novo was the most advanced in fermentationscale-up. Genencor, a small start up biotechnol-ogy company, was enthusiastic and Procter &Gamble worked initially with them. Novo becameinterested after initial product successes beganto have an impact on their enzyme business.While Procter & Gamble was not known for itsbiological science expertise, its track record of com-mercializing new technology in products havinga well established market success, and in largequantities, was appealing.

For nearly a decade Procter & Gamble jointlyworked to develop a basic understanding of thebiological sciences required to utilize recombinantDNA technology to tailor the genetic make up ofthe most promising enzymes. This allowed us todevelop more effective, genetically engineeredenzymes and gain some useful industrial expe-rience with these materials. Akey organization inter-vention was the linkage of the basic researchorganization with the scientists within the laun-dry technology group who would have respon-sibility to determine the efficacy of the newenzymes as soil removers in the laundry cycle. Thenthere was the final linkage to the product devel-opment groups worldwide. They had the ultimateresponsibility to assess the final product perfor-mance and process conditions required to suc-cessfully manufacture bio-engineered detergents.

The culmination of this work was the identi-fication of a unique, highly specific cellulaseenzyme which was recently and exclusivelyintroduced into P&G detergents worldwide.This new cellulase has the capability to selectivelyremove the fuzz and pills from frequently laun-dered color cotton garments that otherwisewould look aged and worn. In doing so, thisenzyme allows the original fabric color to showthrough. It, therefore, provides the “new fabric”color even after extensive washings.

To dimensionalize the technical challenge,we identified a single, highly effective cellulasefrom an enzyme cocktail. This unique materialwas successfully isolated, cloned and the entireprocess scaled up to a commercial level. Methods

were developed to stabilize it in the detergent man-ufacturing process and tailor its usage for differentfabrics in our key worldwide geographies. All thiswas done while solving the key technical andpotential public relations problem of determin-ing the correct level of enzyme so that thedefuzzing and depilling process was no more detri-mental to fabrics than other detergents.

In the process Procter & Gamble also learnedthat fuzz and pills became damaged after mul-tiple launderings, and in addition to losing theirdyes faster than the rest of the yarn, they inhib-ited the basic detergent cleaning process. That is,if clothes don’t have damaged fuzz and pillsthey are less likely to pick up and retain dirt. So,what started out as a color-retention technologyturned out to have an even larger benefit as a clean-ing enhancement ingredient.

As Procter & Gamble has restructured itsapproach to basic research over many years, ithas developed some effective guidelines.

First, constantly reach out for new scientificadvances. While many discoveries occur entire-ly within Procter & Gamble’s own lab walls, his-tory says that no research organization has the fullcapability to make all of its own scientific discoveries.

Second, when an important emerging relevanttechnology appears, consciously create internalexpertise and build a strong core technologicalbase. To deal with the breadth of capable outsideresearch organizations, core internal expertise iscritical to maximize the value, choose relation-ships and transfer expertise.

Third, utilize the company’s technology leadsto attack the most important business problems.This is the linkage of technology and businessstrategies, or, marrying what’s needed withwhat is possible.

Finally, commercialize technologies. It isimportant to drive new technology advancesinto superior performing products, packagesand processes. This may entail the use of anexternal partner who can supplement our capa-bility to accelerate speed to market. The innovationgame is played in the marketplace. There are nowins in the labs! Great science is only effectiveif it results in new commercial opportunities.And the entire organization must realize this isthe only true measure of its effectiveness.

72

Knowledge that Makes a DifferenceMind/Brain/Behavior (MBB) is a university-

wide initiative at Harvard University that inves-tigates the complex interaction of biology andculture that gives rise to human thought, feel-ing and behavior. If our brains are so similar, whyare human beings so different? When and whydo human beings act irrationally? How doesthe brain change over time? How do cultures andbrains interact to influence health?

Acollaborative effort among some of Harvard’sleading faculty, MBB was created in 1992 byHarvard’s President Neil Rudenstine along withfour other Interfaculty Initiatives. He envisioneda Universitiy that could unite its nine distinct fac-ulties and bring its vast but decentralizedresources to bear on problems of importanceto society. MBB helps realize this vision by focus-ing on questions and problems of human poten-tial and vulnerability that require a creativeuniting of forces because they fall between thecracks of traditional disciplinary inquiry.

MBB engages experts from diverse disci-plines in explorations of themes relevant to oursociety. Imagine a discussion about drugs andaddictions aimed at limiting the abuse of drugsand reducing the undesirable side-effects ofprohibition, that avoids rhetoric and partisan-ship, and incorporates the latest research inneuroscience, psychology, criminology, andpublic policy.

Mind/Brain/Behavior represents an innova-tive approach to the study of human nature.MBB is committed to thinking about the impli-cations and applications of knowledge as it is tothe adventure of discovery. Where other institu-tions have brought neurobiologists together withpyschologists to advance knowledge about humanbeings, Harvard has cast its net far wider. MBBenables public policy experts, lawyers, theologians,physicians, anthropologists, historians, philoso-phers, and organizational researchers to joinforces with laboratory scientists in neuroscienceand psychology to develop projects that are inno-vative and designed to make a difference to insti-

tutional communities—from hospitals to schoolsto work organizations—outside the academy.

Faculty Fellowships and Working GroupsMBB’s interdisciplinary work takes root in a

Faculty Fellowship comprised of 35 scholarsfrom across the University. The Fellowship gath-ers six times a year for a series of seminars anddiscussions centered on the complex interac-tion of biology and culture. The faculty havediscovered that the single greatest barrier toeffective collaboration is misunderstanding.Each scholar brings to the table assumptions, ter-minologies and ways of approaching problemsthat are unique to his or her discipline. These dis-cussions enable faculty to exchange perspec-tives and methodological approaches in theservice of generating new ideas.

Through their interactions in the Fellowship,faculty identify specific topics ripe for interdis-ciplinary investigation that no single disciplineor department has managed to address effectivelyalone. Smaller collaborative groups—calledworking groups—are formed to enable morefocused study of these topics. Each group ischaired by a faculty fellow and includes a broadcross-section of faculty from both within andoutside the Harvard community. The workinggroup topics fall into one of three broad categories:reason and emotion, learning and development,and health and illness.

MBB has already seen concrete results from itsworking groups. For example, the working groupThe Brain in the Sociocultural World produced alandmark conference and related book on theplacebo-effect in healing. The Drugs and Addictionsgroup recently briefed top federal officials oncurrent issues in U.S. drug policy, and the work-ing group Early Brain Development is planning anational conference on the brain bases for read-ing difficulties.

Research and EducationA committee comprised of members of MBB’s

Faculty Fellowship awards funds for innova-

HARVARD UNIVERSITY’S MIND/BRAIN/BEHAVIOR(Reprinted with the permission of Harvard University)

73

tive, cross-disciplinary research projects. Facultymembers, students, and post-doctoral scholarsworking in the mind, brain and behavioral sci-ences are encouraged to submit proposals tothe Research Committee. Recently funded pro-jects include: the relationship between visualperceptions and mental processing and the neu-robiological origins of anxiety and fear.

Faculty fellowship members have also creat-ed an undergraduate certificate program. The pro-gram is hosted in four departments in the College:biology, computer science, history of science,and psychology. The program integrates MBBcoursework common in all departments, inter-disciplinary seminar work, electives, and hon-ors thesis. Innovative new courses designed byMBB faculty include: “Human Behavior and the

Developing Brain;” “The Philosophy ofNeuroscience;” “The Biology of Morality;” and“Remembering and Imagining.”

Structure and LeadershipMBB is led by co-directors—a psychologist and

an historian of science—and by a steering com-mittee consisting of faculty from different schools.The directors set MBB’s research and teachingpriorities and oversee the Initiative’s activitiesin consultation with the steering committee andwith the Provost of the University.

Mind/Brain/Behavior is an ambitious exper-iment. In keeping with President Rudenstine’svision, MBB seeks to reconnect with life outsidethe university by uniting fractured faculties in theservice of complex problems that matter to us all.

XEROGRAPHIC PROCESS RESEARCH(Prepared for CED by Xerox)

This case study demonstrates the unpre-dictable, non-linear nature of the entire R&Dprocess, as well as the way in which firms employa “fill-in-the-gap” approach to conducting basicresearch. In this case, the need for fundamentalscientific insights came after a product hadalready been brought to market. Yet, improve-ments in product performance were not simplya matter of fine-tuning; there were significant gapsin the knowledge of how certain processesworked that could only be addressed through basicresearch. The story unfolds over six decades,from Chester Carlson’s invention in 1938 to theawarding of the 1997 American Institute ofPhysics Prize for Industrial Applications ofPhysics to a Xerox researcher.

The invention of xerography by ChesterCarlson in 1938 represents one of the signifi-cant technology achievements of the 20th Century.Xerographic copying and printing has exerted amajor impact on the democratization of infor-mation on a global scale. Although the basicrequirements for xerography were identified byCarlson in his early work, the successful com-mercialization of the technology through theintroductions of the Xerox 914 copier in 1959required a number of subsystem and material engi-

neering advances. In a six-year period followingthe invention, Carlson was unsuccessful in per-suading 20 business equipment companies to con-sider the commercial development of the process.Finally, in 1944, Carlson established a workingrelationship with Battelle Memorial Institute, aprivate research foundation in Columbus, Ohio.Many advances were made at Battelle in xero-graphic materials and processes. Some notableinventions included two component developer(in which the toning powder was triboelectricallycharged by mixing with larger carrier beads), amor-phous selenium photoreceptors with 1000 timesmore sensitivity to light than the sulphur usedby Carlson, and electrostatic transfer of thedeveloped powder image to paper. The HaloidCorporation, a small photographic paper com-pany in Rochester, New York, acquired a licenseto the process in 1947. Further technologyadvances enabled the marketing of severalcopiers in the ‘50s. The company’s name waschanged to Haloid-Xerox Corporation in 1955 andXerox Corporation in 1961 following the wide mar-ket acceptance of the legendary Xerox 914 whichautomatically produced plain paper copies froman original page placed on the platen of themachine.

74

Although the first Xerox products introducedin the late ‘50s and early ‘60s were commercialsuccesses, the scientific foundations of many ofthe xerographic process steps and materialswere poorly understood. Up to this point xerog-raphy had been the province of inventors, tech-nologists, and business people: it was now theturn of the scientists to analyze and extend theprocess to enable high copy speeds and improvedcopy quality. For example, the Xerox 9000 fam-ily of products, introduced in 1974, generated 120copies per minute. A research center in Webster,NY was established in the mid-’60s for the pur-pose of studying the xerographic process toenable improvements in subsystems and mate-rials. The intent was to spawn new productsthat embodied improved image quality, higherreliability, lower manufacturing cost and high-er copying speeds. Major efforts were mountedto study charge carrier generation and trans-port in amorphous semiconductors, the electri-cal and rheological properties of toner (powder)materials processes for developing electrostaticimages, gaseous ion charging physics, mechanismsof triboelectrification and particle adhesion,processes for fusing toner to paper, and cleaningmethods for removing residual toner from the pho-toreceptor. Due to the multi-disciplinary natureof this effort, teams consisting of physicists,chemists, materials scientists, electrical engi-neers and mechanical engineers were assem-bled. The research efforts by Xerox and itscompetitors over the subsequent years haveadvanced the performance of xerographic tech-nology to a level that approaches offset printingquality and productivity.

The xerographic research effort was orga-nized around the individual steps in the xero-graphic process. These are: 1) photoreceptorcharging, 2) light exposure to form an electrostaticimage, 3) development of the electrostatic imagewith charged, pigmented poser (called toner), 4)electrostatic transfer of the developed toner topaper, 5) fixing of the toner to the paper with heatand/or pressers, and 6) cleaning of any residualtoner from the photoreceptor before the nextcycle. Research programs at Xerox have enabledsignificant advances in all of these process steps,as well as in photoreceptors themselves. This fact

can be illustrated by exploring the chronologyof advances in the performance of xerographicdevelopment systems.

The designs for xerographic developmentsystems have gone through a number of gener-ations since the cascade development systemused in the Xerox 914 with two componentdeveloper. The two component mixture wascascaded over the photoreceptor to develop theelectrostatic image. Only line copy and the edgesof broad image areas were developed with thistechnique. To improve the development of broadimage areas, the next generation of magneticbrush development systems was introduced inthe early 1970s. The developer material con-sisted of polymer-coated, magnetic carrier beadswhich enabled the developer to be transportedon a rotating roll containing stationary perma-nent magnets. The proximity of the roll to the elec-trostatic image on the photoreceptor and thebrushing action of the non-conducting devel-oper enabled improvement in the developmentof board image areas, a technique called two-com-ponent insulative magnetic brush development.Nevertheless, using this technique, the achieve-ment of adequate image density development athigh speeds (120 copies per minute) required alarge development system with five developmentroles. To provide improved high-speed broad areadevelopment with a compact design, another gen-eration of magnetic brush designs was intro-duced for which the carrier beads were madeconducting for the purpose of bringing the effective development electrode closer to theelectrostatic image on the photoreceptor. In thissystem, however, the effectiveness in develop-ing fine features in the image was comprised com-pared to images developed with insulativemagnetic brush development. In addition to thecascade and magnetic brush systems, otherdevelopment system designs have been utilizedfor special applications. For low speed copiersand printers, single component developmentsystems are widely used. In this case, the toneris triboelectrically charged by a metering blade inself-spaced contact with a rotating sleeve. Forthe development of colored images, most colordevelopment systems use a gentle magnetic brushof insulative two component developer in com-

75

bination with an AC electric field to generate a tonercloud in the development zone. For each gener-ation of these designs, extensive research hasultimately resulted in Xerox products.

Magnetic brush development system designsare favored for high-speed copiers and printers.The insulative magnetic brush systems provideexcellent development of image details but arenot efficient in developing broad image areas. Onthe other hand, conductive magnetic brushdevelopment systems are efficient in developingthe broad areas, but compromise image detail.To understand the shortfall in the broad areaperformance of insulative magnetic brush devel-opment, the physics of two component developermaterials and magnetic brush development sys-tems was intensely studied at Xerox during the‘70s and ‘80s. From this work, it was determinedthat the limitation on broad area development witha magnetic brush was due to a net carrier beadcharge caused by toner deposition onto the pho-toreceptor. The identification of this limitationmechanism and a corresponding broad areadevelopment model provided important newinsights into critical material and process para-meters. In understanding the cause of the per-formance shortfall, a new system design calledHighly Agitated Zone (HAZE) developmentwas invented which provided both excellentbroad area and image detail development withhardware that was smaller (fewer rolls) andlower cost. This design was first incorporated ina 62-copies-per-minute Xerox product, the Xerox1065 Marathon copier, introduced in 1987. Theproduct received “Product of the Year” awardsfor two years in a row.

Over the past 10 years, this product family hasproduced more than 100 thousand machines inthe U.S. Furthermore, the HAZE design has sub-sequently been incorporated in several addi-tional major Xerox product platforms for copiersand printers. Particularly noteworthy is theinclusion of the design in a highly successfulfamily of high-speed (135 prints per minute)duplicators/printers introduced in 1988. Thedevelopment system technology continues toplay a key role in a number of new productofferings. The combined revenue from all of theproducts embodying this development system

was nearly half of Xerox’ total revenue in 1996.In recognition of the contribution of the HAZEdevelopment process to the success of three gen-erations of Xerox’ current products, and of theinvention of further development systems like-ly to be used in Xerox’ future products, theinventor of the process was awarded the 1997American Institute of Physics Prize for IndustrialApplications of Physics.

In summary, xerography evolved from the keyinvention of Chester Carlson and subsequentmaterial and process inventions at Battelle whichprovided the foundation for copier products atthe Haloid Company and ultimately led to thefirst automate electrophotographic copy machine,the commercially successful Xerox 914. The suc-cess of this product and the need to generate sub-sequent generations of improved versionsmotivated the establishment of a research lab-oratory organized around the xerographicprocess steps and devoted to generating anevergreen stream of improved xerographicmarking engines for both copiers and printers.Improvements in each process step generated byresearch were incorporated into new subsys-tem designs which in turn were offered to thepublic in successive generations of Xerox copiersand printers during the past four decades. By wayof an example, the generations of developmentsystem designs were reviewed to illustrate the sequence of advances driven by researchprograms.

A more detailed description of the HAZEprocess serves to illustrate how fundamentalinsights gained through basic research can resultin superior product performance over multiplegenerations of products. Although this casestudy has emphasized the role of the develop-ment system design in contributing to theadvanced performance of copiers and printers,comparable pay offs have been achieved with otherxerographic materials and subsystems. The netresult is that investments in basic research haveprovided a highly leveraged source of profits forgeneration after generation of Xerox products fromthe Xerox 914 to the recently announced Xerox6180 digital press which incorporates the HAZEdevelopment system process to produce 180prints per minute.

76

The Research function in IBM is located in threeU.S. laboratories and four international labora-tories in Beijing, Haifa, Tokyo, and Zurich. Thetechnical population consists of nearly 2000engineers and scientists, most of whom havePhDs. The scope of activities is broad as wouldbe expected of a company covering nearly everyarea of information technology. In addition to workthat address the product needs of the IBM busi-nesses, there are activities in new and alternativetechnologies and areas of basic research thatcontribute to the scientific understanding under-lying the technologies in our business. The largestpart of the Research funding comes from corporateheadquarters, followed by funding from thevarious IBM product and sales organizations. Avery small amount of funding comes directly fromthe outside in the form of government contractsand occasional external revenue sources.

Research remains organizationally independentfrom IBM’s development functions, creating bothopportunity and responsibility with respect to itsrole in the IBM Corporation. Such independencepermits the division to explore new technologiesand applications, carry out basic research, and pro-vide independent technical advice to IBM regard-ing current and future product directions beingundertaken by the product groups. With such inde-pendence goes the responsibility of contributingrecognized value to the IBM corporation throughsuch means as creating intellectual property, cre-ating and developing leadership technologiesfor rapid infusion into the IBM product line,enhancing IBM’s technical image through pub-lic awareness of its world class science and tech-nology, and creating a technical view of the futurethat can provoke and guide the product andsales groups in the creation of their businessstrategies. To carry out our mission in IBM,Research works with more than 30 organiza-tions within the corporation.

To be a successful research organizationrequires that the right programs be defined andthat these programs be effective, both in carry-ing out the research but also in translating the

results into values that can be leveraged by therest of the company.

The Research planning process begins with anannual exercise to identify the future trends ofthe information technology business. Althoughfocused on the basic technology trends, e.g. thereduction of size and increased speed of circuitsor capacity of storage, societal and businesstrends are also examined in order to define theexternal environment within which IBM willexist in the near future. Although IBM science andexploratory programs often create the gamechangers for its industry, Research is keenlyaware that many of the major changes comefrom laboratories and companies outside IBM andtherefore we expect our professionals to be tunedto the external science and technology commu-nities. The need for a clearer view of the future,particularly in areas other than basic technolo-gy, has motivated Research to establish a smallgroup of specialists in long term market planning.Using this view of the future, IBM’s technical strate-gists prepare plans for the year that take intoaccount not only the anticipated trends but alsothe nearer term requirements of the IBM units withwhich it collaborates to provide technical lead-ership in IBM’s product lines.

Having created an annual plan of activities itremains for the management to monitor andtrack the success of the research in delivering valueto IBM. At the end of each year, the top divisionmanagement gathers to review the work of theyear. The review includes evaluations by otherIBM divisions regarding delivery of results, thedegree to which collaborative relationships havebeen successful in delivering needed technolo-gy and direction to other parts of IBM (alignmentof strategy, people relationships, etc.), the num-ber of U.S. patents filed for the year, the numberof external awards and other recognition receivedby IBM researchers (Nobel prizes, NationalMedals, Fellows, etc.), and self assessments of keyresults for the year, both in the field of science aswell as those affecting IBM. These measures aretaken quite seriously by both the management

IBM RESEARCH(Prepared for CED by IBM)

77

and non-management research personnel, asthe final result gets translated into compensationfor Research employees. Most importantly, this

thorough review of Research’s activities is an excel-lent source of feedback about how well IBMresearch is contributing to the company.

BBN AND THE DEFENSE ADVANCED RESEARCH PROJECTS AGENCY (Prepared for CED by BBN/GTE Internetworking)

Maintaining the nation’s economic well-beingand preserving its security are two of the federalgovernment’s most important responsibilities.It has long been recognized—and often establishedas public policy—that innovation and technologicalpreeminence contribute to both goals by creat-ing new capabilities, and giving rise to newindustries and jobs. Nowhere is this more evi-dent than in the development of computer tech-nology; and no office of government has beenmore effective in this arena than the DefenseAdvanced Research Projects Agency (DARPA).

Originally created as the Advanced ResearchProjects Agency (ARPA) by President DwightEisenhower shortly after the Soviet Unionlaunched its first Sputnik satellite in 1957, itwas designed to be a fast-response, R&D mech-anism that would ensure American leadershipin future technologies. Over the years, howev-er, ARPA shifted its focus to “high-risk, high-pay-off,” long-term basic research, and researchsponsorship of new ideas. By drawing talent fromthe nation’s finest universities, public, and pri-vate research laboratories—and by allowingscientists to apply their ideas through experi-mentation—ARPAbuilt a community that includ-ed some of the best technical and scientificminds in American research.

For more than thirty years, BBN Technologieshas been a vital member of that community,and its long and productive relationship with theagency has led to a host of innovations; mostnotably the ARPANET, the forerunner to theInternet. At the time of the network’s develop-ment in the late 1960s, agency scientists were usingdifferent computers, running different operat-ing systems, in different parts of the country. ARPAreasoned that, by electronically linking these

machines, researchers at various institutionscould more easily share resources and results.But that would require building an experimen-tal network based on packet switching—a newmethod of transmitting data by dividing electronicmessages into small, uniform segments.

When ARPA sent out its request for propos-als to develop the packet switches—which werefirst called Interface Message Processors (IMPs)—BBN responded with an extensive and detaileddescription of the entire network, incorporatingtest programs and performance checks, expla-nations of how to handle congestion and recov-er from computer and line failures, andcomputations, equations, and tables dealingwith queuing of packets and transmission delays.Once it was chosen for the job, the Company’sengineers designed the software to form a fullyintegrated network. They studied interactionsamong programs to ensure the actions takenby one program did not conflict with actions takenby others. And they created a system where thecomponents worked together smoothly, so mostusers did not sense its underlying complexity.By the early 1970‘s, BBN had successfully devel-oped and quickly expanded the first network ofpacket switches.12 But it took more than IMPsto make the emerging network truly useful.

The initial goal of designing, implementing,and fielding the ARPANET generated new inter-est in the kinds of protocols and distributedcomputing technology that could further enhancethe utility of the experimental network. By 1975,protocol investigation within DARPA led to theconcept of a network of networks—which ulti-mately became the Internet—and to TCP/IP asa means for universal interconnection. At the sametime, BBN, as part of both DARPA’s research com-

78

munity and the university computer scienceresearch community at large, began investigat-ing the impact of computer-to-computer com-munication capabilities on new applicationssuch as distributed air traffic control and unifiedmedical records across hospitals; and on theoperating systems software needed to deliver com-munication services to those applications.13, 14

Early work in these areas proved conclu-sively that such applications were complex anddifficult to build. This led to two parallel tracksof development. One concentrated on the directimplementation of a very limited set of imme-diately useable applications, including e-mail, tel-net, and ftp; while the other focused on additionalinfrastructure to enable more general-purposeapplications to be developed using wide area com-munications capabilities. The latter activityspawned what has come to be known as “mid-dleware,” because it was strategically placedbetween the network and the application to pro-vide a richer, simpler network environment fora wide variety of uses.

With continuing DARPAsupport, and employ-ing the experimental method championed by theagency to test and refine new ideas, BBN devel-oped, tested and evaluated two generations of mid-dleware systems throughout the later part of the1970’s. The first generation provided solutions fordistributed computing in a homogeneous sys-tems environment, based on a message-passingparadigm. The second perfected this work anddeveloped additional solutions for a more general,heterogeneous system environment. But one thatwas based on a remote procedure-call form of ori-entation. Though these technical investigations werevery successful, the impact of applying them eco-nomically on the prevailing computer technicalinfrastructure was less so.

Indeed, the computing landscape at the time wasshifting from one of a few large computer systemsto many smaller, cheaper systems enabled byinnovations in chip fabrication; and from a few largenetworks to many smaller, locally managed net-works enabled by technologies such as Ethernet.Moreover, the transformation from a program-ming environment dominated by custom assem-bly language to a variety of high level languages(e.g. “C,” Lisp), enhanced the conception and exe-

cution of more complex applications. It was becom-ing clear that the proliferation of new technologieswould depose many of the older computing hier-archies. It was clear, too, that middleware wouldhave to change to keep pace with the size, scopeand variety of the computer technology base.

Consequently, BBN’s third generation mid-dleware system, Cronus, which began in 1981,was based on the development of distributed objectcomputing.15 This approach also enabled TheCompany to address the challenge of creating com-plex, distributed applications in an extensivelyheterogeneous and rapidly changing environment.The sort of environment required by the militaryto effectively operate the variety of new computerand communication technologies which werefast being developed and adopted.

As BBN’s knowledge of the middleware tech-nology matured, financial support for develop-ing a new distributed, object-based, middlewareinfrastructure moved to the Air Force RomeLaboratory. By the mid-1980’s, BBN engineers hada working Cronus system suitable for evaluationand use by government projects. The first trialsinvolved developers working with early adopterson prototype concept demonstrations. Each ledto further refinements in the technology, and tobetter understanding of how it could be used byapplication developers. In the late 1980’s, theNavy’s San Diego R&D Lab joined its Air Forcecounterpart in testing the technology on opera-tional problems that were being addressed by BBNapplication developers for a number of Navy andDARPA programs.

These trials were extremely successful andfielded new network-based capabilities quicklyand cost-effectively, particularly when comparedwith non-middleware-based development pro-jects. Distributed application engineers at BBNworked with Department of Defense (DoD) con-tractors and operational personnel to deploysystems such as CASES, TARGET, and DART, andwere able to tackle significant operational issuesfor distributed command-and-control applica-tions.16, 17 Along with conducting various tests andevaluations, they trained DoD contractors and uni-versity engineers, programmers, and systemmanagement personnel to use the new technol-ogy. Subsequently, the systems were incorpo-

79

rated into advanced concept demonstrations byother DoD and DARPA contractors. Universitiesand industrial research labs also synthesized theresults with their own technical investigations, fur-ther extending middleware’s use and refine-ment.18 Over time, successes such as these, indistributed object computing, led to several large-scale, distributed integration programs makingworld wide military command-and-control oper-ations more responsive and cohesive.

By the start of this decade, capabilities like thoseof Cronus began to appear in commercial prod-ucts from major companies such as Digital, IBM,SUN Microsystems, and Microsoft, as well as fromsmall startup firms. But because no two sys-tems were alike, operating between them wasextremely difficult. To remedy the situation,commercial vendors and users of distributedobject technology joined forces to set acceptableindustry standards. They formed the ObjectManagement Group, and established the CommonObject Request Broker Architecture, or CORBA.Their grassroots efforts attracted a great deal ofinterest, first, from vendors, and then from userswho saw it as a way to voice their requirementsfor the newly emerging commercial offerings. Intime, Cronus was adapted to the CORBA stan-dard and given the new name Corbus.

Since its development, middleware technol-ogy has played a major role in the growth of dis-tributed networks, including the Internet. Infact, the World Wide Web is, itself, a special-ized form of middleware. Prior to its emergence,many people only saw utility for middleware anddistributed applications in the context of local area

computing. Now, with the increasing popular-ity of the Internet, middleware becomes even moreimportant as a means to simplify, and moreeffectively manage, an extremely complex envi-ronment. In addition to the Web, there is also Java,with its own brand of middleware, to competewith DCOM and CORBA, and scores of nar-rowly specialized technologies employing mid-dleware solutions.

Today, middleware represents just one ofsundry successes that have resulted from thelong relationship between BBN and DARPAthat have continually raised the technology bar.Just as important, it is a notable example of howthe transition from basic to applied R&D is notalways well-delineated. This occurs in manyfields, but it is especially true in computer science,where systems are man made and abstract innature, and differ considerably from the morefamiliar patterns associated with the phy-sical sciences. Innovations such as middlewareactually combine basic and applied activitiessimultaneously. While the earliest investigationsrepresent fundamental research, in that theyseek functional organization of ideas and abstrac-tions, the later tests, evaluations and refine-ments are clearly applied, because they identifyparticular uses and users, and establish practi-cal applications. This is often the case for manyDARPA-sponsored programs, which are intend-ed, not only to develop potential capabilities, butto demonstrate—early on—the feasibility of newconcepts prior to their actual widespread accep-tance and availability.

80

Page 5, James Q. Riordan

I would add another paragraph to this list stat-ing that while federal support for basic researchis necessary, we also need to have basic researchconducted in the United States by the private sec-tor. In the past, CED has supported tax incentives

for private research. This report does not call forsuch incentives. We should, however, call forthe elimination of all tax disincentives for U.S.research. Every dollar spent on research in theUnited States should be treated as a U.S. sourcedeductible expense in computing taxable incomeand available foreign tax credits.

MEMORANDUM OF COMMENT, RESERVATION, OR DISSENT

81

CHAPTER 2

1. The Procter & Gamble case study (see Case Studies section) illus-trates the necessity for diverse basic research activities to meet inno-vation goals. P&G saw promise in biological research, a scientificfield in which the company previously had no expertise. Recognizingthe potential, they developed a basic research capacity in thisarea, primarily through cross-industry alliances, with a considerablepayoff in the form of a better detergent product.

2. As explained in Box 2 on page 12, these characteristics of basicresearch, including the economic benefits that flow to those notinvolved in the discovery, provide justification for government fund-ing.

3. The theoretical framework for these growth studies was devel-oped in the 1950’s by Robert Solow of the Massachusetts Instituteof Technology, who was awarded a Nobel Prize for his work. Thepioneering empirical work in this area was undertaken by EdwardDenison while he worked for the Committee for EconomicDevelopment. See Edward F. Denison, The Sources of EconomicGrowth in the United States and the Alternatives Before Us, (Washington,D.C.: Committee for Economic Development, 1962).

4. Dale W. Jorgenson “Investing in Productivity Growth," inTechnology and Economics, (Washington, D.C.: National AcademyPress, 1991), p.59.

5. The new theory is associated with the work of Paul Romer ofStanford University. For a brief summary of these issues, see J. BradfordDeLong, “What Causes Economic Growth?” (paper prepared fora symposium on tax policy sponsored by the American Councilfor Capital Formation, Washington, D.C., December 5, 1996).

6. “The internal rate of return provides a concise method for com-paring different investments. Suppose, for example, the R&Dinvestor could alternatively invest in a savings account with a con-stant interest rate. The internal rate of return on the R&D invest-ment answers the question: what would the interest rate on the savingsaccount have to be in order to make the investor indifferentbetween putting the investment in the savings account and mak-ing the R&D investment?”

7. Most rate-of-return studies use an R&D measure rather than amore narrowly defined basic research measure.

8. M.I. Nadiri, Innovations and Technological Spillovers, WorkingPaper no. 4423, (Cambridge, Mass.: National Bureau of EconomicResearch, 1993). Nadiri’s consensus estimate was derived from asurvey of 63 rate of return studies.

9. MIT: The Impact of Innovation, ABankBoston Economics DepartmentSpecial Report, (March 1997). “MIT-related companies” are firmsfounded by MIT graduates.

10. E. Mansfield, “Academic Research Underlying IndustrialInnovations: Sources, Characteristics, and Financing,” The Reviewof Economics and Statistics 77 no. 1, (1995), pp. 55-65.

11. Francis Narin, Kimberly S. Hamilton, and Dominic Olivastro,“The Increasing Linkage Between U.S. Technology and PublicScience,” Research Policy, 26(3) (1997) 317-330.

12. For examples of this “new thinking,” see Terence Kealey, TheEconomic Laws of Scientific Research, (New York: St. Martin’s Press,1996), and “What Price Science?” Business Week, 26 May 1997.

13. Microsoft’s Bill Gates has suggested that the scientists who arecurrently sequencing DNA might provide the “algorithms” incarbon that can be transferred to silicone for purposes of creatinga “computer that learns.” Remarks made at the American Associationfor the Advancement of Science Annual Meeting, February 1997,Seattle, Wash.

14. Total basic research spending from all sectors was $31.2 billionfor 1997 and GDP in that year was $8,083 billion.

CHAPTER 3

15. According to the National Science Board, federal governmentsupport for basic research was $17.1 billion in 1995 (59 percent oftotal support), industry support was $7.5 billion (25 percent of totalsupport), and other support. including universities’ own funds, non-profits, and state/local governments, was $5 billion (17 percent oftotal support). National Science Board, Science & EngineeringIndicators—1996 (Washington, D.C.: U.S. Government PrintingOffice, 1996), NSB 96-21.

16. Bush was the wartime director of the Office of ScientificResearch and Development. His influential 1945 report, “Science,the Endless Frontier,” was an important point of reference forpost-war R&D policy.

17. Harvey Brooks and Lucien Randazzese, “University-IndustryRelations: The Next Four Years and Beyond,” in Investing inInnovation: Creating a Research and Innovation Policy That Works, ed.Lewis Branscomb and James Keller, (Cambridge, Mass.: MITPress, 1998).

18. Brooks and Randazzese, “University-Industry Relations: TheNext Four Years and Beyond.”

19. The industry case studies section illuminates the way in whichindustry views basic research.

20. Before passage of the Bayh-Dole Act, there was no government-wide policy concerning ownership of inventions made under fed-eral funding. Agencies were generally reluctant to grant ownershiprights to an invention to the inventing institution. Bayh-Dole andsubsequent amendments to it, all of which were finalized in 1987,made it government-wide policy to grant such rights. Bayh-Doleand its related amendments today constitute the “operating man-ual” for technology transfer offices at universities and other insti-tutions receiving federal research funds.

NOTES

82

21. Universities’ licensing of their intellectual property has result-ed in a rapidly expanding source of revenue, though still very smallrelative to total research dollars. According to the Association ofUniversity Technology Managers, royalties on licenses were $274million in 1995 for surveyed universities, an increase of 108 per-cent from 1991 (see also Figure 3, page 23). The survey sample includes87 of the top 100 research universities in the United States, rankedaccording to research dollar volume.

22. In “Research Universities and the Marketplace,” (page 22), wediscuss the problems that can arise as a result of university-indus-try partnerships and the patenting of university research. InChapter 4, we offer principles which we believe provide theappropriate safeguards to facilitate technology transfer without imped-ing basic research.

23. Wesley Cohen, Richard Florida, Lucien Randazzese, and JohnWalsh, “Industry and the Academy: Uneasy Partners in the Causeof Technological Advance,” in Challenges to Research Universities,ed. Roger Noll, (Washington, D.C.: Brookings Press, 1998).

24. Medical school professors are a notably large exception; manyare clinical professors with minimal or no teaching responsibili-ties. See Linda Cohen, “Biomedical Research Support and theDecline of Medical Services Research Subsidies at Medical Schools,”in AAAS Science and Technology Policy Yearbook, (Washington,D.C.: AAAS, 1998).

25. One-quarter of scientists and engineers who have received gov-ernment support have obtained funding from more than one fed-eral agency. See National Science Board, Science & EngineeringIndicators—1996 (Washington, D.C.: U.S. Government Printing Office,1996), NSB 96-21.

26. National Science Board, Science & Engineering Indicators—1996(Washington, D.C.: U.S. Government Printing Office, 1996), NSB96-21.


28. This competition for researchers is not limited to universities.Prominent researchers can be the focus of intense competitionbetween universities and other nonprofit research institutes. As wedescribe in “Nonprofit Institutions,” (page 26), these nonprofits canbe attractive alternatives for researchers because of lower fund-rais-ing and administrative burdens. In this way, competition betweenuniversities and the nonprofits puts pressure on universities to keepthese burdens to a minimum.


30. Over the life of the patents, which expired on December 2, 1997,the universities expect revenues to total over $220 million from 369licensees.

31. Wesley Cohen and Lucien P. Randazzese, “Eminence andEnterprise: The Impact of Industry Support on the Conduct ofAcademic Research in Science and Engineering,” (working paper,Belfer Center for Science and International Affairs, John F. KennedySchool of Government, Harvard University, 1997); David Blumenthal,Nancyanne Causino, et al., “Relationships Between AcademicInstitutions and Industry in the Life Sciences — An IndustrySurvey,” The New England Journal of Medicine, (February 8, 1996);

Walter Powell and Jason Owen-Smith, “Universities and theMarket for Intellectual Property in the Life Sciences,” Journal of PolicyAnalysis and Management, Vol (17) No (2), pp. 253-277. For a sum-mary of studies in this area, see Harvey Brooks and LucienRandazzese, “University-Industry Relations: The Next Four Yearsand Beyond,” in Investing in Innovation: Creating a Research andInnovation Policy that Works, ed. Lewis Branscomb and JamesKeller (Cambridge, Mass.: MIT Press, 1998).

32. Basic research conducted by the federal government accountsfor 9 percent of total basic research in the United States (NSF,Science and Engineering Indicators 1996, Appendix Table 4-5).

33. The Washington Post, 11 September 1997, p. A23.

34. In this case, the federal subsidy does not come directly froman allocation of federal dollars for the project. Industry will pro-vide the cash resources to cover salaries and other expenses.Nonetheless, there is a substantial indirect federal subsidy involvedin making the three national labs available for the project.

35. See Kenneth M. Brown, Downsizing Science: Will the UnitedStates Pay a Price?, (Washington, D.C.: The AEI Press, 1998). Brownprovides a thoughtful discussion of the “twin” problems thatplague the labs: disappearing missions and bad management.

36. Secretary of Energy Advisory Board, Task Force on AlternativeFutures for the Department of Energy National Laboratories,Alternative Futures for the Department of Energy National Laboratories(February 1995), p. 41.

37. The most prominent are the 1983 Federal Laboratory ReviewPanel, chaired by David Packard, and the task force chaired by WilliamGalvin, which produced the 1995 report Alternative Futures for theDepartment of Energy Laboratories.

38. The IBM case study (see Case Studies) indicates that IBM sci-entists must be tuned in to research occurring outside of the com-pany, even though IBM performs a great deal of basic research in-house.

39. See the CED policy statement, Connecting Students to a ChangingWorld: A Technology Strategy for Improving Mathematics and ScienceEducation (1995).

40. National Assessment Governing Board, What Do StudentsKnow? 1996 NAEP Science Results for 4th, 8th, & 12th Graders(Washington, D.C.: NAGB, 1997).

41. Results reported for high school seniors. Ina V.S. Mullis et al.,Mathematics and Science Achievement in the Final Year of SecondarySchool: Third International Mathematics and Science Study (ChestnutHill, Mass.: Boston College, 1998).

42. National Science Board, Science and Engineering Indicators,1996.

43. Committee on Science, Engineering, and Public Policy, Reshapingthe Graduate Education of Scientists and Engineers, (Washington,D.C.: National Academy Press, 1995).

44. Between 1973 and 1993, the number of full-time faculty appoint-ments for doctoral scientists and engineers grew by 67 percent. Duringthe same period, postdoctoral appointments rose by 223 percentand part-time faculty appointments increased 95 percent.


83

CHAPTER 4

46. National Science Board, Science & Engineering Indicators—1993,(Washington, DC: U.S. Government Printing Office), 1993, p. 139.

47. Walter Powell and Jason Owen-Smith, “Universities and theMarket for Intellectual Property in the Life Sciences,” Journal of PolicyAnalysis and Management, Vol 17 No (2), p. 268.

48. As we argue later in this chapter, some projects prove to be tooexpensive to support at the national level and should be pursuedin conjunction with other countries.

49. The President’s 1999 budget proposal would increase the fund-ing disparity. Of the proposed $1.2 billion increase in basic researchfunding between 1998 and 1999, $616 million (just over half)would go to the NIH basic research budget. This would raiseNIH’s share of the federal basic research budget slightly, from 46percent to 47 percent. See AAAS Analysis of R&D in the FY 1999 Budget,available at www.aaas.org on the Internet.

50. As described in Appendix 1, Medicare and Medicaid havebeen important sources of funding for research at university med-ical centers and, indirectly, a significant source of funding foruniversity research in general. Structural reforms in these programsmay result in less research funding from these sources.

51. See Roger Noll and William Rogerson, “The Economics ofUniversity Indirect Cost Reimbursement in Federal ResearchGrants,” in Challenges to Research Universities, ed. Roger Noll,(Washington, D.C.: Brookings Press, 1998). Noll and Rogersonprovide an analysis of the cost reimbursement problem and jus-tification for a “benchmarking” alternative to the current system.

52. National Academy of Sciences, Allocating Federal Funds forScience and Technology, Committee on Criteria for Federal Supportof Research and Development, (Washington, D.C.: NationalAcademy Press, 1995), pp. 25-26.

53. National Science Board, Science & Engineering Indicators—1996(Washington, D.C.: U.S. Government Printing Office, 1996), NSB96-21, pp.4-28.

54. CED has recently proposed remedies for the Social Security com-ponent of entitlements. See Fixing Social Security (March 1997).

55. See Kenneth M. Brown, Downsizing Science: Will the UnitedStates Pay the Price?, (Washington, D.C.: The AEI Press, 1998).

56. Data are for 1995. NSF, NSF Data Brief, 1997 no. 3 (March 13,1997); U.S. Department of Labor, Monthly Labor Review (January 1998),Table 43.

57. See the CED policy statements Putting Learning First: Governingand Managing the Schools for High Achievement (1994), ConnectingStudents to a Changing World: A Technology Strategy for ImprovingMathematics and Science Education (1995), and The Employer’s Rolein Linking School and Work (1998).

58. Some have suggested that agencies that sponsor research (suchas the NSF, NIH, and the Department of Defense) could providea significant boost to mentoring, internship, and other educa-tional outreach initiatives by providing additional resources to research institutions for these purposes.

59. For a recent discussion of this issue, see Committee on Science,Engineering, and Public Policy, Reshaping the Graduate Education

of Scientists and Engineers (Washington, D.C.: National AcademyPress, 1995).

60. Committee on Science, Engineering and Public Policy, Reshapingthe Graduate Education of Scientists and Engineers. See also Committeeon Science, Engineering and Public Policy, Advisor, Teacher, Role Model,Friend: On Being a Mentor to Students in Science and Engineering,(Washington, D.C.: National Academy Press, l997).

61. Equally discouraging for new doctorates and those consider-ing doctorates are certain aspects of today’s academic researchenvironment. As described in Chapter 3, the administrative and grant-raising burden placed on university researchers has become a for-midable disincentive to pursuing academic research careers. See“Problems and Solutions in the Administration of Federal Grantsfor University Research,” page 36, for recommendations in this area.

62. Immigrant scientists have long been very important to Americanbasic research, particularly during and following World War II.Between 1933 and 1970, 27 of the 77 Nobel Prizes awarded to theUnited States went to first-generation Americans (Eric Hobsbawm,The Age of Extremes, New York: Pantheon Books, 1993).

63. National Science Board, Science & Engineering Indicators—1996(Washington, D.C.: U.S. Government Printing Office, 1996), NSB96-21, Appendix Table 2-35.

APPENDIX 1

1. See Bureau of Economic Analysis, U.S. Department of Commerce,“A Satellite Acount for Research and Development,” Survey ofCurrent Business, November 1994.

2. See, Fixing Social Security (1997).

3. See Linda Cohen, “Soft Money, Hard Choices: ResearchUniversities and University Hospitals,” in Challenges to ResearchUniversities, ed. Roger Noll (Washington, D.C.: Brookings InstitutionPress, 1998)

CASE STUDIES

4. Merck had partnered with Chiron and with zymogenetics in thedevelopment of Recombivax HB.

5. Science, 249, 932-35.

6. Eventually, Merck discovered a non-nucleoside RTI that was licensedto the DuPont Merck Pharmaceutical Company in 1994 for clini-cal development and certain marketing rights. This compound,efavirenz [Sustiva (DuPont Merck); Stocrin (Merck)], can signifi-cantly reduce viral load and enhance immune system function inpatients with HIV infection when used in combination withCrixivan and other HIV anitviral drugs.

7. Nancy E. Kohl et al., “Active Human Immunodeficiency VirusProtease Is Required for Viral Infectivity,” Proceedings of the NationalAcademy of Science 85 (July 1988): 4686-4690.

8. Manuel A. Navia et al., “Three-dimensional Structure of AspartylProtease from Human Immunodeficiency Virus HIV-1,” Nature 337,no. 6208 (February 16, 1989): 615-620.

9. Jon Condra et al., “Genetic Correlates of Viral Resistance to theHuman Immunodeficiency Virus Type 1 Protease InhibitorIndinavir,” Journal of Virology 70 (1996): 8270-8276.

84

10. See Scott M. Hammer et al., “A Controlled Trial of TwoNucleoside Analogues Plus Indinavir in Persons with HumanImmunodeficiency Virus Infection and CD4 Cell Counts of 200 PerCubic Millimeter or Less,” and Roy M. Gulick et al., “Treatmentwith Indinavir, Zidovudine, and Lamivudine in Adults withHuman Immunodeficiency Virus Infection and Prior AntiretroviralTherapy,” both in The New England Journal of Medicine 337(September 11, 1997): 725-733 and 734-739, respectively.

11. For a discussion of the chemistry, see Paul J. Reider, “Advancesin AIDS Chemotherapy: The Asymmetric Synthesis of Crixivan,”Chimia 51 (1997): 306-308.

12. F.E. Heart, R.E. Kahn, S.M. Ornstein, W.R. Crowther, and D.C.Walden, “The Interface Message Processor for the ARPA ComputerNetwork,” Proceedings of AFIPS 1970 Spring Joint ComputerConference, Vol. 36, June 1970.

13. R.H. Thomas and D.A. Henderson, “McRoss- A Multi-computerProgramming System,” Proceedings of AFIPS 1972 Spring JointComputer Conference, Vol. 40, June 1972.

14. R.E. Schantz, “Operating System Design for a Network Computer,”

Ph.D. Thesis, submitted to State University of New York at StonyBrook; Computer Science Technical Report #28, May 1974.

15. R.E. Schantz, R.H. Thomas and G. Bono, “The Architecture ofthe Cronus Distributed Operating System,” Proceedings of the 6thInternational Conference on Distributed Computing Systems,Cambridge, Massachusetts, May 1986.

16. B.M. Anderson and J.P. Flynn, “CASES: A System for AssessingNaval Warfighting Capability,” Proceedings of the 1990 Symposiumon Command and Control Research, SAIC Report 90/1508, June1990.

17. J.C. Berets and M.A. Dean, “Building Object Oriented DistributedApplications Using Cronus,” Proceedings of the IEEE Dual UseTechnologies and Application Conference, Utica, New York, May1994.

18. J.R. Nicol, C.T Wilkes, R.D. Edmiston, J.C. Fitzgerald, “Experienceswith Accommodating Heterogeneity in a Large ScaleTelecommunications Infrastructure,” Proceedings of the 3rdSymposium on Experiences with Distributed and MultiprocessorSystems, Newport Beach, CA, March 1992.

85

For more than 50 years, the Committee forEconomic Development has been a respectedinfluence on the formation of business and pub-lic policy. CED is devoted to these two objectives:

To develop, through objective research and informeddiscussion, findings and recommendations for privateand public policy that will contribute to preservingand strengthening our free society, achieving steadyeconomic growth at high employment and reasonablystable prices, increasing productivity and living stan-dards, providing greater and more equal opportuni-ty for every citizen, and improving the quality oflife for all.

To bring about increasing understanding by pre-sent and future leaders in business, government,and education, and among concerned citizens, of theimportance of these objectives and the ways in whichthey can be achieved.

CED’s work is supported by private voluntarycontributions from business and industry, foun-

dations, and individuals. It is independent, non-profit, nonpartisan, and nonpolitical.

Through this business-academic partnership,CED endeavors to develop policy statementsand other research materials that commendthemselves as guides to public and businesspolicy; that can be used as texts in college eco-nomics and political science courses and in man-agement training courses; that will be consideredand discussed by newspaper and magazine edi-tors, columnists, and commentators; and thatare distributed abroad to promote better under-standing of the American economic system.

CED believes that by enabling business lead-ers to demonstrate constructively their concernfor the general welfare, it is helping business toearn and maintain the national and communityrespect essential to the successful functioning ofthe free enterprise capitalist system.

OBJECTIVES OF THE COMMITTEE FORECONOMIC DEVELOPMENT

86

CED BOARD OF TRUSTEES

Chairman

FRANK P. DOYLE, Retired Executive Vice PresidentGE

Vice Chairmen

PHILIP J. CARROLL, President and Chief Executive OfficerShell Oil CompanyRAYMOND V. GILMARTIN, Chairman, President and Chief

Executive OfficerMerck & Co., Inc.MATINA S. HORNERExecutive Vice PresidentTIAA-CREFHENRY A. MCKINNELLExecutive Vice PresidentPfizer Inc.

Treasurer

JOHN B. CAVEPrincipalAvenir Group, Inc.

REX D. ADAMS Dean, The Fuqua School of BusinessDuke UniversityPAUL A. ALLAIREChairman and Chief Executive OfficerXerox CorporationIAN ARNOF President and Chief Executive OfficerFirst Commerce CorporationIRVING W. BAILEY, IIVice Chairman AEGON Insurance GroupBERNARD B. BEALChief Executive OfficerM.R. Beal & Co.HANS W. BECHERER Chairman and Chief Executive OfficerDeere & CompanyHENRY P. BECTON, JR.President and General ManagerWGBH Educational FoundationALAN BELZERRetired President and Chief Operating OfficerAlliedSignal Inc.PETER A. BENOLIELChairman, Executive CommitteeQuaker Chemical CorporationMELVYN E. BERGSTEINChairman and Chief Executive OfficerDiamond Technology PartnersC. ROBERT BLACKSenior Vice PresidentTexaco Inc.

ROY J. BOSTOCK Chairman and Chief Executive OfficerThe MacManus GroupC. L. BOWERMAN Executive Vice President, Planning and Corporate Relations &

ServicesPhillips Petroleum CompanyRICHARD J. BOYLE Retired Vice ChairmanChase Manhattan Bank, N.A.JOHN BRADEMAS President EmeritusNew York UniversityWILLIAM E. BROCK ChairmanThe Brock Group, LTDSTEPHEN L. BROWN Chairman and Chief Executive OfficerJohn Hancock Mutual Life Insurance CompanyGORDON F. BRUNNER Senior Vice President and DirectorThe Procter & Gamble CompanyJOHN H. BRYAN Chairman and Chief Executive OfficerSara Lee CorporationMICHAEL BUNGEYChairman and Chief Executive OfficerBates Worldwide, Inc.J. GARY BURKHEADVice ChairmanFMR CorporationJEAN B. BUTTNERChairman and Chief Executive OfficerValue Line Inc.*FLETCHER L. BYROMPresident and Chief Executive OfficerMICASU CorporationDONALD R. CALDWELLPresident and Chief Operating OfficerSafeguard Scientifics, Inc.FRANK J. CARLUCCIChairmanThe Carlyle GroupPHILIP J. CARROLLPresident and Chief Executive OfficerShell Oil CompanyROBERT B. CATELLChairman, President and Chief Executive OfficerKeySpan Energy CorporationJOHN B. CAVEPrincipalAvenir Group, Inc.JOHN S. CHALSTYChairmanDonaldson, Lufkin & Jenrette, Inc.

*Life Trustee

87

RAYMOND G. CHAMBERS Chairman of the BoardAmelior FoundationMARY ANN CHAMPLIN Retired Senior Vice PresidentAetna Inc.CAROLYN CHIN Executive Vice PresidentReuters AmericaA. W. CLAUSENRetired Chairman and Chief Executive OfficerBankAmerica Corporation*JOHN L. CLENDENINRetired ChairmanBellSouth CorporationNANCY S. COLE PresidentEducational Testing ServiceFERDINAND COLLOREDO-MANSFELD Chairman and Chief Executive OfficerCabot Industrial TrustGEORGE H. CONRADES Executive Vice PresidentGTE Corporation and President, GTE InternetworkingKATHLEEN COOPER Chief EconomistExxon CorporationGARY L. COUNTRYMAN Chairman of the BoardLiberty Mutual Insurance CompanySTEPHEN A. CRANE President and Chief Executive OfficerGryphon Holdings, Inc.RONALD R. DAVENPORT Chairman of the BoardSheridan Broadcasting CorporationALFRED C. DECRANE, JRRetired Chairman and Chief Executive OfficerTexaco Inc.JOHN DIEBOLD ChairmanJohn Diebold IncorporatedJOHN T. DILLON Chairman and Chief Executive OfficerInternational Paper CompanyREGINA DOLAN Vice President and Chief Administrative OfficerPaineWebber Group Inc.HERBERT P. DOOSKINChairmanAtlantic Health SystemE. LINN DRAPER, JR.Chairman, President and Chief Executive OfficerAmerican Electric Power CompanyT. J. DERMOT DUNPHYChairman and Chief Executive OfficerSealed Air CorporationCHRISTOPHER D. EARLSenior Vice PresidentPerseus Capital LLC

W. D. EBERLE ChairmanManchester Associates, Ltd.WILLIAM S. EDGERLYChairmanFoundation for PartnershipsWALTER Y. ELISHARetired Chairman and ChiefExecutive OfficerSprings Industries, Inc.JAMES D. ERICSON President and Chief Executive OfficerThe Northwestern Mutual Life Insurance CompanyWILLIAM T. ESREYChairman and Chief Executive OfficerSprintPATRICIA O’DONNELL EWERSPresidentPace UniversityKATHLEEN FELDSTEIN PresidentEconomic Studies, Inc.RONALD E. FERGUSON Chairman, President and Chief Executive OfficerGeneral RE CorporationE. JAMES FERLAND Chairman, President and Chief Executive OfficerPublic Service Enterprise Group Inc.*EDMUND B. FITZGERALDManaging DirectorWoodmont AssociatesHARRY L. FREEMAN PresidentThe Freeman CompanyJOHN W. GALIARDO Vice Chairman and General CounselBecton Dickinson and CompanyJOSEPH GANTZ Harrison, New YorkTHOMAS P. GERRITYDeanThe Wharton SchoolUniversity of PennsylvaniaRAYMOND V. GILMARTINChairman, President and Chief Executive OfficerMerck & Co., Inc.SUE LING GIN President and Chief Executive OfficerFlying Food Fare, Inc.BOYD E. GIVANSenior Vice President and Chief Financial OfficerThe Boeing CompanyFREDERICK W. GLUCK Vice Chairman and DirectorBechtel Group, Inc.CAROL R. GOLDBERG PresidentThe Avcar Group, Ltd.ELLEN R. GORDONPresident and Chief Operating OfficerTootsie Roll Industries, Inc.

*Life Trustee

88

JOSEPH T. GORMAN Chairman and Chief Executive OfficerTRW Inc.RICHARD A. GRASSO Chairman and Chief Executive OfficerNew York Stock Exchange, Inc.EARL G. GRAVES, SR.Publisher and Chief Executive OfficerBlack Enterprise Magazine

WILLIAM H. GRAY, IIIPresident and Chief Executive OfficerUnited Negro College FundROSEMARIE B. GRECO CoreStates BankGERALD GREENWALD Chairman and Chief Executive OfficerUAL CorporationBARBARA B. GROGAN PresidentWestern Industrial ContractorsPATRICK W. GROSSFounder and Chairman, Executive CommitteeAmerican Management Systems, Inc.

JUDITH H. HAMILTON President and Chief Executive OfficerFirst Floor SoftwareRICHARD W. HANSELMAN Retired ChairmanGenesco Inc.WILLIAM F. HECHT Chairman, President and Chief Executive OfficerPP&L Resources, Inc.NOAH T. HERNDON PartnerBrown Brothers Harriman & Co.RODERICK M. HILLS PresidentHills Enterprises, Ltd.HAYNE HIPPPresident and Chief Executive OfficerThe Liberty CorporationMATINA S. HORNERExecutive Vice PresidentTIAA-CREFAMOS B. HOSTETTER ChairmanPilot House Associates, LLCWILLIAM R. HOWELLFormer Chairman of the BoardJ. C. Penney Company, Inc.ROBERT J. HURST Vice ChairmanGoldman, Sachs & Co.ALICE STONE ILCHMAN PresidentSarah Lawrence CollegeGEORGE B. JAMES Senior Vice President and Chief Financial Officer Levi Strauss & Co.

JAMES A. JOHNSON Chairman and Chief Executive OfficerFannie MaeROBBIN S. JOHNSON Corporate Vice President, Public AffairsCargill, IncorporatedPRES KABACOFF President and Co-ChairmanHistoric Restoration, Inc.JOSEPH J. KAMINSKI Corporate Executive Vice PresidentAir Products and Chemicals, Inc.EDWARD A. KANGAS Chairman and Chief ExecutiveDeloitte Touche Tohmatsu InternationalJOSEPH E. KASPUTYS Chairman, President and Chief Executive OfficerPrimark CorporationEAMON M. KELLYPresidentTulane UniversityJAMES P. KELLYChairman and Chief Executive OfficerUnited Parcel Service of America, Inc.THOMAS J. KLUTZNICKPresidentThomas J. Klutznick CompanyCHARLES F. KNIGHT Chairman and Chief Executive OfficerEmerson Electric Co.CHARLES E.M. KOLBPresidentCommittee for Economic DevelopmentALLEN J. KROWE Retired Vice ChairmanTexaco Inc.RICHARD J. KRUIZENGAPresidentISEM, Southern Methodist UniversityC. JOSEPH LABONTEThe Vantage GroupCHARLES R. LEE Chairman and Chief Executive OfficerGTE CorporationA.V. LIVENTALS Vice President, Strategic PlanningMobil CorporationEDWIN A. LUPBERGER Chairman and Chief Executive OfficerEntergy CorporationJOSEPH T. LYNAUGH President and Chief Executive OfficerNYLCare Health Plans, Inc.BRUCE K. MACLAURYPresident EmeritusThe Brookings InstitutionCOLETTE MAHONEY RSHMPresident EmeritusMarymount Manhattan College

*Life Trustee

89

ALONZO L. MCDONALD Chairman and Chief Executive OfficerAvenir Group, Inc.EUGENE R. MCGRATH Chairman, President and Chief Executive OfficerConsolidated Edison Company of New YorkHENRY A. MCKINNELLExecutive Vice PresidentPfizer Inc.DAVID E. MCKINNEYThomas J. Watson FoundationDEBORAH HICKS MIDANEK PrincipalJay Alix & AssociatesNICHOLAS G. MOORE ChairmanCoopers & Lybrand L.L.P.DIANA S. NATALICIO PresidentThe University of Texas at El PasoMARILYN CARLSON NELSON Vice Chair and Chief Operating OfficerCarlson Holdings, Inc.JOSEPH NEUBAUER Chairman and Chief Executive OfficerARAMARK CorporationBARBARA W. NEWELLRegents ProfessorFlorida State UniversityJAMES J. O’CONNOR Retired Chairman and Chief Executive OfficerUnicom CorporationDEAN R. O’HARE Chairman and Chief Executive OfficerChubb CorporationJAMES J. O’NEILLChief Executive OfficerOnex Food Services, Inc.JOHN D. ONG Chairman EmeritusThe BF Goodrich CompanyANTHONY J.F. O’REILLYChairman and Chief Executive OfficerH.J. Heinz CompanyJAMES F. ORR IIIChairman and Chief Executive OfficerUNUM CorporationROBERT J. O’TOOLE Chairman and Chief Executive OfficerA.O. Smith CorporationSTEFFEN E. PALKO Vice Chairman and PresidentCross Timbers Oil CompanyROBERT B. PALMER Chairman, President and Chief Executive OfficerDigital Equipment CorporationVICTOR A. PELSON Senior AdvisorSBC Warburg Dillon Read Inc.

PETER G. PETERSON ChairmanThe Blackstone GroupDEAN P. PHYPERS New Canaan, ConnecticutARNOLD B. POLLARDPresident and Chief Executive OfficerThe Chief Executive GroupS. LAWRENCE PRENDERGASTChairman and Chief Executive OfficerAT&T Investment Management CorporationHUGH B. PRICE President and Chief Executive OfficerNational Urban LeagueROBERT A. PRITZKER President and Chief Executive OfficerThe Marmon GroupWESLEY D. RATCLIFF President and Chief Executive OfficerAdvanced Technological Solutions, Inc.NED V. REGAN The Jerome Levy Economics InstituteWILLIAM R. RHODES Vice ChairmanCiticorp/Citibank, N.A.JAMES Q. RIORDAN Stuart, FloridaE. B. ROBINSON, JR.ChairmanDeposit Guaranty National BankIAN M. ROLLAND Chairman and Chief Executive OfficerLincoln National CorporationDANIEL ROSEPresidentRose Associates, Inc.HOWARD M. ROSENKRANTZ President and Chief Operating OfficerUnited States Surgical CorporationJOHN A. ROTHPresident and Chief Executive OfficerNorthern Telecom LimitedMICHAEL I. ROTH Chairman and Chief Executive OfficerThe Mutual Life Insurance Company of New YorkLANDON H. ROWLAND President and Chief Executive OfficerKansas City Southern Industries, Inc.NEIL L. RUDENSTINE PresidentHarvard UniversityGEORGE E. RUPPPresidentColumbia UniversityGEORGE F. RUSSELL, JR.ChairmanFrank Russell CompanyARTHUR F. RYAN Chairman and Chief Executive OfficerThe Prudential Insurance Company of America

*Life Trustee

90

STEPHEN W. SANGER Chairman and Chief Executive OfficerGeneral Mills, Inc.JOHN C. SAWHILLPresident and Chief Executive OfficerThe Nature ConservancyHENRY B. SCHACHT Director & Senior AdvisorLucent Technologies Inc.JONATHAN M. SCHOFIELD Chairman and Chief Executive OfficerAirbus Industrie of North America, Inc.DONALD J. SCHUENKE ChairmanNorthern Telecom LimitedWALTER V. SHIPLEYChairman and Chief Executive OfficerThe Chase Manhattan CorporationWALTER H. SHORENSTEIN Chairman of the BoardThe Shorenstein Company*GEORGE P. SHULTZDistinguished FellowThe Hoover InstitutionStanford UniversityROCCO C. SICILIANO Beverly Hills, CaliforniaRUTH SIMMONS PresidentSmith CollegeFREDERICK W. SMITH Chairman, President and Chief Executive OfficerFederal Express CorporationRAYMOND W. SMITH ChairmanBell Atlantic CorporationTIMOTHY P. SMUCKER ChairmanThe J.M. Smucker CompanyHUGO FREUND SONNENSCHEIN PresidentThe University of ChicagoALAN G. SPOON PresidentThe Washington Post CompanyJOHN R. STAFFORD Chairman, President and Chief Executive OfficerAmerican Home Products CorporationSTEPHEN STAMAS ChairmanThe American AssemblyJOHN L. STEFFENS Vice ChairmanMerrill Lynch & Co., Inc.PAULA STERN PresidentThe Stern Group, Inc.DONALD M. STEWART PresidentThe College Board

ROGER W. STONE Chairman, President and Chief Executive OfficerStone Container CorporationMATTHEW J. STOVER Group President, Information Services GroupBell Atlantic CorporationJAMES N. SULLIVAN Vice Chairman of the BoardChevron CorporationRICHARD J. SWIFT Chairman, President and Chief Executive OfficerFoster Wheeler CorporationRICHARD F. SYRON Chairman and Chief Executive OfficerAmerican Stock ExchangeALISON TAUNTON-RIGBYPresident and Chief Executive OfficerAquila Biopharmaceuticals, Inc.RICHARD L. THOMAS Retired ChairmanFirst Chicago NBD CorporationJAMES A. THOMSON President and Chief Executive OfficerRANDCHANG-LIN TIEN NEC Distinguished Professor of EngineeringUniversity of California, BerkeleyTHOMAS J. TIERNEYWorldwide Managing DirectorBain & CompanyALAIR A. TOWNSEND PublisherCrain’s New York Business

ALEXANDER J. TROTMAN Chairman, President and Chief Executive OfficerThe Ford Motor CompanyDONALD C. WAITE, IIIManaging DirectorMcKinsey & Company, Inc.ARNOLD R. WEBER ChancellorNorthwestern UniversityLAWRENCE A. WEINBACH Chairman and Chief Executive OfficerUnisys CorporationROBERT E. WEISSMAN Chairman and Chief Executive OfficerCognizant CorporationJOHN F. WELCH Chairman and Chief Executive OfficerGEJOSH S. WESTON Honorary ChairmanAutomatic Data Processing, Inc.CLIFTON R. WHARTON, JR. Former Chairman and Chief Executive OfficerTIAA-CREFDOLORES D. WHARTON Chairman and Chief Executive OfficerThe Fund for Corporate Initiatives, Inc.

*Life Trustee

91

MICHAEL W. WICKHAMChairman and Chief Executive OfficerRoadway Express, Inc.HAROLD M. WILLIAMS Retired PresidentThe J. Paul Getty TrustJ. KELLEY WILLIAMS Chairman and Chief Executive OfficerChemFirst Inc.LINDA SMITH WILSON PresidentRadcliffe College

MARGARET S. WILSON Chairman of the BoardScarbroughsKURT E. YEAGERPresident and Chief Executive OfficerElectric Power Research InstituteMARTIN B. ZIMMERMAN Executive Director, Governmental Relations and

Corporate EconomicsFord Motor Company

*Life Trustee

92

RAY C. ADAM, Retired ChairmanNL IndustriesO. KELLEY ANDERSONBoston, MassachusettsROBERT O. ANDERSON, Retired Chairman Hondo Oil & Gas CompanyROY L. ASHLos Angeles, CaliforniaSANFORD S. ATWOOD, President EmeritusEmory UniversityROBERT H.B. BALDWIN Retired ChairmanMorgan Stanley Group, Inc.GEORGE F. BENNETT, Chairman Emeritus State Street Investment TrustHAROLD H. BENNETT Salt Lake City, UtahJACK F. BENNETT, Retired Senior Vice PresidentExxon CorporationHOWARD W. BLAUVELT Keswick, VirginiaMARVIN BOWER Delray Beach, FloridaALAN S. BOYD Washington, D.C.ANDREW F. BRIMMER, PresidentBrimmer & Company, Inc.HARRY G. BUBB, Chairman EmeritusPacific Mutual Life InsuranceTHEODORE A. BURTIS, Retired Chairman of the Board Sun Company, Inc.PHILIP CALDWELL, Chairman (Retired)The Ford Motor CompanyEVERETT N. CASE Van Hornesville, New YorkHUGH M. CHAPMAN, Retired ChairmanNationsBank SouthE. H. CLARK, JR., Chairman and Chief Executive OfficerThe Friendship GroupDOUGLAS D. DANFORTH, Retired ChairmanWestinghouse Electric CorporationJOHN H. DANIELS, Retired Chairman and Chief Executive

OfficerArcher-Daniels Midland Co.RALPH P. DAVIDSON Washington, D.C.ARCHIE K. DAVIS, Chairman of the Board (Retired)Wachovia Bank and Trust Company, N.A.ROBERT R. DOCKSON, Chairman Emeritus CalFed, Inc.LYLE EVERINGHAM, Retired ChairmanThe Kroger Co.THOMAS J. EYERMAN, President Delphi Associates Limited

JOHN T. FEYPark City, UtahJOHN M. FOX Sapphire, North CarolinaDON C. FRISBEE, Chairman EmeritusPacifiCorpRICHARD L. GELB, Chairman Emeritus Bristol-Myers Squibb CompanyW. H. KROME GEORGE, Retired Chairman Aluminum Company of AmericaWALTER B. GERKEN, Chairman, Executive Committee Pacific Mutual Life Insurance CompanyPAUL S. GEROT Delray Beach, FloridaLINCOLN GORDON, Guest Scholar The Brookings InstitutionKATHARINE GRAHAM, Chairman of the Executive

Committee The Washington Post CompanyJOHN D. GRAY, Chairman Emeritus Hartmarx CorporationJOHN R. HALL, Retired Chairman Ashland Inc.ROBERT A. HANSON, Retired Chairman Deere & CompanyROBERT S. HATFIELD, Retired Chairman The Continental Group, Inc.ARTHUR HAUSPURG, Member, Board of Trustees Consolidated Edison Company of New YorkPHILIP M. HAWLEY, Retired Chairman of the BoardCarter Hawley Hale Stores, Inc.WILLIAM A. HEWITT Rutherford, CaliforniaROBERT C. HOLLAND, Senior FellowThe Wharton SchoolUniversity of PennsylvaniaLEON C. HOLT, Retired Vice Chairman Air Products and Chemicals, Inc.SOL HURWITZ, Former President Committee for Economic DevelopmentGEORGE F. JAMES Ponte Vedra Beach, FloridaHENRY R. JOHNSTONPonte Vedra Beach, FloridaGEORGE M. KELLER, Chairman of the Board, Retired Chevron CorporationJAMES R. KENNEDYManalapan, FloridaPHILIP M. KLUTZNICK, Senior PartnerKlutznick InvestmentsFRANKLIN A. LINDSAY, Retired Chairman Itek CorporationROBERT W. LUNDEEN, Retired Chairman The Dow Chemical Company

CED HONORARY TRUSTEES

93

RAY W. MACDONALD, Honorary Chairman of the Board Burroughs CorporationIAN MACGREGOR, Retired ChairmanAMAX Inc.RICHARD B. MADDEN, Retired Chairman and Chief

Executive OfficerPotlatch CorporationSTANLEY MARCUS, ConsultantStanley Marcus ConsultancyAUGUSTINE R. MARUSI Lake Wales, FloridaWILLIAM F. MAY, ChairmanStatue of Liberty - Ellis Island Foundation, Inc.OSCAR G. MAYER, Retired ChairmanOscar Mayer & Co.GEORGE C. MCGHEE Former U.S. Ambassador and Under Secretary of StateJOHN F. MCGILLICUDDY, Retired Chairman and Chief

Executive OfficerChemical Banking CorporationJAMES W. MCKEE, JR., Retired ChairmanCPC International, Inc.CHAMPNEY A. MCNAIR, Retired Vice Chairman Trust Company of GeorgiaJ. W. MCSWINEY, Retired Chairman of the Board The Mead CorporationROBERT E. MERCER, Retired ChairmanThe Goodyear Tire & Rubber Co.RUBEN F. METTLER, Retired Chairman and Chief Executive

OfficerTRW Inc.LEE L. MORGAN, Former Chairman of the Board Caterpillar, Inc.ROBERT R. NATHAN, Chairman Nathan Associates, Inc.ALFRED C. NEALHarrison, New YorkJ. WILSON NEWMAN, Retired Chairman Dun & Bradstreet CorporationLEIF H. OLSEN, President Leif H. Olsen Investments, Inc.NORMA PACE, New York, New YorkCHARLES W. PARRY, Retired ChairmanAluminum Company of AmericaWILLIAM R. PEARCE, President and Chief Executive Officer IDS Mutual Fund GroupJOHN H. PERKINS, Former President Continental Illinois National Bank and Trust CompanyRUDOLPH A. PETERSON, President and Chief Executive

Officer (Emeritus) BankAmerica CorporationEDMUND T. PRATT, JR., Retired Chairman and Chief

Executive OfficerPfizer Inc.ROBERT M. PRICE, Retired Chairman and Chief Executive

OfficerControl Data Corporation

R. STEWART RAUCH, Former Chairman The Philadelphia Savings Fund SocietyAXEL G. ROSIN, Retired ChairmanBook-of-the-Month Club, Inc.WILLIAM M. ROTH Princeton, New JerseyGEORGE RUSSELLBloomfield, MichiganJOHN SAGAN, PresidentJohn Sagan AssociatesRALPH S. SAUL, Former Chairman of the Board CIGNA CompaniesGEORGE A. SCHAEFER, Retired Chairman of the Board Caterpillar Inc.ROBERT G. SCHWARTZ New York, New YorkMARK SHEPHERD, JR., Retired Chairman Texas Instruments, Inc.RICHARD R. SHINN, Retired Chairman and Chief Executive

OfficerMetropolitan Life Insurance CompanyDAVIDSON SOMMERS Washington, D.C.ELMER B. STAATS, Former Controller General of the

United States

HONORABLE ELVIS J. STAHR, JR.Chickering & Gregory, P.C.FRANK STANTON, Former President CBS, Inc.EDGAR B. STERN, JR., Chairman of the Board Royal Street CorporationALEXANDER L. STOTT Fairfield, ConnecticutWAYNE E. THOMPSON, Past Chairman Merritt Peralta Medical CenterCHARLES C. TILLINGHAST, JR.Providence, Rhode IslandHOWARD S. TURNER, Retired Chairman Turner Construction Company L. S. TURNER, JR.Dallas, TexasTHOMAS A. VANDERSLICE TAV AssociatesJAMES E. WEBBWashington, D.C.SIDNEY J. WEINBERG, JR., Limited Partner The Goldman Sachs Group, L.P.ROBERT C. WINTERS, Chairman Emeritus Prudential Insurance Company of AmericaARTHUR M. WOODChicago, IllinoisRICHARD D. WOOD, Director Eli Lilly and CompanyWILLIAM S. WOODSIDE, Chairman LSG Sky ChefsCHARLES J. ZWICK Coral Gables, Florida

94

Chairman

ISABEL V. SAWHILLSenior FellowThe Brookings Institution

GEORGE A. AKERLOFSenior FellowThe Brookings InstitutionProfessor, Department of EconomicsUniversity of California, BerkeleyELIZABETH E. BAILEYJohn C. Hower Professor of Public

Policy and ManagementThe Wharton SchoolUniversity of PennsylvaniaALAN S. BLINDERGordon S. Rentschler Memorial

Professor of EconomicsPrinceton University

JOHN COGANSenior FellowHoover InstitutionStanford UniversityRUDI DORNBUSCHFord Professor of Economics and

International ManagementMassachusetts Institute of TechnologyRONALD F. FERGUSONProfessorJohn F. Kennedy School of GovernmentHarvard UniversityLINDA YUEN-CHING LIMProfessorUniversity of Michigan Business SchoolROBERT D. REISCHAUERSenior FellowThe Brookings Institution

CHRISTINA D. ROMERProfessor, Department of EconomicsUniversity of California, BerkeleyJOHN B. SHOVENDean, School of Humanities and

SciencesStanford UniversityMURRAY WEIDENBAUMChairman, Center for the Study of

American BusinessWashington UniversityDAVID WESSELChief Economics CorrespondentThe Wall Street Journal

SIDNEY G. WINTERDeloitte and Touche Professor of

ManagementThe Wharton SchoolUniversity of Pennsylvania

CED RESEARCH ADVISORY BOARD

95

CHARLES E.M. KOLBPresident

CED PROFESSIONAL AND ADMINISTRATIVE STAFF

VAN DOORN OOMSSenior Vice President and Director of

ResearchWILLIAM J. BEEMANVice President and Director

of Economic StudiesCLAUDIA P. FEUREYVice President for Communications

and Corporate Affairs

THOMAS R. FLAHERTYComptroller and Director of OperationsSANDRA KESSLER HAMBURGVice President and Director of

Education and Special ProjectsTIMOTHY J. MUENCHVice President and Director of Finance

and AdministrationMICHAEL J. PETROVice President and Director of Business

and Government Policy

EVA POPPERVice President, Director of

Development, and Secretary of theBoard of Trustees

Advisor on InternationalEconomic Policy

ISAIAH FRANKWilliam L. Clayton Professor of

International EconomicsThe Johns Hopkins University

Research

SCOTT MORRISSenior EconomistSUE SOMMERFIELDSecretary of the Research and Policy

CommitteeROBERT FLEEGLERResearch AssociateCHRIS DREIBELBISStaff Assistant

Special Projects

JESSICA B. ORKINAssistant Director,Special Projects and CommunicationsPETER DAVISStaff AssociateDEOKI PESTANOGrants Coordinator

Development

JAMES WRIGHTAssociate DirectorANA SOMOHANOCampaign CoordinatorWILFORD V. MALCOLMCampaign Production AdministratorKATHLEEN EDMONDSONContributions Secretary

Conferences

VALERIE MENDELSOHNManager

Publications

DARCY TUCKERCoordinator

Administration - New York Office

DOROTHY M. STEFANSKIDeputy ComptrollerKAREN CASTROAssistant ComptrollerARLENE M. MURPHYExecutive Assistant to the PresidentPETER E. COXOperations Assistant

Administration - Washington, D.C. Office

SHARON A. CLATTERBAUGHExecutive Assistant to the PresidentSHIRLEY R. SHERMANOffice ManagerAMANDA TURNERAssistant Office Manager

96

STATEMENTS ON NATIONAL POLICY ISSUED BY THE COMMITTEEFOR ECONOMIC DEVELOPMENT

SELECTED PUBLICATIONS:

Modernizing Government Regulation (1998)U.S. Economic Policy Toward The Asia-Pacific Region (1997)Connecting Inner-City Youth To The World of Work (1997)Fixing Social Security (1997)Growth With Opportunity (1997)American Workers and Economic Change (1996)Connecting Students to a Changing World: A Technology Strategy for Improving Mathematics and Science

Education (1995)Cut Spending First: Tax Cuts Should Be Deferred to Ensure a Balanced Budget (1995)Rebuilding Inner-City Communities: A New Approach to the Nation’s Urban Crisis (1995)Who Will Pay For Your Retirement? The Looming Crisis (1995)Putting Learning First: Governing and Managing the Schools for High Achievement (1994)Prescription for Progress: The Uruguay Round in the New Global Economy (1994)*From Promise to Progress: Towards a New Stage in U.S.-Japan Economic Relations (1994)U.S. Trade Policy Beyond The Uruguay Round (1994)In Our Best Interest: NAFTA and the New American Economy (1993)What Price Clean Air? A Market Approach to Energy and Environmental Policy (1993).Why Child Care Matters: Preparing Young Children For A More Productive America (1993)Restoring Prosperity: Budget Choices for Economic Growth (1992)The United States in the New Global Economy: A Rallier of Nations (1992)The Economy and National Defense: Adjusting to Cutbacks in the Post-Cold War Era (1991)Politics, Tax Cuts and the Peace Dividend (1991)The Unfinished Agenda: A New Vision for Child Development and Education (1991)Foreign Investment in the United States: What Does It Signal? (1990)An America That Works: The Life-Cycle Approach to a Competitive Work Force (1990)Breaking New Ground in U.S. Trade Policy (1990)Battling America’s Budget Deficits (1989)*Strengthening U.S.-Japan Economic Relations (1989)Who Should Be Liable? A Guide to Policy for Dealing with Risk (1989)Investing in America’s Future: Challenges and Opportunities for Public Sector Economic Policies (1988)Children in Need: Investment Strategies for the Educationally Disadvantaged (1987)Finance and Third World Economic Growth (1987)Reforming Health Care: Market Prescription (1987)Work and Change: Labor Market Adjustment Policies in a Competitive World (1987)Leadership for Dynamic State Economies (1986)Investing in Our Children: Business and the Public Schools (1985)Fighting Federal Deficits: The Time for Hard Choices (1985)Strategy for U.S. Industrial Competitiveness (1984)Productivity Policy: Key to the Nation’s Economic Future (1983)Energy Prices and Public Policy (1982)Public-Private Partnership: An Opportunity for Urban Communities (1982)Reforming Retirement Policies (1981)Transnational Corporations and Developing Countries: New Policies for a Changing World Economy

(1981)Stimulating Technological Progress (1980)

*Statements issued in association with CED counterpart organizations in foreign countries.

97

CED COUNTERPART ORGANIZATIONS

Close relations exist between the Committee for Economic Development and independent, nonpo-litical research organizations in other countries. Such counterpart groups are composed of businessexecutives and scholars and have objectives similar to those of CED, which they pursue by similar-ly objective methods. CED cooperates with these organizations on research and study projects of com-mon interest to the various countries concerned. This program has resulted in a number of joint policystatements involving such international matters as energy, East-West trade, assistance to develop-ing countries, and the reduction of nontariff barriers to trade.

CE Circulo de EmpresariosMadrid, Spain

CEDA Committee for Economic Development of AustraliaSydney, Australia

EVA Centre for Finnish Business and Policy StudiesHelsinki, Finland

FAE Forum de Administradores de EmpresasLisbon, Portugal

FDE Belgian Enterprise FoundationBrussels, Belgium

IDEP Institut de l’EntrepriseParis, France

IW Institut der Deutschen WirtschaftCologne, Germany

Keizai DoyukaiTokyo, Japan

SMO Stichting Maatschappij en OndernemingThe Netherlands

SNS Studieforbundet Naringsliv och SamhalleStockholm, Sweden

Committee for Economic Development

477 Madison AvenueNew York, New York 10022Telephone: (212) 688-2063

Fax: (212) 758-9068

2000 L Street, N.W., Suite 700Washington, D.C. 20036

Telephone: (202) 296-5860Fax: (202) 223-0776

http://www.ced.org

AMERICA’S BASIC RESEARCH · vi The Committee for Economic Development is an independent research...

Documents

Transcript of AMERICA’S BASIC RESEARCH · vi The Committee for Economic Development is an independent research...