Selected readings on research and development expenditures and the national economy


Material Information

Selected readings on research and development expenditures and the national economy
Series Title:
Serial no. 94-Z
Physical Description:
vi, 283 p. : ill. ; 24 cm.
United States -- Congress. -- House. -- Committee on Science and Technology. -- Subcommittee on Domestic and International Scientific Planning and Analysis
U.S. Govt. Print. Off.
Place of Publication:
Publication Date:


Subjects / Keywords:
Research and development contracts -- United States   ( lcsh )
Economic conditions -- United States   ( lcsh )
federal government publication   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographies.
Statement of Responsibility:
prepared by the Subcommittee on Domestic and International Scientific Planning and Analysis of the Committee on Science and Technology, U.S. House of Representatives, Ninety-fourth Congress, second sesson ... April 1976.
General Note:
At head of title: Committee print.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 025790252
oclc - 02426317
lccn - 76601812
System ID:

Full Text






Deri. 20Z rc 26

i, ii==, i ,,, ,,i iii i ii i"""
IS > 1
A P R I L 1 9 76
S m o

P r in t e d f o r t h e u s e o f th e C o m m itt e e o n S c ie n c e a n d T e c h n o lo gy
WAH=I NGTONiiiii :i 1976iiii~i=HH~iiiiH H
Forsae y te upritenen o.Dou. nt.. Goenmn Pritin Officeiiqi
..........................C 2 0402= P i ce ii2.6 0i

OILIP H., HAYGUE Indianhaima
IHMA LOYDWIG CaliforniaZOI~h,'alf rn EOM FQA, AMoia, NewN YorkOkahm HRIFSTOP.ESYMINOD, CisonecientN .WDLR N, WIAEL T.FLOER, AlawaaLRRsIN J a
ROBERT (B OB RNeG er exas, RY J.Vojd

JAMES H.SCHEUEJNsew Ywork T Jr. -. MR, an


MICHAT. T. Bass AaIdea

JAMES ~ ~ ~ EAD T.BANeRDicia TIMOTHY ~ ~ C KaWITwoord
HAROLDA. GUD, euviirco Mitns LI B. SUPEAE,.C~ e

Suscoxict o.- DILoM, Ehiaonutn

AYD THONTON, Acec onutn

:ENRY .. KECM Californi EROME A. AMBRO, Newe Yorkk

lut', X!! -exa -(IIm

WasingonD.C., April 1, 1976.

remain onthesubect"Reearh ad evelopment Expenditures

The edeal overmen iscurrntl spndin alost$205 billion
eac yer fr wie rng ofresarh ad dveopmet actwiites. Th

prvt eco pnd prxmaey$3 ilo ortesmeproe

. ..... ... ..

Of T


ft. 5

....... i.. .~iiiiiiiiii iiiiiii .. H ....

U troducliiti'i- --------------------M 14FL & D EXPEND.R.ES ON T NATINAL BOONOME
'm~cencj~jkmmiaClywhandSoial Chage" by Jacob E Goldman, S ie and: Technology" W. Dale Compton,
U S Goer~metSupor. fr ivlan Technology: Economic Ter
Vers r0~eqjJ~rcW' y Gorge EnA,4 from. "Research Polies,"1
w~h~ A.E~eftls Whos Hlpe Pear FoAks and I'll Kias Bar Hand" JW~~~~~~~~ PoM Qt ro .oilic7.," March-April 1L974------ 1
"New ~ ~ ~ ~ ~ ~ s ftnr LAD.yMra1 Waldenhang from "The Econ", ,a~c an Naiona Poicy byPafrick E. IHaggerty, from "Science,"7
"Vft Scenc ca't aveUs"acoversation with Dr. William O. Baker,
presiftk eg,'TWlepime arais, from "Frbes;' Fterary
Ma Sienceand oney olve AtlOu Prottems?" an intervtew with Sir
Berstdlo*R tv. ".S Net;& WordR egpot" Dembter 1,1975-- 3
"TheBeaand! ed T~e~oogial mplics ank of Percent Gowith" by
.10b eA ,, J dll r~n r f om selt rg and Caltue ,se e csober 1ee8- 4
"A eviw o th Reatinshp Btween Research and Development ad
Econoic GowthProdc~ivt A"hkpll andtproneedings ot a colloquim XatonQ134enc Fondaionintoduction and anmmary by Leonard L.
"Men lne~tir4 nd FconmicGrowth" by Nathan Rosenberg, from
")Wetltratur an Inovaion--A survey" by Morton, I. Kamien and
Nanc -L. Sc V",fro the"Jornal of. Economic Literature," March

"Neoomi Reortof. he resden" transmitted to the Congress, 1972
"Sience Boarato, Naioa SciencoFoun
Leter f ranmital PrsidntGerald R. Ford----------- U
-:,-ResmwforResereh nd DvelopmenVt"hglights ------- 1

'An Deelomet Program in the Fiscal Year 1977 Bnd(-f $10'ed'~xrps fomAdm inistration Documents" by -Doroth M.Bat .$, ipeialst scenc rd technology, Science Policy Research

"SevntyIsses:Fiscl Yar 977Budget" from the Executive Office of
thePreidet, ffie oMa ageet and Budget, January 21, 1976- 16
"Naionl Piortie: RundOnetoBasic Science" by William D. Carey,
from '"Scenc 6,7 .... -- 1
"TheFor Buget NewSigalsforScience" by William D. Carey, from



"Business and Technology: The Beautiful Bride or Wicked Stepmot by Peter F. Drucker from "Vital Speeches," May 1974.. ----"The Silent Crisis in R. & D." fo "Basth eqk," March 8, 187
"Federal Support of Commerelly relevan & D." from "Amer
Scientists," by Louis M. Braneomb, Mareh-April 1978 ..............
"Research and Development in TRUftlr Growth" by WilliamN onard, from "Journal of Political Economy," Marb-April 1971 "T'he Relationship Between -interstate Variations in -the Groewthatf.RD.and Economic Activity" by Ira Iorowitz, from g Man
ment," September 196'7 ------0
"Industrial R. & D. and Innovation" (Natlanal Mate Board) from eRcO Indicators," 1974 : -...,.--- -.-------.........-- --.--.:::::--- 2
"Returns to Researchf askd Developmnent Expenditures In the PilvateS,tor" by Zvi Grilices, from discussion paper, Harvard Institate of 9o
*nomie Research, Beptember 1975----- ------------ -- ..
"Resea~ice and Development and Productivity Change in the U.., 19 1968" -by John A. Shaw' ad Don R. Leet, from "The Journal of lnam trial Economics," December 1973 ---------- -------..218
"Muddling Throug: Government and Technology' by Willan D. O from 'Science," April 1975. .-..---- -.. '.
"Briefing: Sefence and Tehnolbgy, and Inflation"' by. C. H4Adan, fo11 "'S ience," October 4, 1974 ------------------...................21

PAR B--R. & D. DaicoNeg N INDUSTeR
"Industrial.R. & 1 190' by the Long Range Plannin Servie I
' ford Research Institute, Menlo.Park, Ci t.... 1967....... ------ ............
"R. & D." by Howard Wolf from "Electreates,f October 17,1974
"Patterns of Impact and Responss.-inBesearch and.Developmegn4
dustry: Summary of a Study" by Gay Black, frem Georget Walhigto
University Program of Poliey Stalks bint11 abD.0, Mar;ADWd_- S

'"Fear Extinction of Inventor-Led Firms" by Loebe Julie, from "Elect:ni E ngineerin g T im es," July 28, 1975 ---- ---',Research and Development Costs as a Barrier to' Entry" by Dent.
Mueller and John E. Tilton from the "Canadian Journal of Econbo ,
November 1969----------------23
'.'Research and Development and Other Determinants of ]Investment!Idustrial Research and Development : Characteristics, -Costs an i sion of Results" by Edwin Mansfield, from the "'American Econmi Review," May 1969-- - - - - - - - -- - - 21
"-Science Indicators New Report Finds U.S. Performance Weakening"b
Phillin -11 TBoffeyr from "Scence," March -12, 197G_ _----Heyman, from "'Professional Enierg, September 1975_ W
"Technology, International Compeitions, and Economic Growth: Som
Lessons and Perspectives" by Keith Pavitt, from "'World Politis,
January 1973 - - - - -- - -- -- - -- - - - 6

I I V11 t'l


els a. visi-literature, on research and development expenditures an4ib4r Jinpact on the nation'al ec : onomy and, conversely, the devel0 i e M ational economy which impact on research and del"' 116' ieading materials included here are divided into the
'ta sections depending' on their focus: impact
9 primary
c4"WY eivienditu'res- on 'the national economy; issues surroundino,
ra'i ex n ures effect of R&D in private industry;
pe dit all,
iWf;oLitional asp-ecis of United States R&D Funding.
vision of the materials is not to suggest that the issues do not many in ment seemed most arbi01 a .116, 'fad, irf stances place al,
tr 4nd was determined by th y focus of.the author. There
NZ. 96;an effort to introduce the. reader. to varying views on each issue.
The context for the effort of the Subcommittee on Domestic and International Scientific Planning and Analysis of the Committee on Science and Technology to address the wide ranging issues surrounding R. & D. expenditures and the National Economy was stated by the committee's Chairman, Hon. Olin E. Teague (D.-Texas), in a statement on the proposed National Science Policy and Organization Act of 1975 (Committee Print, 94-th Congress, First Session, Serial C). Mr. Teague observed that of the many reasons for the legislation under consideration, the issues "cardinal" to the efforts of the committee were that:
(1) We recognize the prominent role which applied science has
played in producing the goTeat problems of modern civilizationthe crowding and congestion, the excessive gobbling of natural resources, the dangerously shifting foundations undergirding the economy, the disruptive social and moral influences abroad in the land, and so on. Indeed, such recognition was directly responsible, and in large measure, for the concept of Technology Assessment and the formation of the legislative Office which now bears that name. We know the need to understand as best we can all the probable impacts of technologies as they develop-good and bad.
(2) We are further aware, particularly as we look about and see
the critical problems facing us with regard to food, energy, materials, security, economic strength and the like, that the solutions to our problems depend in some way upon the judicious use of better technology. Former Presidential science adviser Dr.
Edward E. David has put is succinctly:

(Jan we be sure that scece and tcnlg ilfn h n

Noi bu~t we cnb ueta ohn u cec n eh

If they do't notiges il

era has beenreonzdopnyotailbevrA mnsrtion of te pat4 ers
andl changing srtge tteFdrllvlo h ainleooy

AtheSbomte himnsae ncrepnec ihSb co mtedebr eadn h ainl o ernso-hstml

exenlfcos An ths ar h fet fifain h tog
prgrs ofR .i te dacd ndsra ainad h qing~ in inutyta nrae euato scuigadvril

industrial ~I R.V &h D. fud no'rvnie ate hn'noai;
reerh Th ernso hssbet ilse odtrieA~
rationl for th e ugt h roiis ihniadteefc

1..:'.'.. ................ Foiiox i AO T I A D O I L H N E
By~i Jacb olma

First caeteior oethntecrir ofi busins and com-~~i~~ tegoenetadteaprtsowaannoithiiii sud-iii~~ii
den This hasiiii been thei p o rs ion ofli
................... ....................... ..... .... ... ..... .................. .......................... .... ii r e s e a r c h .,, .. ........................................................
y~~' an tlzainoisini

Vntil elatiely reently scienific esearc (as isic rpteh
Ailcicl reninern rserc)wa te xluiv poine f!h
4daiemiccamps. ts uilit wa consider atbes vage an th i rad hatlikeditspratie t th evluionof ew ecnolgie s
windng-bth satialy ad teporlly-hat o on felianycom
pulsio to itrudeupon he prvacy nd sactityof th resercher
.)Rmach wa (andis) ynonyous wth acuiiiion o new nowl

,.4e.-Wer bt t uivrstyshul schasirtinsfid hme
..:he cae he erodbetee th to rea wrs-hepeiod ofiiiiiii

gret idusria epanionin heUnied tats.Her anthrei as....

di~cvre ta siec cn eprftalei ladn t adpon i vp,,,nw matrialsmethos, anddevicsiandihusibidgeiteigap b~thin i I e ad i spae-btwen scentfic iscvery an tehno
logicalinventin. Reserchisudenlyibeame a crporateasset-a
-nt ealty t bereconizd bythemor
one, butiofiiiiiiii
proresie, ndagresiv crpoat mnagmets


has spread to the public at large. Science is written about daily in the public press; there are even several comic strips devoted to the serious expoagulation of matters scientific and some not-so-serious ones that will tell you that whoever masters magnetism will rule the world; it is talked about daily in the legislative odies at national, state, and lbcal lvels it I%(s eve tb6feft 89V%'o t i shigistatis eir We practitioners. The scientist in the public eye is no longer the mad iconolast but a normal human being.
But this spreading of the informational and financisleapgai base has brought with it a whole new series of problems and challenges It is to some of these that l chose to devote the balance of my renaks today. .
A question being'asked again and again these days is: Why can't the science that created nuclear energy, that put a man in spac at fashioned incredible y fast and powerful computrs-why-can it scientific apparatus be put to work to solve our inajor social probT If, indeed, the techniques of science are so powerful, so all em ing-goes the query-why can't they be put to work im-provym lot of man-putting an end to the major social difficulties confr us? All we have to do is to institutiionalize the process, a la the aihattan project, and all the ills of the world will di appear'th o the magic of science.

would like to rturn now to the field of transportation. Whom A, you might ask, lies the scientific component of transportation. Tni e, say, communications or ordnance, there are no obvious undtt ing disciplines which can combine to advance well defied frontier Bcause of our intense preoccupation with problems of transporati we are in the transportation business not just the automoNe business-we asked ourselves the same question. To fnd some as we put together a small group-a physicist turned bperatioI1 $searcher, an electrical engineer, and others-to see what dimns is can be used to describe and understand transportation. We ?~Ve learned a few interesting things one of which I should like to 4esoribe for you.
I believe that we all recognize intuitively that transportation 8affets the way in which a region develops: that transportation is a paIt of the urban system incapable of isolated analysis. Further omplicatif this analyicl problem is the fact that we are confronted*ith aj rkfusion of technology-hardware ranging from conveyor systems VSTOL aircraft-with little quantitative knowledge to guide' its application.
One of the first questions which we asked ourselves was are there any mathematical tools or analogies which would enable one to quiantify and optimize transportation relationships-somewhat in the manner of the use of Monte Carlo methods in the analysis of commnuncations networks. Clearly, the philosophy of systems analysis was indicated, and the lack of an exact solution dto consideration-of mathematical modeling and simulation as a means of analysis.
Consider thes question of the growth of a city under the influenceof Population, economics, Induty trnpottion tpgahy, AndA
-other constraints 011 land ise. it is possible to deilop amathiatiil model which, based on projections of basic industrial activity, will

be shil edde
oa t sp ftfordMid A AddPF al 1owerdr *hieffihe 'tIr'
u no ooL~~fig th duhi of popuilaid ni~ h bm .ou trsa.ispio6 !RRtTq4n~iifRQ4P4qaqcog di ft tafo r codhitif inpt.h"
JAP Jlstrid the aition. of better Plannera on ec~0operat ion of th orina~ti
iliqf W othe Unived ify of IIllinois, whose facilities we
#T~~~~p JofiE M9 film: Was generated here that illustrates a,
rqgi, wich is to be served by, a new, road. The planned
,P of Ilteriate' road. Joeaions _by drawing te prov fico9a -raP~e wih a,1ght pen and watching tid
uni~~~~~~ 'fisa'tqyaepqtte by flid comput er. attempt feitrapplate these relatively limited
e e4 at ye sol-ution of 'the mKdh -social ills of the
os~teegytq me, however, that one' of e formidable obstcesto overcome is a i institutional, one. The science and the ,kietot that will help solve these problems will still be concentrated lrey on academic campuase~s and will continue to maintain their ,Ptlnr~y YAIty t6 theii disciplines, their professional journals; and
-per goups. Such instisfitions as th6 Coordinated Science Laboratory l ihps a step in the right, direction, in that they provide a focus frte artilation of needs that tianscbeni the traditional boundaries ofh intellectual marketplace. Weinberg sees ina his new and urgent S6,demand, a Possible role, for the National Laboratories. But more wwa~fifz' else, *e net A cadre oE seextifically trained secomowar.Aho see a, profeweistaF goal in ministering to ths, fte ,6T~, .r1 that transeendd'the four walls of their disciplinary
o he, nstld mdifcaton n the instittitionalization of
in et dkkndfacilities lie two additional levels where new
bb ~ 'U steled'for methodology 4nd support.
twq= e flid and telfiushv rged to help. the
I edftist idvthibe his frontiers. Many of these, such as the new 'Sdstate epptroncia and computer technology, were born and bred
Hifty'Supp-or Th plication' df science to scial probleins
& ane~qioration for useful, methodologies. that may already
ofmay le hidden in the rich storehouse of emerging knowledge ftheI behemoth of presently supported research. One technique, n- anlss whic is showing promise in attacking these com&x ocial problems, is also a direct result of a wartime need. 'and. citantiaion of' goals often conflicting, occasionally O dc t'. ry-and optimum deployment of limited resources to apn thetheff goals is, as much a concern. o-f 'the city planner or public oidras.i iithe concern of the miitr cmander. Methods
ged, byightemti cians for evaluiationof weaon sytMs eff6
nja 96hamleprvide the Wfl picticalayocompari .&A Glfbec6bt~e fat~aieltoS t social: brobl ems: -

Fiiinaly th cetfccm uiyadtegvtmn o* P

must iati!day

bic scii i ec deie.t..potfo.h mltr n ow s~ls
..... frmteA CadfrmteNtoa cineFuiai
Th scientific,, 3mmunit must, now face sqarl the .......... pii res othe semet of i the..............oi et toapoi cec
ifiiiiiiiii~iiiiiiiiiiiiii they% H are to ex ectto ei ve benefit fr m it. The cia k nge toiiioi
ernmnt ~i s equally ber' Those sectors of government rkoiire direktyii~iiii !i~ii~~iiii~i
iiiii ~~s )Aiiiiiiii e m siiill~lliiiiiiiiliiiiii o f sociiiiiiiii ~i~iiiiiiiiiiiiiiiiiiiiiiiialiiii~iiiiiiiiiii~iii iiiiiii ch an g e, eiili~gii, H Ui I D O Tiii
concerned wiiiiiiii~ i th the robl~~~iii iiiiiiii!~i~!!ii~ii
H EW,~iiiiiiiiiiiiiii Com merceiiiii~~iiiii~iiiiiii mus acceptiii iiiiiiiiiiiiii~ a~ shiareiiiiofiiitheiiiresponsibilityiiiiiiiiifor thie support........ of science Th ......... argum entsiiiiiiiiiiiiiiiJii advancediiii toiii createiiiiiii thei m iiiitaryiiii
support agie fo ai n cdmcsineaejs svld=
arun fo upr fsineb hs ohraece.I hyepc
t h e s c ie n t i fi u.nit y............................................................................................
prbe lin wihnterpriw hymutpoietepi a
supp o s ......... bs alytInwfrgnati hscopn.My
wha I'm....... asigfri oilA PAo nO RfrH Dada
OSR for DO .Hwvri sac1jsetepeeethsbe
etishdb hs oe'he iso-retdmltr gnit

ItrmisfrteW tr onstprftfo.tentoa

,&PWA wroe4and development by a wide and afe uforeseeable

Lobgow si ilar in tone also appeared 'th the 1resident's Mesapg
-dAAsheei a 4. Technology, delivered March 16, 19f2, in which the
a~ -of the proposed US. Government initiatives was detailed.
Statements ought to bring joy to the hearts of economists inter&Win Mta rationalization of 'U.S. Government support for commerinoblerinded science and technology. They contain an explicit ague nition that the principal justification for such support is the
Mu6 afmarkets to perform their job adequately. More significantly, inplid in President's messa miiitment that the it pri
gramae, was announcing were designed to-supplement, not supplant, t hemarket. Their aim was to create a set of institutions to diagnose ad mureat market failures: so that the market then could be relied
u=to provide correct signals for private inyestment in teelmological
0j l*ee any such joqy ought to be tempered with anxiety as to
*istikely61, to be the result once these admittel admirable inteatides;lik ao be implemented. Indeed, there is ample precedent for eemetrH .he -ecanomie history of the United State is full of attaenpts
blthbt-aam t o correct through direct intervention what have Wiw rperceiied- by some as failures, of markets to direct economic sMvy elprp aperly. The regulation of the railroads is an early example. Thei ded?%ien to, undertake economy-wide wae and price controls is sidmo eseent. one.. However, in an unfortunately large number of easespthbe attempts have been unsuccessful 'The market failures,
d~he M zea r imagined, have not been corrected;, and, what is worse, a hos at asse distortions have been created In certain cases the prhasfnee aof,,industries subject to direct economic regulation has been swea met that legislation has been proposed thtwould substantially dd MSfhed lmea, of regulation and increaseqh reliance upon marksh lpeatest thugh threy may be. (The Re.uaoyModernization At nameodaed in the fall of 1971 by the Nixo Administion, Wfid proposed relaxattion of rate regulaton and freer entry for stae-feight transportation is one example. The deregulation of warn-d gas prices, proposed. by the Administration mn 1978, is
phe~fre as though I am encouraged by thac tha thie geeal AWea af the market failure justify' gownd Intervention has ben. orrdctly perceived,: and thoughIa ereebyt st
dent as of the current aminiitration to employ solutionsaprrit to. rte problem,: I am extremely apprehensive about what islkl to be the outcome when government bureaucrats, operating 'n a highy
chagedpoltia atopere, atteatopmpt to improve upon te workmgs
of mprsoalmarkets. My apprehension is hiteed by the wayI de.actual U.S. Government science and technology POlc developing. Already there is a large gap between the admirable stated intentions and, actual practice. This gap is caused in part by the fact that the theory, of externalities and the conditigns under which its simplest redction. is & proper guide to policy have not been clearly understood, by~those formulating U.S. Government science and technology
POBYAY i tu~rtxu~n ha. eenabttp b th filue f aon
ondss t prsentthethery o exernhtie in anopettioal orm

cials.charg'd with implementing scieO-6.po'ithowiikkoysdud be apple in specific cases.or to demonstrate to h,~, h uU that the thor is being milsappli d.'This is no he .sy ht eooit are powerleas to. point out gross migapplicati~g"fte hoyTi can, and obviusy should, be done. Izi the-ms 1 ae,-oevrlh proper, policies may not be obvious even to ecomitgvnurIu 4 rent state of understanding. Therefore economss hudnteW quick to criticize when policy makers, coifnfitt i~iratv theo-ry and faced with'the political imperativ fdig oeh] proceed to produce what may look like bad police
Let, us focus on why the ,simplest' prd8titn'o h horfex ternalities may prove an inaccurate guide fo tforuiii ~ e~ nology policy mn specific industries. The basicasupinndrym the predictions of a tendency towards unagarinvtmn tcolg-a change by private industry is that currently dcsosaebikid in am otherwise neutral atmosp ere. That ti :asu~d t absence of a mhechanisni designed to allow capueofxtialisI the only market-distorting force in operating: h. sMupin l~l is .violated in practice. For example, in' certainoipliteart; by tadit agreement of the firms involved, cotionhabet dreeted away f rom price and towards product ipoeet h;frm that has been able to offer a slightly iprovedpouthsbe &l to gain a substantial edge over its ial.This'h t-ft
vive in such iixdustries, firms have been 'kqui:e odvt asae f the resources that others miight ihve 6ph. to their market positioni to research,,ldevelopent n eyrdc design. Prices have beebl hi e tianthey oUdhv'beLuer more competition mr cod tions itudthertoftznhgC
change in the inutysbe 4hbfdued.@I~etm;ase h w
of investment in tehooial cag by (iidulfrm:esoig to such incentives idhy Mitey ekdadi'the scedGtmII~e.O to state the result in more technical language,dut h.eliaks of Market structure and product characteristic hc hn6,ea petition into nouprice areas and which channelnarc optt~. into competitions for product improvement, thopiaertkfrtdr to innovation may in some cases exceed the socia< rt:o dr,4a ing to an overcom mitment of resources to the: prcs f1cnlgcl change. (This point was suggested by 11arveyBok, h ig
out the pharmaceutical industry as. an -example weea vromtment of resources to technological change may hvocre]s ajrsult of this phenomenon. An article by, Comanor[]sg~tstaThe tendency toward higher levels. of inve 4stment i]tcnl~clcn for competitive reasons exists throughout: thecnuedraei. investment goods sectors. The argument presne re otig more than a variant of the well-known pr<)oeinthtt aon of nonprice competition in oligopolistids 'market.4lkytobehgir than socially optimal. Ordinarily-the only -type fnnpieom tion considered is -advertisig, yet

mefiw l ircti of hilagese Am M~~~~~~~ p4r oiac, ;st urchasera ofS Gos and series a ay eresne the estainly

,71I ie n iiZe4,hti
19,W.1-godepileetect of .S. Government pro-sitnlyrvdd oethnhl
of,.feAM& serte and hodteeemets thuh in
h ar designed to promote re,
b ~ ~ ~~~~~a Aqf dmiihaabe cncnrat ion inavrlfwscon, griartes
umrne wit defnse nd Maedciat tcnly[] Tocal com-in
dustres. A~ otctio astimularetedn inflec hsen h~wev et 1 indict g government subjene nectnoomaj sctorsgo
beenevengreaer-prtiulatlin (Shereajriyo simatesa
ricpfirect economicnregheationfcontribs
ectheiinfluence ofathebtex laws, th

non-dfeeinduts i-areatryerventions aristl Wor a iscusion o the cotabnfts of th. o e innpoCumentp~ntheincntie o inortes and, hwee soefence on
'.1venmet rogam whiare aseiced by prthe heryofd all-wl uchas.FedrlHouin certininstries then netrainflu, 7=1_f trficiionMediarins Miad h aben to lacnuat n # fo. impovedpol otmthe estin sev he I t~te, and nay avet alredy movectosted Jestment so t tr r

utbe produce if externalitiesGNcould
St ingenralis ubjct otel calec or the ntituio ofws me reoitm of:.6 nvirnmenaddiottion piaeinvstentinus
atL theprpitndge fcm
are prgrm aplcal to allwin
theseregultory rquisieultorth intellientn formu,
th6 t~lfth. owrd uner'vstenolic rite by the U..oeorn isf
as simp~~~ly exprse.I etit ofute the net ic f the enpeof. urrnt government ti-tst procurvembento andentrade

ftlisrs ~s:if tdisouageaios)ona private intvestoen ins
CtL TOiiiiiiiiiiii
t6IR 109ic gei

r'AAisradlyapaen i ha teaprorit dgreofcoi
PVtiU'& nytedectounernvstwil ar wdey yi

io ..hi~s sugests thatiniutry-peciic -rograis ae moefikly t pm~u Lce al tan re ~r~ram apliabl toa iiii
qik~riate. result

sidcretil soldiegive reqiv n b h gnisca
ths nischarged with regforulating varou se.envrrte

to produce "Technology Impact Stateet" nlg o hlni
roninental Impact Statements that curnl utb1rpt6'o accompay their major policy decisioC sense for the Interstate Commerce Comsin hagcycred with the regulation of U.S. rail and mtrcrirt osdi h effect of its rate and entry policies on vate as it does to consider the impactothssaeplcs onh. environment. (Any attempt to do juttipscnaie ntedcr
ment Improving Railroad Prodswtivie idRpoto e
Force on Railroad Productivity, A Ieprlo h aina bms sion on Productivity and the Council o'. nmcAvios Nvlber 1973).) Indeed, a major candidaefrpprigTcnly inpact Statemeints would be the VnWromnaPrtci. gey itsef., Many of this agency's actionshaeamjo(i urnl, Trecoginized or unacknowledged) infhu-e'nteR .deI~s
of many industries. (See.. for example, umaieRgltr fat
oni th Cost of Automotive fransportain ia eoto b:k
ITo .Conmuittee,. prepared, for the Offieo cec n ehgoy (28 February 1972) .)
Admittedly, it might prove adni Technology Impt Statements even if nyfrmjrdeiin.d0 ever, some means needs to be found torasth IovincncoV nese" of the governmental agencies inovdi irc]nl~e regulations and to provide higher level dcso aeswt.ifra tipn about the current state of inentvs ota n rgmdhv
-propose to institute to correct "marketfalrswihegd W-f6 investment in science and technologywilblkeytprdc Prk
priate results.
To illustrate the scope of such influecIhv opldtefl 10wing list of U.S. Government -police hc aeusgicu impact on the rate or direction of tecnlgalhneint. D)( mercial aircraft industry. By this chocIdontmatootna that this industry is typical as far asthdereoU..Gvnmt: influence on its pattern of technologiachnescoerd.t,1. in fact, -Derhaps the leading, example ofsc obvious of cases, the degree of this inflec snta l udrto by the relevant decision makers, as ce:ti oiypooasd~me below will illustrate. The list. probably i o xasie n-atclr. no attempt has been made to include hsplce hchv~n
impact on all industries such as the favoxbetxteann wre to R.. & D. expenditures. Finally, noquniavetemthsbn made to compare the level of technolog~a.cag ht atal 'it in this. industry with whatever ex .ternaiis a xs, huh8m quantitative estimates, of impacts woul eancmrypr ftp noloev impact statements, if they are to(eusfl

'Ate~ V.. overnment, through tbe National Advisory Comnittee 4 and its successor, the National Aeronauties and Space
AA11instrtinand through iteshltary aircraft development proJUdfr most, if not all, of the basie research and much
014ho-,06d research required by the comrc ial aircraft industry.
Th i : tchnlogcalproblem for the idustry has been to' choose
wor reatielyknown, and in some cases well proven, technologies akr~t6&'heb asanibl a product that prooides ehougli of an ecoii~dtidva nta e e existing models to induce orders in a volume gi~tciA o jutify production. In short, the risk lhas been primarily ebodi o teclnological. Only in the ease of the SST was the inCh*' ia ;ting to push out the technological frontier with a comifek62l ircerf fto any significant degree, and even here prior military &*0opx6kneh as, the B--70 has answered many of the homec
ali etionis [OJ.
_!e t parent study by Sydney Carroll, the U.S. Govern*A *mttedsircraft manufacturers to earn rates of return on tota inestment on military aircraft projects that have been sulatant~if.. % 6des f the 1evel required to allow these fims to attract c~fltd:Tis-asperifted theemanufacturers ed initiate commzercalt
hooc se prsetv fate of ieturn, adjusted for risk,
11'b.6'blow tht hich ,ordinarily wiuld be retired. This. has
liq14tun,.d t redfer number of new commnercial aircraft projects
a" uenlyto more technological change [7].
P~h1* -olid esjust niitioned afreet the supyside. Other inmport US. Government olcis have had a prfondeeton the 16 4 'ndi-e ii 'tnt on the timing of the demand for some.
nieiel~ircrft. tndhis pulFD111 has had its own substntial
im'tw .(i te rate of technolgcal change in the commercial airaf
1(5 1 arkt ulP' impact need not be always in the direction of
stifi~lai creased investment in technological change. Leonard Ikdimin, n commenting cn the original draft of this papr has
stige~edthat in cases when -U.S. Government stimulationofdmn
reuls indetmand exceeding supply, innovation can carry with it the rikkof isoption of the orderly flow of output and an attendant loss ofcirrnt sales. One instance in which the drive to maximize current oi]P resulting, from excess demand has indeed appeared to haie
14da erousadverse impact on the incentive of the industrial enterPieto innovate in spite of relatively high long-term paofsas
b t. nSoviet idsr 18]. The net imatof demand stmlto on private incentives toinnovate must be judged on a case-by-case basis. Among such policies are the following:
U.s Government subsidy to the trunk airlines during their early days of operation allowed them to purchase aircraft in quantities that mado, production of new aircraft types feasible. This was particula-rly itapb~riant'during the immediate post-World War Two period. U.S Government subsidy to the local service carriers at first supplied the carrier group with funds to buy used aircraft from the trunks. This dlemanddrove up) thxe prices of these used aircraft to above- book levels

importantly for. the trun ipep the th i
wilingwould notdeChagslhs up
t. ar i r o.f .4 iteO IT up_ F
lito E rdtvthn these carridia .9UualVI-APF
0 il Arnics Boar rglatoofirc

iiiii~liii th re i-W rhnii4fi, IA
discourage carriers finding themselves in an infe tion from attemptingto overcomes disadvantageby fares. This has tended to produce 'bunching'7 of airrf when any one major carrier ordered a new airerf LYR6 substantial pressure for his competitors to follo.Is of orders is crucial. to the successful launchingofaewtn program. William Jordan in; his study qf. th strategies of unregulated California intrastte crers h this impact of regulation on technological clangy i aircraft industry [10]..
The U.S. Treasury. has established, extrofillibrler~ll0 poilees for transport aircraft..'The mrost,,:c4 TrA allow such investments to be Writteni off-o-Ver a -ieya emIr tax purposes [11]. To the extent that air carriers aculyk4am for a period longer than five .years--andit}4, k pot.r policy represents an interest free-loah frp m thii Tpauyt~te.4 which is translated at least in par-t into. an inasdema,*r airraft.
h Export-Irhort Ban, participLtes..sta...... in'g of virtually all' aircraft s old ovemas. 20sptonU~o limited to .aircraft sold. to under'evelope4 coy~tis Japan, Switzerland, Scandinavia, and The Netel nshv,4 ceived Exizabank assistance in financing recent wiebd-eta7pi tions [12]. The availability of such fTinaicg~canhva q iipact the dlmand foraircraft. Te erspa...... cation in a iece publication noted: "*, *oe *at ofthn est, wlde 'possbl by Eddmbairk participationpurie itni
period of 6tightent allowed by the bank, can be~qae g'c~~ a buiyer who musit plan o .n covering.,all -repaymentqssYil ing earnings. -Am annual cash flow -reduction of asu as P n below that would be required with straight con etoa is possible." [13].
This is indeed a formidable array of programsl vn o.cret any' tendency, that might. exist for the commereil.arrf to underinvest, in technological change due to itib ilt omtM fully the benefits this technological change creatsfr o 1r tain programs or policies may have the opposite ct Zrk ernment regulation and subsidation of local servicecrirlhst4e to discourage the development of aircraft design rv es
cient service on short-haul, 10w-density routes. (gead 1 p O 125-142.) Exem ption from economic relation .......................................................... i~ii iii~ ~i = = "J," I,',, ;iiii J

J44m~ithfines filedl saiantith saftkb i atWMct thd' tto te, agreement amounted to a dvx~,i
Asokth d development qualladiretarded innorating ']a~qrl
1.S Jstie Db0ptassnn spres released, 9 -Ma
I" vw~ithis, a APresidential Camiissindeak y cocldeabh h Ukf~dSt too can no'longer reyon tliedfrepe.masket topd uc o praircraft: in the quatt andi variety. required ao neat the 0of foreign competition. Te Conumission proposed the creatioq~bf.In agenc-y .withinth Dlk epartment of ,Transportation th~at woulddecide when new aissift types were 'required' and also the s.Pmyhnl would build thli ew aircraft 114].uAt abutthe samne tm:he then chairman of the C ilvil Aerohautics Boad, Senet Bvsn asW touring the country raising the spectradthat wqithout sub s" As a n='om th U.S. Government, no more noummercialnairraft: "ttlI teodneed in this cutyafter the production of the curr~femeration of aircraft is co5pl.ted [15].
-#Vi8tui h ainples shom~o a hbory that has substantial potem tills as a guide to decision-making ha e used to justify-bade polapoists tsheawthhatother po beigtoutedis, infat,
The theory afteraite in itidsimplest forme predicta under a certain setbofhasixtaptions there willbe a geeal tteby lor Private industry to undqrmnvest in technologicalkchage sod Iates! that governmental ihterventish aimed at! norrecting this tendency may be proper. The practical outcome is that someone- etaW ,.emfe a party having asubstantislfpivateofinancial intearest in theothomeb-perceives that an ihdustry is achieving a rate oaltechnological change below the Tevel that the particular party, believes is fesablat After suitable 6ub1icity has inbresed public* Aviarenes ftt*-robleM'xi dits,a prest ians panel i thqrupdtpvdd Afe dida q oprister interval it prkokce a report stating this while, of emikse ,everyone knows that the edonoay wbuld operate bebt:if :th market were left free to operating the partionlsr case at hand4th market has 'failed' arid cannot betrusted to bing about: thesocially desirabiek Iretut. It is at this paint that the theory of external itiets ine61dd. IIn the case at hand, the faotathat no I T. S.eammitifah-aircraftmanu facturer is producing aircraft equivalent tfstde A-3800, the Merand, the VFW-614, the Yak 40, or theoncorde is takenrasqproof, Pasiivwe of market failure. Nobattempt is made to hae a 4-truly dies a~ktoested: outside party -determine whethdr the alleg.,d market failure hAiddedoceurred and, if: ithas, td pinpoint dthebuerf thatifatlure, that a rianedy "designed talcorrakt: the pkibtakwitiva mthdnur~ot

Ptedential Advisory Commission or to Mr. Browns that'the failure' of th U.S. co mmercial aircraft inidustry to produce a&nalogs to th,6 si r afft mentioned is perhaps a signal that the market is indeed working properly. In this industry s in most others--supply does not
a promising commercial aircraft was hftot produced'due to' the failure, 0-fithe market )properly to6aeiscmnrilpopcs but I hav6 unc .over-ad many cases in which aircraft were produced after having:

Thusf ad attention has been confuned to problems rate~dbytwfj that conditions in industry do not generally correspond totow0 must exist if the simple theory of externalities is toproeeM ambigons guide to the appropriate level of U.S. Governmtjat" vention in the process of technological change. We turn nowttl. dangers that may exist if a premature attempt is made to utU Government science and technology policy eon what might apa-t first to be an economically more rational bosis.
There has been much concern expressed by the scientific. den that- he fiscal 1914 budget implied a downgrading of basic sfoc a favor of applied reesarch. Science policy was said to have coesuuw the reign of economic decision-makers ("14's Austerity. i*. o Basic Science" [1}. Thewsubbead of this article' states:," ?Y micians are out. And so is the emphasis on backing basic resre") Williani 0. Baker, president of Bell labs and headeof the rog created Sqience and Engineering Counceil, stated: '"The- Adminorr tion is anxious to apyto government more of the values-o at aui neas omnt-fiiny cost-efecetiveness chains oft *~qr'.r I17] The th1us of tie fismal 197t budget was said to have. en v,4 ward "naissig short-tarm payoff inrresearch" in energy eQ and transporation-aa doo'med to bebf "sen taial se
It is difficiult toshort out the rhetoric from the reality to learn-A, how signifloant any such policy 'shift' was. (For a view that the aon corn over-the "shift" is overstated, see Wade [19j.) If histerd" guide, atr least some oftwaet was publicly esale a 'shift' mrlr~, resentedtherapakaging of existing programs so-that they perd to be'nmere ini line with current domestic political cone ehighlightinigeren it only with rhetoric, of the desire to makeUS Government sciene and technology policy more 'businesslike'. as a valid concern over what is likely to be the result.
One pedictionisthat cdones out of the simple version of the th 6ry externalitaiesisthatthe egree of nderinvetmentby privae nM s try-and heads thersize ofthe proper role for government,-isgrae the closer one gets to the blasic research end of the technblogica kum spactrm. As the Preaident's Council of Eeonomic Advisors ntsi the paissape quoted arlier, this is because the returns to basic uesess 6 are uinlikely to be fully perceived at the time the researchi being carried out, are even lessflikely to be fullydeapturable by the aun patyu an are lieyt be realized only in the distant future. h uther toward aple research and eprodnet development one poim, the there lkl it will haththe Person ending the ms e rh-i1P able to perceive and capture fully the total social benefits resulting from his outlay, and hence, the less likely it will be that there will. be an underinvestment in technological change.
All of us, and most particularly we economists, favor aloaigour scarce resources where they Will bring the greatest return to society. And, it certainly is possible that in our desire to. see that -science: not go undernourished we have instead o-verfed it so that some simn down of unpromising programs is justified. But, how should an evaluation as to whether such, a 'simin downM' is, needed be. condtteed?. ft economists currently were able to measure externalities, this could be

Asitlatisely easily. The nit social rate of return (thelocial rate of
*Ot semsr the private rate of return) to funding additional reak Iffs ai Aeteais hschoogy, or stronomy could
dedto: the not social rate of return from additional govern1113,1 4:iAdfo the development of 'People movers', quiet Jet engmes or bfder: reactors..Searce governmental funds could then "be distributed
*0 1 tomaximize their contribution to society's well-being.
a pocess just outlined reveals the polm so oaeoo
4doored soutinely measure externalities. The necessary tools have
m4oe developed. Therefore, to allocate USS. Goverment fundq on
asis:of demonstrated potential economne payoff is bounto lead
9.aiundedfunding of basic research by the government for one of i~ae reasons that business itself underftihds basic researeche,r, ts economic benefits cannot be fulty perceived. I... hi fortunate tendency is likely to -be accentutated by the fact A4.Xhepwygaf to the government from support for' applied research maybeoverestimated. For reasons that have already been mentioned, Isuporers of funding for basic research are often unable to poiht to llyseifio economic benefits that will flow from their projects.

Ajundeand are successful. They can idso point to secondaryf' benehe production of these goods or services will create --emPL n~ntand exports, for example. It does little good, to point out .that.MMajority of cases, these 'secondary' benefits can be shown, if 14yp~e to' disinterested critical analysis, to be at, least overstated .a in etims to Lbe entirely spurious. TIheir claimed existence, ad articularly the fact that they can be stated in dollar magnitudes, ea~ds a powerful additional appeal to the projects they are associated
h., Fulrthermore, even the direct benefits of governmental suppot appled research lend themselves to oversatepment. Inmany cases
,gyrament aid merely acts to speed up the availabilht of some new nology. However, supporters of government funding for such
,poeats invariably credit government, support with bringing the a*.~lz into existence rather than properly counting only those ..44itonabenefits that would accrue to society from having this techqio-g, available sooner rather than later.
7'afore, it is possible that if government spport for science and
oytruly were put on a more 'businesshhke' basis before the tonl exist that will allow the full social benefit of, research to be maaared, basic, research might be undertuned to a larer dlegre .han it: would be if its support were left entirely to prve industry* This is not to suggest that this outcome will be likely. Despite rhetoric Aout. shifting U.S. Government R. & D. funding from a 'supplypush'}o a 'demand-pullF orientation, it is to be expected that most of
-the actual decisions. concerning basic research support will continue to be made (and appropriately so, given the current state of knowl.edge) on the basis of -rough rule-of-thumb calculations about what p~ro-nortion of the nation's resources it can 'afford' to devote to basic
Towevo,. fbe-dtortin that a mrematurm rationalization' of TES.

i ibiidistdibY a wi rat nth the 666 St .

iilbliiib n
sti e
p m icy fcult b6 i nisn t;fnaies exe le8 prou onc 'ri hasn pas'e~fse out ofade the rea tranlatn bstrct hienomiencIes ittae,'xrihtel M

that'd p lcm.a r. c un erta d The...............
eteniis aryota whel aswt st

mthpoiy mkrsdin toisc ern hilbow ln well thdep %twy And in whatouc dectiopm nst polic ima]eea aebi w 6i*a.1

hor are at 'ere if hese sum ptiin w on some mtonedtobfon, uc a U.S ST proet to 'blowthey whiuste

Pi u nte .reE.ocs Cenriso a p tffiond ]uct reainsceibility fFsci hati os nthp fulsto avid the lureof 'prtissipn. ToWfea ntee be economic conep hare misusverd~ tcoomi u i cpe aepbii
ipopr ohepersnbing the] his n t] rbtdi a-tf cent arewbein scuse o blte proferain h otpetg'i
ally finds disatahstperoThisn the, difsslhnree mkigh thfenr l operation. Wet econos itsiniet b.Usfilf rept o narlitha asse ouget ofteramo h cnri un
n n o d iy p a t c .E o o i f u t' a e a g a e n e e M
i! A
tn slatin abtac......onet..k etralte it aiga

that! poiymkr anudrtn.Thymsiipai lal o
exenlteIa ercgieadpihp oeipratywa

nor, 6,ret Strctr Prduct-$ Disterentiationad I
biW S7. Bonn 81(97)65

61 i @& Afi Makt bcue a ,tn icnmic .Peo
thtago 1976)' p. 519. 1 aor m
j I NeVo t,1970 :w
Il 016 pro; in0 theAifr'mIduty, Quareten J.e elts
5 pi a501 b.,.
plto of'uce ss Ind o rs nSve nuty cnp

& IAt'pfthe Pet*ien Airlhingeftnt (The) Brokng 126.
.,,t~~a 'ma orxkt Structu,H PBe#14t is(Brvtardo Tnnivesial
jV.,Xftxfie4 .d aj, Rsearli nd D&vaion n tea a~n leopeoestio

os Ititubio Wercashton, 1967A) p. 108-204.Th1
IF.1U", hport of theia Mr-est Srcuand Emc Perf170 Atorac j)(Mitat~l etaiof 1- credit A5ith N:IL~illertem for Abera eTcni Development on Modpensie Ato ti~, patt)[orial70 ;a .4itoel agay17)p
k'Teog RedMarke bytrcr (D.C et iran 1711A

-WW7panytt, Chiago1191)
IIIosts As Advocates, Budinesos' We vel (2 IJndury, 1973) pp.a

TranpBy Tons CottleUierit res Boir~y March/April) 197..ol me4,p.33
e? apArmnen hogusesion TnAia Streets behnd mpethe So
46mio .ks e, ,w..fionleEmnn dhvwltyf

lies. Years ago Vwen these Wel-t-do:.whie amilies.. .wM
porBlack fmlemyof whssha *
moved in, the. houswer covered idt... a
as mny as, fort d united each Mor, remain on the'roofs n msual eigsct 0i -t the buildings remind iMpresent resdents dfist ince'the W here, once artisans built buildi gs here, once life was rk'.e Today Taylor Street is one of many streets. in atpoor hidden from the main corrents ofBoto by the deve End area. Still, the children who ply agana theirpre somae of the rooftop towers on Taylor Streek elain theyam** f Boston, particularly the taller buildings and oee the' Charles River belonging to the MassachusettsInstitute of They report, for example, sightin htlok ote predicting equipment atop one of the MIT buifldings. Sevf- ,f children have decided, moreover, to explore the Indit=t! o close, prowl about the buildings. The idea of, anientis near to where they live intriguies them, and beesoon I co--e to visit with them, it is only natural that they would telo o plans and turn our conversations to the work Of sciential&
One of the boys who climbs to the top-of an saparkwtln Taylor Street to spy on MT is tenya-f Kei th ]Dowm thin young man with large round eyes and a high forea&h wh~ goes anywhere without one of his hats, which disturbs hi- a!* practically have to pull them off his head when hbe enters
"P1ve got my thinking hats," he told Me one afternoon. if t r ence with them on, particularly my asmy, one. I can go upiq t h tower and look at where those scientists work and just abu self to imagine what they're doing.":
"Which is whatI" I1 asked him.
"Seems to me, they spend a lot of time working on. -ox&6

space for a long, lon tiM so they dent eve have to: ear
fo fe ad l teother stat the nedlketei1 could work up there, 11 bet, for,. whole centuries, haveup live up there, have their babies born too, and when po; dn the -bodies, infi space up. thereyo donlt need t bry poi know. You just throw them out of one of your spaceshuprI; s than the way we have it here. Like, when my grandfath&d4fm father said -he didn't, know whether it: was -cheaper to hawntAv or bury him. People who buried him and di

o~bami ih Downey walks with me an the streets of his neighbor#v4the appearance of being oblivious td almost eerything.
_Ef*(1PP~oaibsme, peers in windows af groery stores and rsao
r" I for his Frmndlf hut he doesdnot seemnespeeialy annsiive
Wtj3b ~ulnmet. hen I1 go with him into a d6ntewa areas,
howeiespecialY to college comnpues he begins toasee and teach eye ~that he ean. We wiB1 be walkingdtogethen and Kdith Vill bom along a building or fened feeling the meals or satoee o can ,He may pack hway at chips of mortar and rub theme between
Ifi O94 ,or hit a pipe to listen to its special -stund. When we visit
_~&I is3 06one of &we. In the nutrition laboratories, for example,
whiftaN of our recent visited took us, a graduate student explained to him ttate of some research on artificial foodenubstances. Keith wMial interested in this work.
Isater, On the trolley going home he told me, '"You see how much e~at4a h for people. That food labors Meesaw there has to be an IM Frd Asla When they get done with 'i work there won'tabeoa
apcmpeso in this country go mg to senda any more. Now the Preal. Ottm1)* that week too. That's the day, madn,1Iwant to see. Come into thh* take adIm4 tell my mom, give me the breakfast pill, mom. She'll h6Mi a for mh e. and I don't have to dcnme home gain till supper,
it can stick my lanch pill in ny peaket too. That mnan at g.the right idea. Never go hungry adanever hara to waste di be ome sitting around at the table hastening to all your baby brothers and sisters screaming in your ear while you're trym to get something to eat. Scientists man. There canbe nobody on the earth do better things than they are. What else did we see there, Tom?"
9-yi~th.adKeith of the medical laboratory we had visited whe
sliiippci ",wrimenans were just getting underway along with soes
oisef heart muscle. The latter required theritse of dogs who
1 rI& Jruid be sacrificed. Keith had wanted to see the insertion of a c"AIixi a maj-or artery. We had watched together, robed in white lahgb~ ,and only once did he wince. The rest of the time, almost onte hoi 7 b-ooodtransfixed, fascinated. Hhere again a graduate student
tesreac tohi, but now, alighting ftomthe ttolley ahd
thed thne blocks to his home, his fantasies began to play on his<
rp~mtaf quired knowledge.
N"IMres what scientists are going to be able to do, maybe even have it dbne ber6fore I die. They'll haive a brain or a heart that's dying. The ]phrson isn't dying, see, ji.ust his brain or his heart is. Or maybe his stodih will be. rotting away. I heard of a lady whose stomach did thf,'eInterrupted himself. "The doctors said it was because she upver ate the Irighft foods. So they cut her stomach out and then she didn't have one. But see, when all those scientists we've been visiting
getdoethey'll just look you over, you know, X-ryyuadwaee the do t ake your heart out and put a new one in."
Thysort of do that now, don't they, Keith!?"

work"I'little Mo*A11st1R)1iM4A#iiL6 keep people from d 1t dRePVA *OarsiOU4,
fes just 1ikB:,C1U4.-,n .,ftzMftV*-?V-.ft"FffWW #1 'JOW put somebody eNa's :ca aMuSMR and f6r
You1mb* -whut I i -s *6idd Jw said Ilia
composed himwlf,, "]Lvisklt eyld bviM4itS',thzt.vu1d ro ket only not into th el kln r, -ki m d of wAr d
cQU4 Ithou
90111g into: thoe;e rockeatsh* a Jn al aw
I wish they'd make cars g,6 Jkd A- M OW
uqp Then I wish they'd inm -*Hs:to fm 3roti vlmwpW & was kftmg his mind wander'over idewhe 1 "JAW
visit to the Institute, almost ho&4,hm Cft:-MT601
-Invention that w&a1dwhoHy.'p1easo1hiM' ..."I got iql he
Ste. P
-1D Aia. YONW
pills and drugs that don't hurLT VL couU &L you, A... 7 i*y
before you even leavwyour ................
pain, man, and yod-Ve."done sbihething -Aw -ei Juv
croes to -the denfiA4 Unit i&11.
"The fives of,: sekntists reaRy iMkd*M-yw,,jAPn1&*y, I WtW 11f,1
"Better believe: i4..nmzL" He MMS aglow -Wiui i NMMt, JMVHWAILMIMT oUr excursion, hswaU as OIL the view, he-getwirow.his Taylor StreeL 441sure WO'uld Elm W be oiw Awieday. V libGratories like, we: inventiJugthin .. thatimooftax 4.110* of maybe get famov* diwovm UWt iWdUld WP,
ria of denied. 4" he..Ai3ikd "help babies gntlbornjigh4 w" strong. .1 Ir
14WQ dift't even *Wy -a, lox ie I
tea&er tells us we gqt hisj,-wbxt t ey c 11. s W.crn out of oil and g&s,.he-sqys, wo-arepn, A X0 or
you kno
tile houses -% W*Y 1'aft M An not, Meg,
b ut heq 6A tS6 14" 1
w -tal a it at sel 7 =000 tEs
day, arejust going tq ve to fino, the-"Worh. F d "'t )MVe 010
of people are go*g to ie. If iou. aen If you don't aet wa e4ot4 h. 47, 1
P61146 ff,
water 'cause'Ws- S u 'Pu'reg0mg tol'ai6
scientists will have seiid ronauts J, zRam
they can find up tbqre,,Qrselxd thum other-, f6W 4 the submarines to 4* or oil. 7. lble ry, hi? CWUM thO secret Of tJW iiA* i6iUthe OMMM Ma t to me ...Coo Y to 'get Qm6 Of
ru c0nd* W'OL
thp- sun Ad I ery da f thei 135.
could tidim I ill I. p 'yon waxm e
hat of the San away M t ev4y
4VO-00io SUR t.hq next da It' would JUA dat
agaiq duripg,.,O*. t bybe a find
roug fi im the, v o
they b I t 6 _pn Wo W 4M 1
out of them that will CU"M -As dau T irl,4,7his head.

-Sinists, you kngw,. have to keep huntin.tsslworjt
likewe aw over there today You have to be patient, whieb is why l'Uu gWb**.e A saientist.'d've got the patience for it. You take a Probla u yon stay with it. Day after day, year after year, if you 1AAW4m: Auyou %denltquit until, you find the answer you're looking.
nfinfd a way to cure somebody of something -, even
oj~qpeson resphody a different, you know, so you need different
king far different folks. You understand 9"
il'Wegot t go bck heesomeday 'real soon, Tom."

t "al. heres ihwaothing else -we have to know about."
!3*nk o Jue,-hat yon- call it, but they do: these experiments on
b~bjO,*d~cU drens real little kids,. you know, just born, to see how thev4rePimtoNbs when they're, adults. What they got in 'them from:
twi 40a~ri mean..
I said.
"YeLh.G Betics." He repeated the word with reverence in his voice ?G q a neoukrw some people got things wrongr with them when
4 and, only scientists can. figure why, they got them, and:
w*w9gt to -do be get rid of those wrong things."
Umd, s of, le hae. lets of not-wr6ng things with 4thi," I
clumily which geihaticials aleo study."
think I'm going to be a geneticist," Keith said as we
re.01d.410comrne of Taylor Street. He was looking up at the rooftoP whrehe and his friends play. Children his age were running onthe street, taking advantage of an unusually warm March Ji"Lv. b t th* took. little notice of them, "I can see your college from rfgh there," he said, pointing to hisf avorite, tower. Suddenly I W va gestf Galileo working alone in his-private laboratory.
4,the tower., h,:Keith?"
uptheme, man. That's where my career in science is going
t q t t 1 o u h e lp ."
'Yo otto keep taking me to those places.- I've get to, get pre-ni got to help me get books and pass all my courses, -which am eU going to get hard in a: few years, like in high school. I'm
eut ig.oyu tor stay there at MIT and, get me mn there with you s ca becyoe aIenit o a pretend one like I am now, but a o g ne,. A4d if I, get sick you got to get me a doctor and get me moneyPV, and get me. all I need."
hiend caln tohi o join the finally made it impossible
fto -US to finish ortalk. WithinoL a few seconds he was running off, elesl making .me promise to take him to. another college in a week oftwoan remining me of. A the things I was gomng to have to do orNI hi n the next. ten or fifteen years --"Illtake it -from there,"1. YW-'houtida~s he'disapeard~it the burger sihop fon the corner of T4 vlbr andl South Plamne. a ou got to'take. care of me, mn, keep

Dtowney nith'sprnsesteraateti o h o loa five floors below lak h oe oke hefwrdaoswr.Ai since the heat is tre f ttna ih n o~undo gi "So ou took my boy to sealthfaoscinsstisfero,

"it was that," Iansweemyfinoffvyeradthmterf six children. I sad down i h iin om stletdeduus:A shook her heda h ette utoto h ouhplos Sin all the rest. Sittigoe hr hr et pe ntep n

don'tkreally et. Maigtinscmliae he elyhtw

over anotoldvrteea'orpaebtal costeJfd'

4mNewhere babies. Three hundred Black babies, ad making
41t,4 tet~am like little animals." She smacked the aewapaper as

eyeeit made. you Sick. They ea&t seem to find Black chil.4rglfW te shools, they cn't find them when they want -odo someP* je them, habmt when they do their exebenfe they find
4*1 i xiht. Three hundred of them! New, you gaing to tell me

aO POUPI cent" Iaid.
_dam -right you can't Not a tsci in this country right us nterested in doing anything for Black folks. Yushw ne
-A zielt* ho's helped poor folks and I'll hiss her hand She turnd A~kkja to la~te-e. "Yea4uget what Imean?' I ftdqd 'You want me to stop -taking him tolMTP
ous:I don't. What would that solve? Wer h slappy with ay
ecan get. But I vould like you to tell me why it is we
V~n't g, aq-iie and doctors and dentists and -thee-right fodoq II 10O14' ik 'on :to tell me how -come they got buildings over there, j~m; their experimenting and we don't have the money to
1 sbuilq ag we live in. That's what, I lie, someonplike
IYOU me.I' like someone to tell me whre thyget the money
tw,0 t all. these scientists and have to aut ou te funds on evry

wn sh~ce peered down onto the street.
,,,.,1qY 4o, if what Keith says is even halfway th truth-and that
Wfmt 6e to me or to his f ather-then the y ot better conditions
Pr ba~ ogs they do .their experiments on t*a we've got.,ferour AiuvnI'You O-know, they actually feed them better, clean them
a care. of them better, and you want to know soehnif
.01 4.gsand mies get sick, thycure them. It' not lile heremWa
14A got here, scientists to 'The hell, we do.,Cil gets
ehe'a wors' f than one o those dqgs. Now, you're supposed
0 scee kind of a doctor or philosopher and I'm just an uneducated a, ncoinwu here from a tiny farm inGeorgia when Iwas
years old. So we ought to be able to agree that I don't know
on.,z do. So suppose you tell me how this society can keep going 1Fh& our clogs on that side of the river are in safer hands than our
-on this side of the river? Yo go aedand aswer that milon for me. That's the way you got tlook atsceeeofi%
,4aoinedat me, stil pacing across the fand oftealllvn
pna.?It fits into the society somehow. Scinwan be soehng so spcial that it doesn't affect the rest of us. Dogs ind chlren are pA of the s Iociety, but the way I look at it, the children right here ,qa.Taylor Street aren'It getting as good a treatment as the animals." Alast she turned to me and stood still.

"Ibame the scientists," she said, seeing that I wasn't about to 'seeaT" bila them all. They're Snpeia.11y ednated. every one of

Js- ikti 66 yeous h 9A=i a f M
manyan time haven we1 goe tht I

cilded him upithisecairse wit s ,- iitayeball folled badk 'Td h g
tael skehow me ich ldren "hav4, dntd e in this .,ho cit ev e m t i theib in-ea ih;o 1 o can'ht ge he oper aton: tfhatsw eedfi allzt ro'tailU4. eanti s et at e mo ey tonme wtheuti ? exper tm .9" : cidhu those stcyroos being hshe w1gnbc fi 1" niato e ounri children abhats gotruhta!YWvgmt howurn hieldrscenof'etists nd wealutj '.r gakh ithisliot ch oler mthn ye odstmer bew-an ih
rsarhe we dntheaybutw oyt ele WhileeadmOTi.W u4 4 ants h mergdelly witot seeing, hem ttyoe windw to thre children. ane ttsli fhtiveala aot thre o tryintideifnshe anlgt orplcmnot6 there'st myc ole thn my est soePlcmnnotr aun helik tantllAreitheyt going toe find son ea ete. hp u goiig bc help rh ahygin topn guarariyt la0A&
widow nan the childrenadthtst!iTfive stre elwIt '1Mi bombs th ro cetsn to thcie mooh n te Shintn hr l
straight~u at me. "I wattempoledw onWher' moon iatthe n ti richig fotng suicke el.that's Arih, she on on sho e ersi h imtm ork onv ouar robm fo The gonx te unte ye:ha c thanto here' hem money froring ouaraeifty search fwo Blp folk, ithe ad of't~ fre li pa fo oks mce, find al hose ciett pigs they go n ohlta!Wt "Ands somthingckelse whicth yor" face pnoZ n straight in theci. "wat abepote yo ki entwlknaon ou thelmon towt them etin kh ind ofol' scientiss orb o sting aell Tht'rishes bussin prorams ad'"id6clsLo b working or no w~org ex. Ar' you is Whr'sar o o gerfo quiet. "Soethintes me yo viefedg are.I~ flma e the sicllte ll i s eethey pokeoo ar"but souehn folsae resph obl focreid e mfto-"Iea ftied withe ar cells abvlclls yourkof scierritE
yo eogt h eti 'kn fsinitto?. eln k l h
0 l !.,C 4
.-eies o.w iv n'o u hirn f!ou nAhai

Stdigalteebsigpormin~ei 10w T1 lvlb

1 1


#4xde asiegra Madirathe ,onesmbo made it sick, and so: ar you all
MikingM~itdo'n yor esearsbh and. gaftg about your bdsi
aesnt caring a damm about. -hither my child hies or asny other
-U=W-po n',sa ehima in some experiment.
J; ..y~ou, thermPete's all finds of experiments with infants, with
rei ith children as old as Keith, that are going on in all
PYou can read about them in the' paper. AN inaeer
Wpys being, experimented. on. We're aas te dogrs. ppl on us, .adhn you leave us to die. But not a word comes
no mwry' She walked toward me and pushes her
yonewpape IM stilhldirrya."Trehude"1 c
1 144v heir babies to those scientists. That means all over the
're experimenting We never get to say as word about it.
W76as~lWe can't be sure thy're telling us everythmg tha gaing
you* dgswaiting for you to ply your rihgames. But
yq 44 ow your white faces around here. You never even say Pi 8oM Pa h~ hpeig Pm sorry that see got our
!!4 jpg on the moon while you Black folks are falling
aiali withenger, and your stomachs re igg. I'me aory
ypur6oyisan epileptic!" We were staring at on another now.
voElthe anger and feelings of betrayal still could sense
whatit Inaively hunted for.
It~~u oi't have to worry, Tom,. I'm not crying about this anyvbeen your slaves so long, it's like nothing's changed. Only
now he aid glancing at the paper, "they call it scence. Everything
of science. But anyway you cut it, you re the masters,
we'r dp, and I just got to wait and see whether a seizure someAav ol akemy boy away from me. And I suppose," she finished
ty,"aay from youto.

"NEw Sives* Pon R. & D."
By Murray L. Weidenbaum
N.: 1: IP.1From The Boosomy, March 1974, p. 11.
A8K~birca turns to development of new domestic energy sources, we Wli~ lia one .thing: that recent changes in government priorities
*1*4, hd'astrildngly negati ve impact on the research and developt e need. That impact is unintentional,-but ver 1eal As long as
ms with a large R. & D. content (notably deese and space)
aso the priority areas, R. & D. had a preferred position in the
I buget.But he shift in recent years to programs with -virrio RA D. content, especially -Social Security and Medicare,
chi~anged all that. Because of its key contribution to domestic wellg s el a t or ntrntinal competitiveness, R. &D. needs a givenn priority in its own right. Otherwise the United States will tinue unwittingly to downgrade the importance of perhaps the most
4tvefocefo ntonal progress.
L~~6 itietlook closely at th arious.mechanisms that can be utsed 4roe, research development and innovation inthe private see-

localities andax and credit subsiies.*
Clearly, we need new types of Governmnt spending Igrs
oriented to R. & D. For example, some Western Europeangoen ments share with private companies the cost of developing, yVdcn and marketing new products and processes. Others offer I aun i aid," fo approve ed projects. That is, they lik''e the -bank9-iatM arise duingithed first year 6r two after a d'esi prodnet i &4 iddI7 These government subsidies can be tecovetid-Inter. The cf61ig bf sutecessfutl exploration or product debidliblnent, thi6e t a df
may reqire ryalty payments equal to some pernts as 6fAes In the other industrialized nations, tfix benefits are thed me~i-I quently used mnethod of encouraging private 'research, de anld innovation. Although the U.S. tax system thrdly'riydsN 'h the scop o f such incentives, it is not devoid of them eithr largest tinentives are at the manufacturing tage i li preitiall and tax etedits for investment in new plant and'equijiA
Only a few ta-x provisions 'are expressly designmedtonc 4 Search and development- such as the dednetibilit~v of eql chased for R. & D. work. Additional taxrincentives "to prom 6 toe gical ininovation have been suggested from time to. time. The sttf qttentif encountered idea is a partial tax. credit'for R. &,D'iiaY the existing investment credit.
Other Government financing devices aste could be 8 11 Although few existing Federal loan programs have I a si
R. & D. content, Federal credit lends, itself to encour aging R. &]Lii product imiovation. As in the case of 'existig small business ptov Washington could work through interiinediaries. TESBICs (Thh Entrepreneurial Small Business Investment Corporations) doWd set uip like the recently organized MIES13ICs (Minority rnerbie Small Business Investment Corporations). MESBICs are design; t encourage holders of private investment funds to make venture ap tal available to minority enterprises, with the Government sharn sonie of the risk.
In view of the continued need to hold down Government spendig attenti on should also be given to mechanisms, which do not afte budget directly. Loan guarantees, for example, .do ,not: regaire #is suitaatial amount of expenditui~e or loss of revenue. The amis trative costs ari, typically small. and more than ampl covered Iy ue
Over the years, the Federal Government has embarked on seal buinss-type enterprises that subsequently have been converte4t privtet ownership, such as Fannv Mae. COMSAT is the rare case.i which Washigton took the lead in setting up a privately owned~c" poration, particularly one designed to use. the fruits of 6overnnio sup-ported technology. Along this line, proposals have been nad lately for setting up a COMSAT to promote development, of4w domestic energy sources.
On a more modest scale, -the Federal' Governmenitmight conse followingy another Western European example -'fosteinog cooperate technical associations .in a given industry. To a limited extent,: suc

unnaintions now exist in the United States. However, antitrust and tax COnsdrations generally require that the results of their R. & D. he put
....... an. thas become arilable to those who did not
fpfit'E& te work. Some changes in legislation might result in greater
on~d othsCOOPertv device, p articularly if there was some Federal
Ay t encourage R. & D. would le to provide that the resulting patknik b6 made available free to althe industrial participants. To assure that the benefits are widely transniitted, other companies could beB Allowed to use them, but only aftdr payment of royalties.
Whatever approach' we take, the Federa1,Government should not 'dn the bulk of the responsibility for determining the areas of 4. be emphasized. 1or should it decide on specific projects and 94mers. It would' be better to use the tax alternative because it 'elies 'Primarily on market forces 'and normal business incentives tp Alledte resources. In this way, R. & D. does not become free, buit merely dhes r than it would be in the absence oif Government aid. There is stil presure on the firm to ake sure that it is getting a favorable return on its R. & D. investment.
But no'singrle mechanic should be relied on exclusively in designing A program to enhance technological innovation-be it tax or other ineebtives. What is needed is' experimentation with each of several
404~es. As in any newr indertakig, there is no substitute for trial id error-and a modicum of good luck.

By Patrick E. Haggerty
. .From Soiene, June 28, 1974, volume 184, p. 1848.
iAll of us, of course, are well aware that there is a strong undercurthkin 11our industrial societies of antisciene, antitechnology, and antiind'nstry, and antieconomie growth attitudes. Yet, when one examines thilneeds of theb. overwhelming maJority of the citizens of this world, it*diffcult to conclude otherwise than that more and better science, tebhhology industry, and econmid growth are taquired..concede that
*Wtave not always been sunficiently donscioutiof the overall quality 69life, but I would argue that only through a vastly improved knowit Odge of 'ourselves, our environment, and our undiverse are we likely tW be fable to attain and sustain an improved quality of life. I weald further argue that economic growth is anything but obsalete and that artch almost universally accepted indides of qualty of life expectancy, infant: mortality, litersey, andyears of edlhooling completed all correI%-stikingly with even sueh an admittedly limited measure of econ'Mic welfare as gross national preduct per capita.
0 76 -

i i i i i i i~ ~ ~ ~~~iiiiiiiiiiii i i i ii~iiii~ n o
.. . .. .................. ...... 4 0
iiiii i M E A SU R E S O F E X4 X X W ZA E

Wrtnin17,eoomssWlimirhusad;aaoT'i atemtdioanwriiicagebiciicificnoiigoth*.it
h avini b e n giilin s iii
ii!!i~~~~~iiliiiii iii===~ ~ ~ ~ ~ ~ ~, ..................................................................
ithaleooitte r wr ta rs ainlpouti o
veygoimaueoieooiielaeihyiosrad J~
arii ,i di ........................... ..... .......... ........... .......... ..........
fo h oeovosdsceace ewe rssntoa ric o

ecnoicwlfre mog terthngthy mutd o riale
fo hesrvcsofcnsmrcaial orliurfo"teio, to
h o u s e h o ld.............w..r..k....a..nd................................ ................................................
ion he onlcedtati heUni tae ma cooicw=

grewiiiii at 1.1 i pecntpr ait eryaroerte 0yer fim,4

195a oprdt icn o e ainlpout hr
MEihsthsien!rwigioeilol!ta nt aioa
....................................i s b e en................................s s.......................
.... aina cut is notjus..myt.......prats....n

iiii lorieii m e a s u re....................................
Unquestionably,= bugoigppltosmk tmr j u l
impoeicnmi elaean uliyoflf.-Ytii s ntbi u
ic~leaiitthiiutiniiiels.Aneierdctoiae fI*l
proucezer- ppuatin gowt wen isuiabeiiiiiiiiiife hs
ibuini tand hentrpouto ae ntelntdSio
roppedii frm17 ereti 90 o12prcn n16 ada si
matedii ....... 9 pecnin17 2. vnwt e epouto aeo
iiiiii l o wi ............................................... g o o n................................................ ....................
slwl fo nter2ir5iyaswieih ug i g itrbtos
whiiiiiiiiiiiiiicis= ..... products ==H H= of~= ............. mo e f rtiiii isi a e .O v r t a
spaniiiiiii= of tiiiii me f oe


*&isAamsin mechanism in our economy that form the gradual
4rom resource-intensive goods to other things, and hence autoWt4ollyworks to reduce requirementsper unit of national output th
r ,,4oe*s.m steadily and gradually wi time. It is true that there are
fects in the market system in that the cost of such disamenities as :W poj)utionwater pollution, noise, and visual pollution are usually jWt fgwwpassed. But surely this is not an irreparable defect and our

t i94Q atWntion can be much more profitably directed to curing these e t the incredibly expensive and self-defeating demand
for:zero economic growth.
eed, for most of the people of the world,. the choice is very clear. Tbay wffl organize their societies to achieve what they feel we in the ,Uai0t..ftates, Europe, and Japan already have. Further, since here in 4" United States, as recently as 1971, about half the families in the limciiiiiiw"&, 'sA incomes of less than $10,000 per year, and almost onefifth had incomes of less than $5,000 per year [4], the overwhelming vax$wfity of. our own citizens will not see themselves as having attained MiAftiente sufficient to accept, other than by force, mechanisms of political and social organization that would limit the future growth of 9M. 630-nomy. sufficiently to prevent their attaining an appreciably i kftlof material welfare. Thus, I would conclude that the need the-Vressures for growth are not only great, but completely
Edward Denison has made extensive studies of the sources of ecoaammi gwwth in the United States and in other countries. He found 4butiiover the period 1950 to 1962, 58 percent of the United States' inexesiswin.hational income per person employed came about as a conse(Vaer": p -he impact of education on
Of im roved knowledge, combining
our labor force -and such applications of knowledge as technological hoopoion. and improved management [5]. Denison's findings are bp",onl the usual nutional income accounts with their admitted deficiencies in measuring economic welfare. Yet, it will almost certainly bi tr6& that knowledge will play an even more significant role in at,ik :.
Wfim9a satisfactory growth rate in true economic welfare, which by its.?VerY wture must be much more complex than attaining satisfactory gri vates in the gross national product.
Until early 1973 there w-as in the Executive, Office of the President i6fiflWUnited States a science advisory function that included the Presu*1onVo Science Adviser, the President's Science Advisory CommitAxwo wwi: tht Office of Science and Technology. In January 1973 t1w(Mce of Science and Technology and the President's Science AdivisaiyCommittee were eliminated, and the remaining functions were 1*91peferred. to the National Science Foundation. Dr. Guy Stever, director of the National Science Foundation and a fellow guest tonight,
*w"'Adoa the PresidentIG Science Adviser. a i i][ muf sure that many of the science qM technology activities of Sig-nificinice. to the Executive Office of the President can be handled as well wrKmn, lbettor with this new mechanism, but I am also convinced that 4"be and, technology are such significant elements in our overall culture and so vital to economic welfare that the complete elimination of tU::Yeieb66 advisory mechani-sm from the Execuiti've Office of the Pres,i&,)qt would appear to have been unwise.

I certainly would niot suggest restoraion oftheldf lnAiewb 4Wdr mechanism. Instead, I would advocate an approah whibh -,31 shle science and technolot and the forces of lknowe& iitA ad diffusion more intimatl in the. policy-mking 'eiite ty Executive Office of the Presid* ** After extensive -debate on the econonly idllowing-W6,4 f
Congress enacted the Employment Act of 1940 *whidli tdd Annual Economic Report of the President, the Coduncil of' 146 Advisers, and the Joint Economic Cominittee ofth'dg
I suggest that there would be -no. more fitting wa~y to telehaoth th 20th birthday of this nation than with a new, act,, the: "Ntia l velopment Act of 1976,"' as a natural evolution from that'Eldpoped Act of 1946. ***P
As T enrviin it+ this nw National. Thrmlopmen+ 'A of swalA~w that it is the coninin polic an responsibility. .4 of thA~ government:
(1) To seek for every citizen an ever improving standard at, livi
(2) To encourage all practicable means to foster: and. prmy" rf competitive enterprise to fulfill needs of our citizens, for go a services.
(3) To use government intervention, but with caution -n noi abanding, to modify the market economy, to affect the pric~pa of goods and services so they reflect the value of such Puiblic gev
society (such as for health or safety) is concerned. .
(4) To require modes of federal, iaterventionthat will ave-id*nnient of workers. .
(5) To use all practicable means consistent with its needjs aWat gations to assure that there will be useful em-ployment o p4Wx~ including self-employment, for; those able, wilig,anmdL teeking
-and to promote maximum: employmentt: production, an&5d mn power. *.
(6) To conduct its affairs and interventionsaoas to providd4Wal and growing -economy with a minimum inelination to infatmi 4
.(7) To encourage a broad enalar gement,of educational o0f "R' with emphasis on equal opportunities~for @,ll men and wome hl& out life,- including especially com bined ,work-learning progra4aie at consistently upgrading the skills of workers everywhoAk their broad cultural betterment.
(8) To foster the growth of knowledge throughout the'e
fields including science and techology, art, and the human44 particular emphasis on, those basic area, vital ,to, the contindoe nomic growth and social development of thxe.United StatesAnl~i1 the use of research and development as key, tools. in attain ingobjectives.f7ikk
The National: Development Act: also would call forv the R sui ttnioteCnrs@i aur.,feLh:Ya-antpa e4
ipetrprrveig h vrl ult flf i hscuty

ion notonlytheivtesat teoi vernmn lino suhusa
ii~~~h ia rs acta proud alo crete natjomal pde ctbu vriey o sgh mebessrypcs of thq ueandt ofg h if as U.111. iprovd eveopment radeortnal andk rconurl t
.... I wuldesabseveral concit of thonr essetd vises; itheExecto adfnce o the Polesentablmised fiv o~mr~tit ecnomic,deaten sic, woud sechopgan t pr program nd a )tvitel opmtent Act of 1976ian igt on hplc dbdba-d b actThe aaeuld tiso coutre as ito teeomn obnuite mdem ofegh quality of ite (ii)t n poetgp oftb House o allre eiates tor m ve othein stdvfmters~ MAC tw'the nahisnaleandlutilieoth nomous cpab nd& 6M:,a findngs technoerl cmtrof the apCatires eln
Abeliv tat t e aicdearge tihe w redcompand dhbrea r~~~ianona suc a this National Development c f17 n t ncmn ther~mteral fflenet quaingol, i) o aont appche

t~tisaat~win scToi nd "IshnGrowythrObsolete aincaoni re

lhmgge a mecan V)c (stsNational ueuDfeonom t Resea o
knblede bcom anintgoxas Inrumntnin Adance Eooicyakin

-IL V".'Wbrdilies and persons, inGro th e .s.te" in conumecGot Tunr er e 9 73) Seie P-0,N..7
31 tima b Tantrues erdPvtarce Econerice Pling e stitution ,Wshngon D.C. 1967)

J. orrste, Wrh Dr.aic WimhO.-Ae, Presride, s,17)
*"y icom offaeliephondeLrorie S., nCnue n
t. f omere, Fuebu 1973, 1974s-, volue13. 68.

A Convesatio ism, sa Wiarl, oldresulet, wel

fkcd'wth uinus competition, tried to squeezesom
laboras ttriger evouin. (One, erious flaw in that aronh

course w as th at... it co m pletely..v.rlo.k....h..i mp...................
d e v el op m e nt,~ii w h ich..........e..e...................................
andiii iiiii newiii meth ds o man facureio kep al patiesreaonaby hap7
As wera h a-n rdcin fmutn nlto

downtown and rising unemployme~~n o h ..i 94 ewnee
wehdalfrotn htbsclso

Measured in constant eFdea ovrmnthscu
onil ree rhs e d n in reen.....n...vte...sty asbo l
hditson avw llfrote htreerhwruh
Weptta usin oD.Wlim0 akr h e edo
Telehon Laoaoistentonslretan otpesiim e
....... an deeomn a.W ea b sighsi:eci,
ator Henry Jako'i(e.iah.ieetcalfraie ah"
P r o ec toii~ b rainsto rm............ a .......... o utiii ............ the""iii e nergy......... crisi ................................... D 15 )

"SntrJako asaget n civ at i ceceadfh
n o l o... ...... ... ............o...... ... ... ...............................................................
we~iii subscib to ........ thssta neteti eerhaddvlpe
ha enfr o opatclry nteidutilcmp n.W
hae hnol ,0 eerhlaso n idi hseuty|n

we coud suport a leas teniime asmnpras10tmia

Plies t and .Oeofomie pgresais. ell the'e ntech nobgyisto~o heasnpouiest A s a coseqnc, thben huige tago abod,10 seiesnce, and thechnolms have bec omewmuhtouean we need another tioarc fforftha kind; tuhtf i nedtob eadthafu he Gvrmntsbiisdida "No, Idont tinkust," sand make ary oo eaere'gernismt J. over the" Bero akdner8 eledythe compsaneveie c eraly o omphtsitepamtarcthy aoud ap~tfoad oInuflear ey resarc tore godoinsomea ardeas.Bnomayb thy'ress doin the rigtt
They~e aton prettly comphtictive Asarinces, why shul thue fgve bwuthererhf science, adtcnlgha esewhrthr
asmanto reoremeut hnyud h hne atig
le'sv-Wodois thedathivatintryohasobecome dinhat Ws goingokedd har for anyd, dneer tracied oftcnogyial
Iffdont ist, collutivi countries. e' e g id Myeo govehmt res t nlog" Bas aeplind ofh reigompn.iey cetil b en don.,-th. estoc they old afnodedgedof scin9 fand thgitneeri
fro-rn lin r, for. xap, rita t he s i eoy aecntly toe autocompneve earne u hevow eat to make i Ththiams ceurt and trnitot foreKhrusohptie' pownfall whyd shauldwhy Aorrval~tuseofhuan esores laingfotsi whud the-theynll doa th
The trubneai theare riae into y en ebehindsnhned 1
Nowastoyres fonveortn yvers ag teheooia tur bildilles of

P~ ~i work~ iecnol re ndosan rluinare voyage a u reol adrsomt ey sowe i ik the cha'koldgfirnc [who reieeang. eneate power byn pastn anxncluhruha ak f iizfr example, v tc l inE eir Mass.m is d qut or the~rip h theso stac garsels fowrfplan. Anywtats s pran Ha~~ nd ther dare plan dos w.theyati themel ay thingt6gtpfesomie knhw1nuwnfed of magicoic advce In othe.hasve or veyeuar pgoher tsure, il kls re esinm nt juthreckattorhg-ehooyinutislk

trunkcs. We were Jamged, almost ordered, to: giwnay ;,g-Ao it of our technology in the postwar. period uardds the nu8~.,~" Europe and Japan. We're being threatenedvith tlset as that is, the Government is trying tb give 1esi andolr~ v countries a great deal of our technology. We'e this." .V
What about Japan? V t
We've been startled by the scareity of innovationth,.asoi4 of that country,", replied Baker. ''Oh, we food a wmieitth one having tinds, with palladium alloys-perhape ofwai~ things that look pretty -good-but gee,, the total output e So what should we do ? Forget about crash progrM -d -2kt feuded industrial research? Lot the economfita nstht (o of money and people and pray for great scientists? "Yes, mn a way," said Baker. "I suppose it comes bacse l~pb that cerebral energy is much more f ragile than Irineti I 4
fragile, it fact, that we don't know how to assure its plent

.Russia as,;40
in Besea

Research Sp ending U IE T,
(coat 1 958 dollea) 30



.iiiiiHi .

i0i |z

SOSOURCE Naioa Sieenc Foundati11

SORE ainlSineFudto

Interview with Sir Bernard Iovel
tri .S. Nte asnd World Report, December 1, 1975, voue 79, p. 53]
Qwgton.Sir Bernard, some people feel modern scieintific research
to e a curse because of the undesirable impact on our lives.
yi is a blessing. What is your concliuaioni
A~ser.1:think the answer is both: Modern science is a blessingcer example of this duplicity is the hs of the rocket. Th evlpd-o alike purposes in World War II, although, of CQ~m already had a long history. Since that time, the rocket has N014 f a ar of the military arsenals of the world. ,
ro also is used to launch scientific instruments into space,
g',enhanced. our laiowledge of the universae in a very remarkk w. i relatively minor changes, a rocket today can either
Carr wareadsto destroy mankind or scientific instruments for the inv~tiaton of the remote parts of the universe. !%i L*li cear example in the physical sciences, but other scientists,
ithat exception, can give similar example of tis thin dividWlne ben the good and evil results of science.
teo tier hand, there are more- fundamental isse in modern s itferesearch which I believe to be altogether a besn.Many disn s ae presented us with very great intleua chalezs For
P I 1y own field of astronomy we are now fced wihprobtowrilfigthe nature and origin of the universe as great as man
11~ -Y feve, as I do, that the attempt to understand our environmj~n mi our place in the universe is one of tha basic driving forces in diikdvacement of civilization, then these aspectsaofmodernascience are ltoehr ggod and, indeed, are an essential pareof man's progress.
V "to~,Yet you said recently that "the simple belief in automatic 6dehiYogress by mean of scientific discovery and application is a "PC mth of our age." Exactly what do you meaa? Answe. In the last two decades, the governments of the great powers fifvl ith ,ersuaded that a massive financial investment in what is elee"will automatically produce goods and machines which
nr vlable to the state in a purely material sense. That is, they will
4.itiityimprove the economy, and generally hoostrthe power
(th ~t i civil and miltrwas
Thi isonsense. ...4.
Itsaveryv odd thing, but in many ways we are entfrino from the
-ii bbs6Vane of science to the Allies in the second World War. So 0Aceud kentists emer~d from that conflict very high 'regarded by 1 6 Ovrnments. Scientists certainly did play an invaluable role in .Ad t* te dommhon enemy, But this led on -the peculiar belief that
po fne 'uenoggh scientists and enougmoe ito the developdInn Of Science you can solve any problem.
Thfact that, sciene in the war was integrated in a, massive and
-rzlhiantl organized military operation was disregarded,,-and the nee-

essay social and orgammzationeal'Tramew~orkgs av iwmen a have been able to solve postwar problems did notimrg. You cannot invest i n science and expect tog get --etsatof icfy,
even in fundamental research and certainly not in a efhological aspects. Science is not a magic wand you can wtand 4tit poor country into a rich one. As soon as. yo try, a host ote -r, lems arise which have nothing to do with scienceancn by scientists or the methods of science. i.W;1
Question. Has such an attempt been madeI
Answer. Of course-look at the Soviet Union and theft Iil of t1ie grain harvests.
But perhaps the clearest example is here in the United ME~~. 1963, Harold Wilson [then -the leader of the Labor' Pair(YIa4 Prime Minister] spoke of the "new" Britain that would eo i
the "white heat" of a technological r-evohution to be initig~ socialist party gained control.
iiii~i~iii miiii. f t i .i

A year later they did gain power and created'a Munstry o %-1
ogy. This department commanded some of the best. scienile ofai mn Britain. But as soon as it began to apply modern concepir oo and -technology to solving certain industrial problems, there w 'A47 5sive social reactions and labor problems to which neither the t ptsR nor the Ministry ha~d answers.
The point isithat science and technology is only'oqps h
an essential one---of the process of improving society. rp; besgM an expensive cog in a very big and -complex iMachin,1e.ig toe political body to realize this has led to a regrettable reacr nait science and scientists f rom which the Western world is stt pg.
Question. What form is this reaction talking 2
Answer. Well, the most obvious example is9 the seve 1U j Abudgets for scientific research in a number of countries. I n the United Kingdom, for instance., a des-nerate situate -s h*en worsened by the economic squeeze in the, universities, whit i Il turning them into teaching factories instead Of Centel~ excellence and fundamental research. Idn the United Sta-tes, -a fierce reaction ocetirred a for the pl shnient of -the near scientific and technological mirac e of landinog the moon and getting them back' again.A The erosion of the faith shd hopeswhich ingp'red theqppw ad the rapidity with which the -buddets '*re Fashedse emerge as one of the great ironies of the tweniethr ent xxy, Question. What lies behind this kind of re action?
Answer. Quite simply, th& explosion of them th that4
sole all ]problems given sixficient resourdedi and toi .elz4P -- b politicians, thht -the promises nide in the Lbie4 Siei 110 years ago have -not beentlrealized. Someties: tis hsbeda reasons; more often, because, of a failure to understand, ta miace is only one part, of a, fir greater and more complex social prob- Question. IS there some truth in the view that science.a cete as Many problems as it has solved ? .
Answer., That. is a, very disputable. point. I thin we a -g~c i4 an
epoch which renders it extremely difficult to make any fina menj"t.
In many aspects of research--and I am thinking partiulrly of
those whose technological anplication may lad to th& conafination
iiiiii..... ... ..... ........


afth#- ftvironment and the atmosphere-I would say the situation noads watching with some anxiety. Many of these contemporary devel -fflks can be of immense benefit to mankind, but can also be
very harmful either through neglect or sheer accident.
qJ111owever, when I think of my childhood and the state of the world
....... ...themahd the facilities which I had as I grew up compared with those
my, and grandchildren enjoy now, it is undeniable that so far
science and technology have been of immense benefit to the world.
H: WImtherthis situation will continue is a critical question. The pre
V ftrious balance of power in the world is maintained by a weapon whose
V.c&AivAi*1 for destruction defies the imaginatiem. I find it impossible to
eaft"pe a 1eeling of deep anxiety.
A.4 Qvwtim. the recent move by genetic engineers to apply selfimj)ased, con*rols to their research point one way ahead? Should other
stientificidisciplines follow this lead?
4.&AAAswer.1 hesitate to give opinions outside my own profession. But vh6&:the..:-dangers of research can be so clearly enunciated as they are
in this case, I do think other scientists would do well to take note.
,Tha: txouble is that it's not always so easy to pinpoint the dangers
,Uzmoagh in., advance.- Take atomic weapons, for example. One would 1&!MJM6.?t0 stop the researches into atomic structure half a century ag&Ntither Ernest Rutherford nor any other responsible scientist at
thxtIim foresaw any practical outcome of these researches.
_:KVvestion, In view of these dilemmas, should governments assume
closer control of scientific research?
fk2.Ajaswer. Almost without people realizing it, science has already become very highly centralized.
Thisk&k.-A happened in the Soviet Union-which, incidentally, I have
4st xtvisited after a 12-year gap. Here, you see the most-immense .... ... ... . 'in the- sciences and in the consequences of science because
Sovipt science has been concentrated on specific issues. Some of their
arcji, particularly in the astronautics and aeronautics field, is absoIptely.Axot class. Other areas are not so good. I am simultaneously ImprPssed and alarmed by thegreat power which this kind of direction
9f science gives to a state.
subtle and perhaps more gentlemanly sen the same
ppWss is evident in the Western world because science is now being coniralized in. government financed organizations. It's an extremely
W.A.1t. situation, arising because the high cost of research leads to
114W. and for public accountability and centralized control.
1. W ou so concerned about this trend? Doesn't it
OiL Why are
p*lw research more co4-effective?
Answer. On, the contrary., I fear that growing centralization and
co0s. of science lead to inefficiency,
*jt: is becornhig exceedingly difficult, for instance, to do entirely new jqnc 0 Centralizailon means you have to have a massive sup
portive paperwork organization. It means the best brains often are
wrapped up in doing an enormous amount of administrative work.
''t' es it seems one, must know in advance what scientific results
is goingto-get before one can try to obtain the large amounts of
money necess for modern science. The-state is reluctant to allow
individuals or universities to control the monev. So there has to be a series of intermediate organizations responsi6le to the central gov-


ernment. That's the difficulty: The freedom which is an essential prorequisite of the best research disappears as the budgets increase.
Question. Should scientists themselves take the responsibility. for the emphasis and goal of their work?
Answer. Yes. But the scientist today can only work through the complex machinery of government. The power of this machine is so very great that it can change the individual instead of the individual changing it.
Question. And this is largely because modern scientific 'research is so CostlyAnswer. Yes, I think that is so. We have evolved into an agewhere. a large majority of scientists believe that in order to do.any worthwhile research they require multibillion- dollar equipment. They often do. But at the same time there are dangers, because equipmentKlike this inevitably directs one along a certain line of research. One always tends to concentrate on areas that stretch the technique and available equipment to the limits. There is therefore a tendency to neglect other areas which may be important.
The classic example occurred in 1967 in my own profession with the discovery of the pulsar [a celestial source of pulsating !radio waves]. These neutron stars were not discovered-by those of-us-with large and expensive telescopes and computers. They were discovered accidentally by a girl research student studying a, pen-and-ink chart produced by a -relatively cheap radio telescope, which was &sig ned, and in use, for an entirely different purpose.
Question. Is there any way that science can put the genie of the atomic bomb back in the bottle, so to speak?
Answer. I think it would be fatal to attempt to do that. I think if you tried you would stifle the advance of civilization. It may be that civilization can be destroyed because of the applications of science, but that is an entirely different matter.
One shouldn't try to halt the advance of knowledge, whetherIVA be in science or any other discipline. The search for knowledge and for a deeper understanding of man and his place in the cosmos is a vital necessity for human advancement.
Question. Would you say that we are in danger of learning -more than is "good" for us?
Answer. Not in any intellectual: sense. I think that one of the most extraordinary things about the discoveries -of modern science is that We have not achieved a completely scientific description of theViiiverse. On the contrary these investigations of the distant regions of time and space have revealed great difficulties which seem to confront the I aws of physics with problems which may be insoluble.
Many eminent scientists today are puzzled by this problem ., Indeed, it may be asked whether there is a limit to scientific explanation because of the involvement of man himself in the object of his iniO stigation.
Question. Is that the only line of inquiry?
Answer. No. If you believe that science can eventually explain eVerything then you search for solutions which may lie beyond our present understanding of fundamental physical theory.

3 9 ..................... ................... i iiiii ii

.....................................................v ietiscio olio f
theory.i For exmleiiiacnwedethirolmabuiiicocp

oftebgnig fteuies frm tt o niiedniy!adiin

fiieia iesos u hybliv hs niiisaeitoue
byaniadqac f hsia teryw ich the ar tri toove
Dosti enta o ietsiei sts' are.......... ... les concerned
WiIisr fmrldlma htwryyu
...A vower. I don't think they recog nize their existence. W e all owe ani~ii~~i~i~~i~iiii~~iii~~ii~~~ii~~ii~~iiii; i :ii~~iiiiiii~i!i~~iii~ ;; ~i~i~i~~i~i~ii
Oleia ce te tat i w ihwi ve .......... work I':~~~ ~ ~ ~ ~ iia iiiiliil iiiiiiin an crtcim Itsjs c htfws ienists-iii
~m ori= W e t o r ove the mo a iiiiii iii= iiiiiiiiiiiiiiilit e of........... th i work.iii T h m ajoritiy.....................
re ~acenc asa rofssin b w ic they==== eanter iig
Qw~m Haeiyuicangdiyurii n abutthimpcto siec
f.scoya o hv ere oe

ind 1930s seemed a a m
pf~bal on a en dcae uigdaai hsso

he ,16, when the list join eaomfk

...................... o n e r
of Technology and the American Enomic soitontaAKa the annual meeting of the latter in WashinigtoDC stecar of the. initial session, I remarked in my opnnaeeitmta technology and economics had been living in omtlwmr~ since the days of Adam Smith, although, euioul, ahao h
union seemed honor bound not to menion h ae fe obri public. Alfred Marshall and J. M. Keynes, .............
economists of their respective times, both poone hom-o economic society that began with the prenitha h uniy # quality of capital-that is, ts technology-wer ob ae.a gwu and changeless, while the greatest teehaph lmnd otf cn tury-its practicing engineers or its philosophies ce their machines or propounded their explamain ftednm4.o science and technology as if the economic miiuwr .ie'":n changeless. Yet in fact everyone knew thatthsigemt naic
force within the economic system was its clugn rote.of. e cal possibilities and that the single most impratnfuew n h rate and direction of change of that frontierwstecnsricAMn posed by the economic relationships of thesytmHecitws'4 time for the illicit, relationship to be brought note pnad ejt. mized. Whether or not they would settlepon n iehpiyee after, I remarked, it was certaily tune that ecnmc naejnjg got married.
The theme of the initial joint session was 'TeEooi ad* nological Implicatigas of 3 Percent Giowth."Wsacniuq)f growth along the historic trajectory of the-pspoibefrAltr century ?" asked the organizer of the initialseio..Teiasr"t such a question, he pointed out, depends in pato ogrn]eaon ships among the economic variables of thsytmRaerhn attempt to cover the huge complexities of eacsieoth usox their short papers, he suggested that the- spoemn o ehoo concentrate on the problem of "running out oftcnlgclipts while the spokesman for economics should pa tetoIo h.PSI. bility of runningg out of consumer demand.."
Michael Michaelis (Arthur D. Little Co.) apoce h is us tion boldly. There was, he stated, little or no rao obleeta h supply of technical stimulus would decline oe h uue ned Michaelis went so far as to declare that "we no a vo hwo t get, all the technical knowledge needed to atsyfrceb h tt wants." The problem, he suggested, was nottegnitolo e
knowledge but its application-bringing the ital liiale.1 pabilities of scientific and technical know-how o- aro tesca n economic framework. Continued growth akogtehsoi ahwl
hine. therefore, on new breakthroughs in scineadtchooyls

t h e c o n t in u e d o r a c c e le r a t e d t r a n s la t io n o f s t e a d ily a c c u m u la t- i ~iiiiiiiiii iiii~iii iiiiiiiiiiiiiiiii
............................................ k n w e g in....... .. u se fu fo r m .iiiii~ ,,,,i~
M ich a e lisiiiiiiiiiiiiiiiiii ''" to o k th e p oiti ha t su c aiiiiiii" stea dyiiii~i~iiiiiiii~iiiiiiii a pi catiiiiiii iiiiiiiiiii i oniiiiiiiiiiiiiiiiiiiii o f tech-ii i iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii nolog.lwoldiioiiak iiiiiiii spontaneous butiiiiii~i woul requi thimr
....w deiert adoption ofa ysem-aayssppoa in whii i chi thei
IdeeneI c......c-anitrdcto....hnlgycarcersi

of akteooyip-cdbaplneaddsiedproh
whig...... saetfcadtcnooia eore remrhldfo h
i n m gy................................... .. .................................... a d

in..~~~~.... tha! readtelogdanou.n.ahaad"outo"t. h

Ioi~rbe fkeigcohscea ouinta a eutdi

un ve iiiiiiiiiiiiii un o-r ina e advancesiiiiiiiliiiiiiiii iniiiiiiii textiiiiiiiiii ile techniquesiandilaundr
an rcenn.tcnloy ial ocliat ntenwcneto

tiew then arises ato whether the at~dv osmpindHCo of this amount (in terms of .todas pie)pesns; contradictions
Villard admitted that the composiino osmto tt li
world of generalized wealth might peroc ifrfo hto v& of skewed wealth. "If 1 percent of befpoutohsfltm*n*.# it [will] obviously be impossible--soto nulkeyr-niwrn of the steer or an immensely wasteful rdcino:s" the average consumer with file nign the rise of real incomes ofconsumaeqieartfo nrdsrbtion of income among consumers, act ocag h opstm4 the "market basket" of goodsenjoyed b hfrelyih This difficultyis not, howevar,a crucialeone....
i yhologia pro s a
whether mass. consumption at the poetdlvl sfail~.T~ hinges, Villard believed, on the typeofcnupi tash&0, ciety wishes to enjoy. That is, some knso osmtoAsg~ by Villard as "neutral," do not affectteevrnetwo xop
the kinds of food ILeat (although ifalcosmret bstmi* of food, there may be a substantialeniomta fct ersf different fac iinputs required to Other kinds ot consumption, however,..x.............................
verse"3 environmental effeets-for eampeh clgcl i-ato the automobile, the airplane, etcee w ail oesp & tAp taint inelasticities, especially those ofshespctaaned ou serious asoial impasses.
The problem, in a word, ,is that nihoho:efet edt-m
with the sqtuare, andnt the absolutenme fteppltiu d by extension, by an exponential funcino helvlo use.' Nonetheless, Villard did notfidtepolmu angw.,
provided that we "dev ote a significanpatothicrsen vR come to the reduction of the adverserprusosi:usin concluded, however, with a pessimistiwangthths agrneeclusion does not apply to the world tlre hr a nedn
population crisis does raise problemofpentyumn .pa dimensions.
Discussing Villard's paper, Ben S. Slga Uiest fMsa
chusetts) emphasized strongly the,-ptnilydsutv xei4
ties of uncontrolled consumption (strsigth amlefc)!Tw ever, Seligman focused his attentionmanyothprdcie flr
than the consumptive side of the grot xrplto n one
to the potentially dislocative effectsofterdecinf now away from the highly mechanized seorinwchtholg te
toward the low-productivity service aes hsmyrsl nafi~of incomes to match productivity, a es o ag etoso h
population whose ability to maintainlingsadrswlbeacringly impaired. Seligman seemed to nis adl-etrcoM arising from the uneven entry of tehia.Porswt s
perhaps shrinking, nucleus of. advancdtcnlg.adrih

xSee W. J. Baumol, "Macroeconomics'of UnaacdIrwh""mrcnEooi Review" (June 1967). p. 424.


iftky dy lower-paid occupations, and he saw in this duality serious
_-M continued steady expansion of output.
tgnmAtbWtup all papers, Victor Fuchs (National Bureau of Eco-*"U6 Dmarch) f6und cause for cheer in the avoidance by all partWP"t9-. of the extremes to which long-run projections are proneMutielAiffol:66fence fiction abundance on the one hand and of resource #khaustitk or vast technological displacement on the other. The prob1&=: ot v #tinued Fyowth can be encompassed without recourse to such apboAly-ptic visions, lie suggested. And yet a crucial problem remains in the values for which growth is attained and to which it is nut. The essential questions to which a study of long-term trends
-are therefore the social, politic4 and moral aspects of exisitenew rather than purely technical obstacles of either science or esonemm proper. -The real impact of continued economic growth and technological advance, Fuchs maintained, will be found in such 6M.
as as the obsolescence of the nation-state, the family, and the value sym&ms of society and in the need to discover new viable forms of institutional structure and spiritual belief. : Inevitably. in papers of the brevity enforced by the occasion, probIms are easier to identify than to explore. The session served well to hi' Effilight many issues connected with long-term growth, in parr the*n4d for tho social and political changes required to produce ined tekwh ieal change and to absorb it. Indeed, if any sinale point of agteeVion't emerged without prior consultation, it was the need for P67 puWic (or ublic-private) social arrangements, both to guide 't,6C6ol6gY and to restrain the private uses of output..In a ver real J y
the.n,' both 'from the technological and economic sides, a con
effike cq e problem of Social adaptation and social change emerges w qT ptral theme of this first attempt to extrapolate the parameters
"Yet? 4her questions remain7. to be dealt with, hopefully, by future contributors to joint SHOT-AEA meetings. Three of them are suggested herewith:
: L. The tremendous expansion of scientific and technological activity of i o&rn times hasnot been accompanied by a change in the over-aTl rht ,bf gro Wth, at least in America since the Civil War. This sugges s ,that the relation between scientific advance and economic expansion is 166ri romplex '. than has yet been recognized. If an "exponential" growth of science has resulted in no more than a continuance of the .,given higtorical rate of advance, this suggests that future advance will require an impossible rate of scientific application--or that economic 9i s on something other than scientific progress.
2 The increase in the standard of living brings not only problems of 4*ternalitieg but changes in the motivational foundation of the market bystam.-][n5r "ed wealth per capital seems likelyto effect the acquisitive (maximizing) behavior on which the behavior of the system is
-iMiamed to rest, whether in terms of lowerinLr the marginal utility of th*dirrot or padding the marginal disutility (;f thestick.
the problem of how social routine and vested inter lesta-can be overcome W cope with new priorities and problem th t find their roots in the socially disruptive effects of technology and a thanged economic environment &like.

"A ay join conieraon berwnial possltsanceqa~i xxxa Ecnsi-ome h nwan /aonwirrecoraint Panau oante ouma capblt Coorm atr NAome Sc o hat a du
within ~B Lh xsigntoko eair l lran soil(n .ooiw

hequsion tof hate iscnowny.T afit h TAE relations bercand dslpent andt it irtrcooaissnprwe thodulrel.w standng.rBoth t remsr asdbdselopent cpomms t li ewp An relativ s rall ne m ofoiit have cnducstanig thReVub je, bE iini s i iibout the i-0 The Ns Funt As Montining inOWte/Pretigiii theEsu bjc both

funder ~B Leofr reerhaLdad1 ern mancywt

eoi ruesearcof ha n kon bu the relationship betwen reserc sacadeeoment and economic groitth/prodcii dic yit.

string idSetemeh90 the Oeslicc and Manelo mentmn andmeb6 NSF to put oetnhae bee quic rewi he sbject fogmeie Ati requtielst a ntomter loin qesonmstshv:odce eerho
the (a Whajet doinin weino about the relationship betioa~c'oc
ecndonmi c ot/rouiiet intesbetbt a ifca
(b)e ow gooachad s h curre ent ateofte-arw th n t rSine
polc Areniwles nApoitingy to F m ake any sre judgmel t whnmi eeareh Unte Statsi d er vrnestirh ndge
lyn fronh economic growth/productivity apc

(nmd)-Whtemight be70 the nextc oMgical t ind :ugtse
(a td eknowledgeuo the relationship soeaseto educe th
This is o an o efor te add resubstatey --ato knowledge ct1
butr rather a moetpfoton eie whate isnownt boncrnn
aspe fthe suec nid wate unexistin owlnedigi & ue
lFrt y f ocuigatnino the economic effhprdcivt sects?
not imlinWhat only the et oi&Darep impourtheigo Ono de rry increatid hi knoweg is clruea ofe greaoug absove an beod it stric ativ el tnoi benisowegvfhesbet buti rather 'i ........... m o es eff ort................................ to'- .............. w h t i n w bo t t e e o o i
asp ct of thei @ su b jectiii~iii~ii~~ andii~ w h a theiiiiiii ex isting~i~ii~iiiii k n o w edg "a d d s u p, to."iiii iiiiiiii iiiii iii iiiiiiiiiiiiiiiiii iiiiii.. i iiiiii i ~~ iiiiiiiiiii iiii~ i~ ,, = iii~iiiiiiiiiii .. ...

Oni the cotay increase knwldg is clalyoigetmorac above and beyond iiiiiii~ i iiiii i ts tri ctly ec no i benefits........... .................................


Vwa.:. ia this: paper was to confine our attention strictly to economic matters. Second, by looking at our Nation's rate of economic growth
* nd productivity increase, we are not assuming explicitly or implicitly, th"'1.qqWQ.Mic.gTowth is, in some simple sense-, what public policy should.,,Attipmpt to -maximize. Clearly, the desirability of a particular ,r,9"9 th rate depends onthe way it is achieved, how the extra producis distributed, how growth is measured, and man other things." (Mansfield)
-T wcqmplish the purpose set forth above, the NSF commissioned MAWS oi the subject from leading researchers in this field. The papers t4aijollpw are the output of this effort, and are necessarily limited hy thppoppp described, as well as by constraints which each contributor faced- terms of his own time and the short deadline placed upon him. NSF is indebted to the contributors for producing this material under sjWh circumstances and asks that the reader view the output with an understanding of the scope and limitations described. As a followup tothe mparation and limited distribution of this material NSF plans a, symposium during which readers and the authors can exchange on subject. The preliminary. papers contained herein (perhaps
)., ,"d proceedings of the symposium will then be considered fpTp4b. cation.
The first paper, by Professor Stewart, is A Sumwwry of the Stateof-t7w Art on the Relationship Between R&D and Economic Growthl Prat zativity. It descriibes the methodologies employed, the estimates d4rii bd, and. the results of research on the subj ect. Professor Stewart's papW previews the most pertinent literature reporting research dealing ivkh the total U.S. economy (macro), the industry level, and the firm level- micro). It seeks to answer the first question: "What do we know about the relationship between R&D and economic growth/ produetivityi"
Professor kansfield's paper on The Conhibution of Re8earch and Dovedopment to Economic Growth in the United States begins with a discussion of t-he second question: "How good is the current state-of he-art:on the subject?" including a discussion of the "Fundamental ProbUrns of Measurement" and an "Evaluation of the Evidence." He
id his vi ws on the third question (Are we in a position to make
I. judgments concerning whether the United States is under or overimkui- In R&D purely from th-e economic growth/productivity aspect?) and presents ideas on the fourth question (What might be the next logical steps in furthering our knowledge of the relationship seas to reduce the area of judgment?)
)V s- Generating Sector8
_,rofessor Fellner's paper entitled The Progre8
CWm. to H qh P7siority discusses the questions raised from the perspective of his research utilizing the residual methodology in estimatIng the macro economic contribution of progress-generating activities. He summarizes the results he has achieved which were most recently
orted in'his Presidential address dehvered at the eighty-second in .ing d the American Economic Association on December 29, 1960. Professor Fellner's paper discusses: the problem estimates of average

rates of return, sho~R~w lau ates reAMith19 dtt
fudig and cannfeMEonev. The forh contribution, by Prfso Gr lis A'Ai
on Research- and rdeM. 1rofessor OirIchb disuene t for a positive return 'fom O R&D, fi & nft 'the^ tl questions, some hunches and iWeas. camemi the allo (X sources, and suggestions for resadch topic&io
The following is a saxmnary of te Views of the: cet. *th regard -t the questions th et nder the Mtdp "vad Puti The sumnunary with regad to ee -quesion its organigd flte briefly the major overriding, conclusion, andit kd to presen from the papers which p-rvide support for, and seim9) major points inetlding the range of Views:a
()What do we know about the relAtionship between R&D tAd nomic growth/productivity 1:m
Although what we know about the relay tionship? beth n't&D a&R economic growth/productivty is'limited,. allaalbeevaaegi cates that R&D is an imotn otribultr thoeconomi r productivity. Research to atesekn to measure thidl (at the level of the firm, the industry, and the whole eoiay in a single direction-the contribution of R&D to economick growth/ productivity is positive, significant. and high, ..4

Various studies have attempted toWestablish different thigh wir they have examnd the correlation between R&D. tim serion (obt)lk aggreated over a number of years) and time series of prodnd I* crease (or productivity increase ovranme fyeare). tive and significant correlaton has usually boen fouznd,:,L_. $Os ha ve calculated rates of return and, marginal rates of return. on invesh-A ment in R&D. These have usually been: quite high, but they varyJ widely. (Stewart) ,***it is clear that the current state-of-the-art in. this a is not strong enough to permit definitive estimation of %these relatineekipar Nonetheless, although the reults, are. subject to onsiderableeor,a establish certain broad conclusions. Ig prIua xsing eooer studies do provide reasonably persuasive evidence thaR pa g nificant effect on the rate of productivity increase in. thindsWJep time periods that have been studied.1 (Mnfil) ..

***these studies rely on the results of svea 6o6nometrio inv tigations that indicate that, for the industries, an d'Sieds indetitatigation, the marginal rate of return fromn nvestmnti n rsevee development has been very high (Mansfied).

All reas onable, way of loointg at the iha~r feaf o ec I* that the rates of return are very high* asPoqpelt' us.
ofrts f-etr n ia fru'1n Fefe 7

4 7liiii~ iiiiliiiiiiiiiiiiiii

.............................................. h a c l e a r iiii~ iiii l e e n o n e ilii~i!!'
of::& %, ajorsou ces fgr wthin output...e.........................
4 g.. b e e n a g o o d in v e s t m e n t b o t h in t h e s e n s e t h a t it y ie l d e d a p o s it i ve

* a o r tu n a n d ~ iiiii= i ~ iiiii,,,i=ii th sen se~i ~ iii th a tii= th ,iiis ra te iiii iiii i,,, o f, ii retu rn h a s b ee n a s g o od== ii i~ii = z =!ii~ i~! ~~~ii~!iiiiii~ ~iiiii "iiiii ...
jj4.A etrta tert frtr note rvt ngbi
The......... evidence for, theseiiiii statementsiii@ lii ,i tered m u i ti ~~iiiiii~~~i~~i
but it is still quite strong. (Griliches)i

................................................... ................................... b e
unatpdi em fwa ee s ne td egidvda no
...... fost eim ,t e n usr = the economy y........ as a wh le), wh t is -igiiiiiii
.............................. c o n t riiiiiiiii i iiiibu o t o e o o iiiiiiiiiiiiiiiiiiiiiniaiiiiiiiiiiiiiiiiiiiii
t. teprdctviyofa idiiua ndstyorfrm, n wa
M ethodology is em ployed in m akin the m measure ent (e g., direct iiiiiiiiiiiiiiiiiiiiiiiiiiiii
iino e t i m e a su re s a g g re g a te r e sig u a l m e a s u re s ) A c co r d in g ly the
p s r e a d t h e f u l l c o n t e x t in w h ic h s u c h e s t i m a t e s a r e p ro d e d.

While there are differences cons ming .ths tdsanguafm vt search Aindins these seem to aetthe degree of aofdnb l tributors place on the estimates rather tha the dAim n magnitude of the estimates The current state-of-4 tnttrm enough to permit definitive estimates of the enMrbasin O economic growth/productivity and results se sulbjee io error. Nonetheless, the results cited in the'-papbr eat e" Fb authors to the conclusions stated above.

By the standards of the scientific community one eano i assertive about sofi the to be eM ticular about the underallenation of resources PM*eggW Ar iti activities. The conclusions will be formulated actfdinl.00* ever, in most practical decision problems it is Ilsea ~o rf
materials here surveyed are strong enougii to be; considered to indications on the basis of which -act wriis indee iln ir 'j real-life situations. (Tellner)

In conclusion, technological change Ims; cetainly C01nnbt. i ah very important way to economic grk vth in the UniteStft eAl though existing studies have not been able'to esndmate thi -Ct~l tion witli great -aecuracy, theye have certaily '9iiftt 'Afti cohtribution has beeil large. Moreover, although _econonilcs%& of the relationship between R&D and productivity increaeM vi~~i subject to many limitations, they provide reasonabIV Pa iv d ee that R&D has an impotrant. effect on prodivty ai the industries and time periods that ave Ieen studied. RMansiO

The research results reported above are aubjet to a number fra Nortant limitations of at least two major kinds: f J::
1. Limitations concerning the specific finding and gpoqi O
odologies employed. Examples include: the dif~eicutyqg g from contributions of technological change to the -sec :tw tribution of R&D; the problem of estimatin the time lag ewet
R&D investment and economic impact difclyinR&D expenditures for price changes. Thee are bs b''
simultaneously with the findings, arid the r6u erewospr,
above (a) should enable the reader to do this.
2. Limitations that are more p-neral and sifect ime
of the work to date.' Examnples mcluae:: diictt of i
one factor (i.e., R&D) Jfrom the: complex, interaction of t
contributing to economic growth/productivity; measu that inadequately reflect quality ch anges; positive'qcorrelati
do not necessarily imply causation. The following excerpt
the papers summarize this second kind of limitation.
The state-of-the-art is least satisfactor inmasr0g(i
ship between R&D and productivity gain. this is true for t wo rhi
1.RD(r cnmc eore dvtdtoteavn'ee.nt fpo

........ ......... i i ini iiii
(X h act rs i rod u c ivity g ain w....................................... istr.........ed ...............
IIIiiniliiili= and manager i pro res andiiiiiiiiiiiiiiiiiiiiiiiii economliiiiiii i es of s ale ai
j~~1............ Bttercnrbtoisomreeilyqunifiedta
9CR&D Te majoriificut seems........nt........ence
Firk,~ ~ ~ =N= ;; thei~i~i mesre aesoigotioiutu hchtee si

mae ebsdsfe rmavr ipratdfcpriual o

tdesets ot natiol productivity nee.a~T3y W

themselves to the private nonfarm ecooy te'to4dttW private market economy; others inchuegvrmetan1l.V*0w son for excluding government in somes are of government output, and therefrofpdutvy"haftA a result studies including governmaetcuti tcsOoa ,uar stating productivity gain, for the econm 'a n hl.L 3*j*V% assume a productivity gamn ia governmnteulothtm O 6r
sectors (not all of which have measuree otu n hrfmifs ured productivity change).(Stewart)

Technological progress has harmfulaswlasbnfclb and the harmful byproducts also exp:estesle ag. market results which may be regardedal eaiennlk V.aus These negative items evade rough nuriaaprsl Me,0U
thumbnand this is partly bease it isn are asociated with the technological rges isviosPaes (Fellner)
The rate 6f utlization of plant and as a percentage of capacity. This resultsi ylclv.rain ai tal (and total fator)' productivity! Bu fteeiE towars dfuller utilization of capacity, totfcto rdc will be slightly higher than it would hv enfra'oe.o utilization. (Stewart)
(c) Are we in a position to make any jdmnscntnn h
the United States is under or overinvesigi & uet < lt economic growth/productivity aspect The authors agree that, based upon h vdne odjdmn would lead to the conclusion that the 1..i rbbyud~k" in civilian sector R&D from a purelyecnmcgot/rutity point of view. However, nothing can i sion, as to where particular R&D invetmnsholbeadV4 this judgment means is that there is g addressed to this question seem to concuewihvk d
confidence, that we ma e nerf R&D in. the civilian sector of the economy n h siaeivga, rates of return from certain types of civlinr.each4 evlpq seem very hig& (LMansfield)
Reasonable estimates lead to -the conlsit :6tursae high in the American economy. A groodraecnterfr efi

igenasing the weight of phagrbas-generating inputs in the economy as ap h~.0 late we have been moviag more in the opposite direction.
Meak dronomists if queried would assert that there is underinvestm t a research by private Arms because much of its product is not a muable(appropriablel) by the private Airm. (Griliches) The Arst proposition is that, bhecanse the results of research are often afH direct value to the sponsoring Airm but of great value to other
Asi useare io good reason to believe that, let to its own devices, the awakat 'would allocate too few researnes to RAD-and that the shortfallie 14ul be particularly great at the more basic end of the R&D
:The second proposition is that, .because research and development is isky for the individual firm, there is good reason to believe that the maniet, left to its, awn devices, would allocate too few resources to BUAOA Of course, the risk to the individual investor in R&D is retter than the risk to seeiety,.since the results of the R&D may be useful to .someone else, not to himself, and he may be unable to obtain 1onwthe ser the full value of the information. Because the economic apatesa bee limited and~imperfeet ways to shift risks, there would be as adewinvstmmentin R&D) For this reason too, one would expect undernaestment to he"e'aest at the more basic end of the R&D speeto Maneseld) T
What miht be the next logical steps in furthering our knowledg ofawelaionship so as to reuce the area of judgmentI
.0"Contributors agree that many of the limitations discussed in the p~pw are a result of the relatively small amount Of research attont ianse to this subject. No one is satisfied with the aurrant state-f t6war and::priorities in this respect are suggested in the papers.

"80sexc, IN-verwrON AND Ecoxoxxo GROWTH"

By Nathan Rosenberg'a
[From The Econoate Journes, March 1974, velume 84, p. 90]

N$too many years ago meet economists were content to treat the proess of technological change as an exagenous8 variable. Tedhnological shange-ad the underlying body of growing scientific knowledge upon which it drew-was regarded as moving along according to aertain''intemnal processes or,1as of its wn in any case independently-9f economic forces. Initermittently, tehoopflchanges were introduced and adopted in economic activity, at which point the econonuc consequenes of inventive activity were regarded as interesting andI important-both for the contribution to long-term econonuc
growth aud to short-term cyclical instability. Schumpeter, for example, saw the engine of capitalist development as riding in this innovative process in the long run, and at the same time he developed a business cycle theory which centered upon the manner in which the capitalist economy absorbs and digests its innovations. In Schumpeter's model, x The author is grateful to Professors s. Engermnan, W. B. Reddaway and E. Smolensky, and to an anonymous referee for their helpful comments on earlier drafts of this paperThey are, however, accorded the usual absolution for all remaining deficiencies.

exgeou ecnooiclicanesstmlaedivetmniLira
th a ia i n in t bf y x
I niiiiii'' the...e.............he..e......W....................o m ist's
grad all changed ........ T h vast..................... exp nd tur s o ......... an ...........i
ment .t.. ven.i.eacti.ity.......
ecoi c neds i suc nees reeive suficint, inanialsupprt)
Clal mc f h eac ctvt o ndDws ihi'prosv:
bus;iness' fim.........o e tcniusin pc~ aegre
ofpoutteipn uhmne pnti erhadh-vf
sei me igl scesfl Smlalgoenmn aecesW- 01

Wh" 18&e"hat lawsa cge the 'gowth of mdas a xgnu
Pfter og taton veywihqutiny sstemnflticalycnoi
f~r,'it1questcoin:()g to greplain smthvrins ihcnbn two expaney atculronom str foe tIe, fatr nd (h) hov n do ttd citth fi iveainteaiity betweng industriesaddao
1,186hothe.ll Scookler'obvioundyent answerin to theecnms h~oa e atesm toin u inventive activity w esnithl tenmcpe n~i136anht can herefoequithl chnesto in temof h rftfflbr anay iasoipaated wi the risi perap Ioe ati

man bptraofeconoI m i rowth. ilaremn'ih uhta
he hastrcal oreofr Schmookler's bookyis in attemptd o ug sideat ionst are the mtajrdeterint olfutvaristtepst ~Al aftie eois of inventive efopctiity ndustris Inelatinhpt

a 6kery, uthmooklineres oun ao dioe orepnenncentneo hi book "Waiaw oen the purchase of ralodeupmtans conseyoe
a fty lHgge incrrses n nvetemp atvty sl meauldigbok
rgu~teas-it idca that t bisvariti n sytemasall stdyn it" indetos 1 owt xli the variations in inventive eo cm
er" inhipstinulingtyoe ie and (2roeu reinng aoepli dtV~ftn asts ontheie indus twe n ha indate gve
6pdne WorlmWar Schmookler findaerya high coret uston citl gos invatenpt tin forua industry andvtywt the lumee o
Jkf*Thiu is noajsshthrthoci nt iind terchnloica ipheres anote ornthog thp at hasoen af uchodkerbatd is Evn iftoemp were exaple that the std real isevan Autnmu ere inteed no t nd i that sheraren unretbehe maot eirectl ovinflenceith varials but theynmayebeffoving subecitoia indteroics or eamiin th
feswh ichn e idrhentfie roandte sd, byuipenmis co poets, toaxpai slgtylge nrae inventive activity a esrdb e
pateift on such items. The lag is highysgiiatbcue cmo

ler Burgus~it indicate thatit is vaiioin te al ofeqipen

whciinuete aitin ninetv efr. cmole id

siia eainhisi uligan erlu einnatog h

p"go& nvni r nids

lers onclueiv thwe cagrat inveaslla feqiis 4,am atviy. Jst nlige i aase profiaiiyodnvnin ta ofdureyeue and drcts the esresadtlnscorgy P

nietdeisiwe etermanasef invetiv ainofivetv SoFa lr bent on aemptnos toial asmsIcnpitb ealieconomied-nwkctivit hich tb og t a m wheming importnce, was noalrdy i yeooimoev,,cm~, gard socupsde htw cnsieat inins jutliery. atusiny disussng cnnsume ooducio nn oluptP t it isvethes aned cotndtme demaneto aiiesm -~vi~gnderm eca aaysinanniv ofivt shib intsrecio
"* *cnoi f(I)mwork Wioth ht ar ivee t
lin tehiepogr, aledeintioaetdmn ~esi capita income ill gosratially ris iely suodntIfdpm income int cusng dinfere claos ofenios Scmorarusta At s dter casen cosumodemaecomer iewiha h p beforeat fsifsi the yieldio to inventive effort
chng correspondingly.atAndien pon ftm1ihr eOt ias inuefebrsectivse yielod tieadalwcpW c wail sift.m Thus, evenuaundere. xtr sqenet ~ root~ ino pecwnts disfermncaetl ofixoswl as rdaly.h sucsiveen lsecios of thats strcure reaieymre1~rat ingre the red strucvture offro nt: feetfilswl tn,.t thaei efortspacordingly. This is wl1ute rat htivetvWTr cenuyce but gavesmuciv moreld tt edrinoenti 'etv requ sieet.hs, fe liunre y cetg asm~inta hes0ow Schmoker'nisrment y s ed presi(pores wbr subjsi ectin o thehft bjection thato lyoe iiteey ~w dend simpl ignodsrste whnfoein t netr n rcael thi Schmoklr dcraw dteimply.icati from, hisx mpe meia toods inentortat ". .oo piretieuctioitefrtpr:o tenbtet staetr th atte nuohe terms ate io in per centeh etuyo I eureentncreas re by oaialgodreti otinpcuerdofeelin Schmookler' s important o notente tha Schwol' e~ t thbe ct tohat ou fatal goodsientionthis rhlmn mpai ormthe i lyeta disciples rom wh lte in utofmdrvsine n h 4 Schmookler op.w ct, pp180-1onOfr cousedt n ne-nuty aitosI fot
expendiettir ae nn atiuaclsso goods is,.P 4.E o Schfaors as age stprctr tof aopopuatiSbon kearsls" .dpn I Axo waion will also, nola anrdimnot n utr role. anfet~ P XWpl

.4optin uely, suggests that at least some of the initiative in
,*q. hng~ng atterns of inventive activity lies on the supply side and
O~nvtedenand side where Schamokler has placed it.
,,&A~klrhes anticipated this objection, and his answer is in fact p~ta~as o. He argues that the e noity olasses towards which
rs ircttheir efforts are determined by expectations concern,AaacW.,yoffs which, in turn, are shaped by the. familiar conofdemand and maarket size. Developments on the side of
scieo an teknology are highly relevant to the inventive process, but onl d teining the teelnieal realms--mechanical, electrical, chemiea, ioogca-upon which the inventor will draw. While the growth in~kowldgeat the scientific and technological levels will thus influ....~~~~ iVIteprM8frw ih
Iha~h spcfi haracteristice of inventions, teproe o hc
................undertaken will depend upon the state of the market for
:61ws f fnalcommodities.
T h, pintisthat, while eanmarketable improvement in envelope-making euiptettis probably aboit as easy to make as one in glass maky eqaser today to make an improvement in either field via
;aletrnic m n than through some mechanical change . If differAnces~xist n ichness of the different inventive potentials of the Trodct tchnoogies of different supplying industries, the pressure to Jpqme. n idustry's prodiuation technology tends to be met by the var~ja ofreltively more new products in supplying industries with 'F lber prouctinventive potentials. For example, if new electrical
Piachjuesareeser to invent than are non electrical machines, then the aggmat~ deand for new machinery tends to induce relatively more eleptrcal taon electrical machinery inventions. In brief, inventors
ktoseluA he most efficient means for achieving their ends, and at
,any.ove moet, some means are more efficient than others.'
Sehu lerthus argues for the primacy of demand side consider
fonnot y~sggesting that shifts on the supply side have been unimP000, Qitethe contrary. Science and technology have, brought
transformation an's capacity to pursue his material
Jjt i p peisel~y because, of the -versatility of man's enlarged
jnY~prX.,pfsentificand technical skills that demand side forces rex.P:,014y noglh te, scienceeandtechnoa p ay~p a subordinate role M iflencngthe erection of inventive activity within Schmookler's PO~ss, ot ecause his analysis downgrades their historicalI signifiMN btyater because he regards science and technology in the modvrn ageaig, in a significant sense, omnicompetent. Schmookler bodyof oder scenc and technology as constituting
puty clay" out of which almost anything can be shaped. As
nhe sates mankind today possesses, and for some time has posIt-pUrpose knowledge base. We are, and evidently for
som ti-iehae been, able to extend the technological frontier percepilvay~a vitudly all points."1
ow ti precisely the aspect of Schmookler's argument which
veem to e mst inadequate. If Schmookler is right, then economists

71bd. p.21. mnhasis Rohmocklerks


need not pay too much attention to the internal histories and-s'&66VOrlds of the sciences and technologies in order to understand the direcfiql bf inventive activity. If he is right, then science and technology haVe Abt f unctioned as major independent forces in shaping the timing and: the direction of the inventive process. If economic forces can so powerfWty shape, not only technology, but science as well, in the achievement of its own ends, then these subjects retain little interest for the emnbnAA or economic historian.8 On the other hand, if Schmookler is wf"* ih this respect, then his analysis needs to be supplemented by a mm* clovful examination of the manner in which the state of knowledge at: aty time shapes and structures the possibilities for inventive activitye,

To establishthe independent importance of supplyside consider&tions, it is necessary to demonstrate several things: (1) That ftlew and technology progress, in some measure, along linesdeterminW either by internal logic, degree of complexity or at least in response to forces independent of economic need; (2) that this sequence in -tum imposes significantconstraints or presents uniqueopportunities wbich materially shape the direction and the timing of the inventive procem; and (3) that, as a result, the costs of invention differ in different industries.
As soon as one speaks of the "costs of invention" it is necessary to recognize that the economic analysis of inventive activity is seriously handicapped by our present inability to specify the production fune-tion for inventive activity with any pretence of precision. Inventions, unfortunately, do not come in units'of equal size, whether considated from the point of view of their usefulness or their costs of production. Both the inputs and the outputs in the production of inventiCA me appallingly difficult to measure. Schmookler's basic unit of memrement is, in fact, not an "invention" but a "patent" which servarks a surrogate f or an invention. Schmookler's primary interest is minating the process through which society allocates resou-rcest6inventive activity. The extreme heterogeneity whichis the essente:'Of inventive output is, Schmookler believes, less serious a problem fbi his interests than it would be in an attempt to link up the number of -Mi- Vehtions with the larger phenomena of technological progress'stild, 060nomic grow-th.9 Schmookler appears content to regard inventi#er output
fl "Thus, Independently of the motives of scientists themselves and with due- recognitlen of the fact that anticipated practical uses of scientific discoverips still unmade, are often vague, It seems reasonable to suggest-without taking joy in the suggestfft-thitt the demand for science (and, of course, engineering) Is and for a long time has-been,,dedTad largely from the demand for conventional economic goods. Without the expectation, IAcreasingly confirmed by experience, of 'useful' applications, those branches of setence loiad engineering that have grown the most In modern times and have contributed t dramatically to technological change-electricity, electronics, chemistry and nucleon=. would have grown far less than they have, If this view is approximately correct then even if we choose to regard the demand for new knowledge for its own Sake 48 ; noneconomic phenomenon, the growth of modern science and engineering Is still priniartly a part of the economic process."' Ibid., p. 177.
'I See ibid., chapter 2, for a searching examination of the problenis Involved In using patent statistics as a surrogate for invention,% and also for Schmooklerg Justification for his belief that the deficiencies in the patent data and the problems posed by vast oualitative differences In inventions are less than Is generally supposed. For a careful discusslon of the measurewent problems involved In the economics of Inventive aetivity, see Simon Kuznets, "Inventive Activity: Problems of Definition and Measurement," in R. R. Nelson (ed.), The Rate and Direction of Inventive Activity, Princeton, 1962, pp. 19-43.

4 ately. measured by the mere number ofiventions since, i is s to note, he is not attempting a directlink-hup bewen the
"I process and the larger question of the historic growth ih
productivity His results, he is cefutl to poli ot,
tyf to .the number of inventions made, not to their imht.. One of the problems of research now is to esighllish the a-4Sthe 6onection between the number of inventions in a fidld and
lie rat oftcnological progress." 1o Within his framework the at. quipt th compare a unit Of invention in one industry with a unit of
in another industry (or even two dsixsi h a
aust'ry)' is obviously fraught with difichk-yi, Schmookler is content b~liva the prosp'8@ve value of inventive output is likely to be
te in industries undertaking large amounts of invesent than thdustries where such investment is smaller. An industry's volume 69-investment activity, in other words, is the pimiary determinant'of
profitabilityy of a unit of invention.
'This leaves us-very much in the dark hn attempting to attach a i g significance to a unit of invention. It would l most convenient, fdr analytical Purposes, if there were an identifiable unit of invention
-whic lowere-d the cost of production in a plant by, say, 1%. This would enaable us to assess the importance of a unit of invention by relating lfttothe ie or to the rate of growth of theadopting industry. Unfortdately, the ex treme heterogeneity of inventitve output simply does ib4'allow us to 'assume any simple relationship between the number bf idiventions and the number of such units of inventing or* productivit) of pddisliklyto induce about as much inetion as the same sum spront~dA. a'y other good. Hence, doubling th aount spent on one kind
of ~o isliely: to induce about as Muchk ZMention as the same eum gehbsan other'good." This position rasssrious dificuities to whichw il shortly return.
Although Schmookler's treatment of the ielationsliip between de'mahd -tes and invention is, in general, highly illuhininig, his conb~ptfi Apparatus even here contains some disturbing gaps. This is tp~a hhn'by states that, "From a broader point of view, demand induces the inventions that satisfy it." 13 One wishes to rush in at one 1-*ith qualiAfictions: some demand induces the inventions that satisfy
13 Bt which, and when!? As soon as these questions are raised we are koipeiled to consider the different rates at which separate branches Alide ahave Progressed. Many important categories of human wants ha~e lonig Cone either unsatisfied or very badly catered for in spite of 'kitl-estalished demand. It is certain true that the progress made
BuehindoMer, Op.- eit., P. 163. See also p. 208, footnote 1.
nIt Is, of course tantologically true to say, as Schinookler does, that "A given perentae revement in productivity Is more valuable in a large than in a small Industry." Iba.,
inha n tv@ ahosi Be'hmanarm's. See alan nnD 290 andl 212.

intchniques of navigation i-n the sixteenth and seveateot owe much to the great demand for such techniqaies--m ahsi as any authors have pointed out, But it is: also true tildemand existed in the same period for improvemn igarts generally,, but that no such improvements wea Teessential explanation is t hat the state of methou-ic n o afforded a useful and. reliable. knowledge baseor nv imrovements, whereas medicine at that.. time had' no bs resin medicine had to await the development of thel i e aeilogy in the second half of the nineteenth cents:.A~6W flof medicine was one which attracted great intersomior sus of money, and large numbers of. iseen-tificahytai;,p~ leical progress was very small until the great brealsiugso _a$ terand Lster. Improvements in the treatments. of infciuaies abholtely required progress in a highly specific discipln-,atA1
---y-and the main thrust of medical. "mnventions"inth .Oe ll.dred years would be dilicult to conceive withouti.b ne tN hi1hy doubtful that, with the single exception vaccnto gis 11al-pox, medical progress was responsible for any sg q.
tiutioix to the decline in human mortality before thetltAcn tuy14
'he point at issue here is one of general importance toShoW i arument. The role of demand side forces is of limieexlntr vale unless one is capable of defining and identifyig te wee ety of the evidence that the demand was satisfied.Itwudotre qie a very lively imagination, as the references to eklrooi sugest, to compile an extensive list of "high priority uinne whch existed for many centuries;, which would hv qsiu hig1y profitable commercial activities, but which ye :aie..ui ased. Schmookler's formulation is such that it is cpbe fitd to almost any conceivable set of historical. obserainq ijw arument to be non-tautological, however, it would hv o 1f &ae in such a way that the component element, of kndpi idtified independently, of our observations concernipmetvatvt. Until this is done it is diffoult-to conceive of anyst fl ain which could directly refute'Schmookler's hypotes.I 4, .&eneof a. reasonably clear, independent specfiaticso O o IM toof demand, one can never demonstrate either thetipr4nk= poents of demand have gone unsatisfied or that su S 4 0r plyed an important role in having down the time, paltro V atvity.
In fact, the argument of this pae staif we sAtawi historical sequence in which different ctegorisofv isae bee satisfied via the inventive proces s, we must pay ls fei 1 > seeial supply side variable: the growing stock of usflk1o~* Hitorical evidence confirms tha inventions, are rarelN&Y'V9? blin all commodity classes. The state of the variousn inss maes some inventions easier (i* e., cheaper) and othr adiL moe costly). In considering the manner in which thes1 *ofcen
P4 his Is the judgment recently delivered by medical historians.SeThmsM:0 ad R. G.Bon, "Medadln Evidence Related to linw1oh: Paran angente11 Recrd, "Reasons for the Decline of Mlortality in Egland and WaeDuigh'1t Cnury," Populatiom Rtudies, 1962, no 94--122.

gthn and the manntri which this got a
Od~the poor'inventive activity, oneb
*b~ ofnature contains many sub-realms, whic, vr
theii* relattive com1PlexitY. If one considers thebr a
wdai tgress over the past, 3Q0 or 400 years, tetmn
46 ~dfthe 'growth of knowledge in these separate dsilns
D4 IWIb ,AI bo th,6 relative complexity of each-as wel eatth
66W tyof Ihe teckitelogy upon which scientific reseatch'in the dwi Ie.'dp~ha. rFor example, -tht microbial world and to a gt ta
Tt 6 M60gital 'world eould not be, examined without the asst
"a hbroidscope, and the contemporary study of the atoic Iii C&htt mnolecules awaited the 'techique of X-ray cryd MIlb .o a On the.- other hand, it. ip not surprising that the diseipl1nes
Mb19 the most advanced state ih antiquity werbe stnv idl~i~di;thmechanics and optics. These *ere each disciplines' 41ed1ao the evidence of unassisted human ob
little. or no reliance upon complex instruments oexej~ft .d%1 40 retis?* 5This, a masedrj of the principles underlying the'M h~ji*1 wrld wa's attained long befoes a similar mastery was 061%*U 10i the principles of chemeitry---almnost 200 year, if we 0 h-akkhddteb the publication of Newton'IsPrineile' on 8~~ &*h&dAt MlendelejofI~s periodic table on the elements on the 0 Afly, within the disephine of chemistry itself progress was
10 ati tha is 5r i chmiy. Even though it had I Ue~liipirtthat there were huge-seenomic benefits to be reaped t~~~ 4At vegetable and aniftdal worlds fromn a greater Inckle cAehemhistry, sithbnowledge persistently lagged behind thg L "Miewledge et inorganic chemistry. Organic chemistry I :111W, thtraceaitl -and unresponsive to an obvious andicompel&WK Even ,after it had become apaet that all organie subst"&sw a r posed, of :small numbers of elements--mainly carbon, h #MO~e)&Xgenand nitrogen-wedence quite simply remained baffled, at-61ft~t~i10of the organic, world.i Progress in organic chemisy, w~lWl bao, Myed far, behind inorganic chemistry. because of a bfi6V~ndtinyielding datum of the!in-atural world: the far greater
0, &tractefal eemplexity atf organic molecules?" Siailar osidet~gadedis rh a broad range of, research activities arnd go far to~kg #Xplining the timing with- which commercial marketable.
9 UdWredeed from such activities., Thus, the molecula r struatu re, rOmllhini-B-1, essential 'in .the treatment of pernicious assemia, is m~dh Oe complex than vitamin Bi or C and, as a results it took
Siflyi the comparative lateness of the organic chemist's successf ul isvut upon the structure of protein molecules is largely attributable,
kroto their great comp~lexity. Amorphous, materials, as a
T 1h, heStruicture of BHefegtifto ReVOlutionsf, Chicago, 1962. Chapter VII1 Abtheauthor'A article, "The History of Science," in the, Internatioawl iEcy0 of the Stat Sciences.
g q'Zrea t ti pe t he nt h, ce ntur bratrugsi rganic chemistry in turn laid 'the
the su bet tweriteth entutf revautton in biology. As Bernal points s b4 'baicte isr ad As-oth#ressental part to:play in the httory of
fitra t' to oafuller rangers nfg.of biological 'Processes. In fact, the g bdfty d# er understandings than the ateroscope could provide was totaly bleP .1ithopt. knowledge of the laws of combination and the types of structure'
at a..x.a fi nr ..,agtmi ia imrtna andonsvntary anlonment Of

Thou ahle Iubelierecopie-n ther la reqlii told and arle wie hrid vqired a-ihg7r'.e~ effotrpoisenasrntaodlyPtotally inxpge teatetf a4 ~, een asl ap complteiliy tomu fortl P**.;uQP resaran asts.n -Aoug tlpgah ecaosi fo)l,'a mbly cdompjor trle dinginl uthee cossigQ
ra 7tm.Thsfnigges along its componentx Iu-iil invmntie effctivter hedia pra me a tkwesver~ou t
electicalbiolgica-is~eepl eeteli isnhintecaeo diablieeuia in the extremeety visaAseJ suhaanedra orces o choolerMc cetAD-a
order?"~a GiveM thhseue-)n h d adihly ci ome scmty clneases, w feecmqx t*K Fraiemore, althug Shoole7 are oaingl mrltirposey-el vnowedhyokd pnow exageratie thenextenty total whichou sprt h cnmc l
itene of, prseninge mthodhehs whittd~ics nP" w pth sc s ubtitutio isfequntlyc nore dmoisha.Jwn* imyprfet amoroe in man case detb o:o c tfpV
raeaonis compleentarbdipitys athea: inreativtwentit vrtpsw century transformatio alecprdcto biologica.1seel knowled i thehitrof U0a* herdifit wich the rxtree t psu hw4pasible t.a more ecntly, nore wodhea e nd rice arie .to aderistic ofv thseqescence. intevenopmnto naOthegra in heren comoleity clafe biologi ca airtai ID aineresing l bsrv tion: purpo[Tbesae kodeg e to urdsofli ofaggeif e issmthin ofxaentowirel difere rd erlso ko l anibeat ofplaonts and floes.tsf or n hefr gieaneoiso uan of* apin the igh of urntg wi e o n wledge re a r.s tiuo o'pp n grneeitrna compeiht in tenal comple n~tvP~es o elutionaryhistory through idt living org an.usiuio.I u preset taete Ib ntd.y vo ,p .t ensormtion. th wrdarcluri ag4,, aordsuce ofro the ls cople to ,the m&t Y OfterolP0,;0 pheeysil ni vers s l arly erse in-osil toedereo.14ieyne j 'h m ora brief but highwYeaerceptlieetr atmetie f -udmeW

illeeiam N. arerft~s .Eonom ic salomen la~o fl i ext feiph lmifl
gamerntrn ral C anpl e Ohatobnerna 196L npl. .1 tefL osquneo h,1f
evi onr itr hog he iigogwss iv.rie te~gle 6'hi

iiiiiiiiii15 6 iiii

iiiii mentiiii! an....aXC a g, O-oe 96. p 4

iMfithC chem60 Iinputi. Indeed' theinew high-yedn to pg
nieiy introdied intoisoutheast Asia sae often anoore
6thish the'traditional varieties if they are'gro under te
olalthlfnl 89i efrop and soil management. Their unique fae is
a 1i Af~eedf 'fertiliser- ss onsivene~ssbrought about by genei
4f it much bstrname than miragee" rice wonld ae
Astiisive.1 There:9are no miraces. I~ntfact, the sharp i4, I 0pl per acre, which superfie allysgetnaie
in redouirce prdciiaeraly the result of lar-ge mniliser and other chemical inputs combined with rigorous;
atte~on. llikques of water managehentt?*rthus, these biologyi ,!C, ~ion require, for,their success, large -dses of chemical inputs;
p.he one hand, and pesticides to protect them from the
rp~y,'-tst9which they are peculiarly vuinerable, on the other. In
ai of agricultural tichnologyte ndnohtaeas
1h Omnant relationships are those tof eopplementarity and
ji~i sitltin. nhthis respect, therefore, orfireedom of choicein
different realms of science and technology for ways of
X 1, i I 64pfidutcion is largely illusory. The range within which ,74 .PYL o e i getinine options in the achievem ent df specific goals
erely attenuated.

we9fo8 from th6 realm of sience to that of tecnology,. we
el er d;where economics motives es nach more direct namedid"tr'seat. Sinoe technological conerens Eare dealt, with pri4 ~ JEl Cfmfisthtrix of profit-geeking business firms, one woald
xP to find, as one does, a high degree of responsiveness to tondiF fiet Hmandand profit expecatiaibgeneraly. Bat here
0612 idanitly clear that an,'understanding of demnd forces alone at onlyve ls-I,,ry liited insight into thed irection ad the timingstf 94fi Je akel ity. Here, too, differences in the inherent complexity at
hfi o' 6flblo giedl level shed a flood of light o te inventive proese as nlius6engured in historical time If this. is correct, then the SchrhookPstiotithat technological problems witl e solved 1(one way or
-ithr when the demand for such a solution is sufficiently presming
&6ftaible) is seriously intcomplete, and needs to be supplemented
Y, arehlscrutiny of supply side variables.
.qC~d~ibr one of the central events of the industrial revolution: the fdb shtttln of a minerst fuel for: wood in industrial activities. The
of we and the desirabilitel
Pk llineremisgly clear in Great Britain as early as the second half Wt& sitthenth dontury, during which time the price of firewood rase fa more rapidly than prices generally. By 1600 the growing prebsure Ifo the limited supplies of firewood and timber had already-proCA num ,erous atlem sf to intro duce co al into individual industries.
yin spite Of strong Ad prevasive economic -inducements, i
verk 20 yearsh before thssbtitultion was reasonably opee
hat particularly interesting from our present'vantage point
-su, mn som industries, the, transition to the new fuel was effected
S0-bCompleleity and costliness of Water* management methods In the arrowivR of 14tP Winkjor Treason why the new wheat varieties have Po often been intrelued more "illand with greater: succese that the new rice varieties. This has been the case 'fo exa ple, In India.
20"eknow from experience In the U.S. that the rapid introduction and widepread 119 Ofsw crop varieties accelerates the biological dynamics of crop disease-ost plant

very rtly intrdain otheicungsa ote-4:' suraing lug sa of pig irea that theuirans coplte? cmlt nwrwud eln n o- ~uf &
Ths th tmin of anutis who eie A rae ingr dereres ofd copext ionty frnthsmligong c iTself hou var e osd teralnts cemriosdrbeefot n i: Z mtoanherne dvthed ntuo the ch camseliating recieyf tht ifndtry ets aeri teO'ti greatre:te iurnac ndustry. wi&ws vetuf"-h
f.ll "T lo ihe Progrins of Technoloryn and the Gkr smrlting 15of64, onomi Hirstrheve yArhmDryi 79 4fw antd pliarouc" i n l are Sinery netio tait Arn HistesM lutio, cokeomig History Rew Auestrce 196 e mlatirnbanho thIt io indurstn tondt thaca the h isto 11tne obeue oi
-ndustr pntert even today, inr splite ofait ekensivItwa o a* ofnr f ct, oneoucio of the reasong fore the relatively 1t the fanie of ptrog cmiion from thetrasto tomiuelsel wsf
o ut the tifigrd oat Ahougseie othertin onetd ih, itouto coal nn man uet httutindesto onny mntemtnfamtt .,
int racb e s i fficulties.n d fe en n us re M r ov rth

1 11A~tei&,f m arumen heofi tht industryctan tof theolder
*WU43 ha tn, hepas ben eral nedointl rese an oftirhchs t6&th*'Wihsup1,.,sde orcs diiohav determine thupy the b
succss wthinany and, ultmfre as el a the prx or#Ptcivedeatof roduinga ses othe o. But ne aditionee wase lb '?t te popostion whcal fact:othtodnd odtioc alene
....W~h llcaton f ivethaid wosolres pr odtifo reason

'IrdethA-kpplyside orc nologriceall tovaiv mnensatonvfr mre c ~~~~~~~rtog fscreoreugh: and. rtvel homutoguceneous invm66w1,~~ ~ in etmigo hs orgatinic, ficke, eanautlyn varie aztnm aM compositrs of udientiv uptnsar l achin es, akad
ing-of~ h re peaio fsipystae ocons. aesiveppl
Si&,~r detrmin whehe th cetpury lsongae the asoitehniue tbz=ziev lchmit o te mdnuciproved, mter isontine tobe

the t**onof nveniveefrouctin ofd invan e nof h cotOminoxmt abut uppy sid oc (to e ool. And eventii ow thr -a

ilinglevel whitch mcetnologyca wasls tesciinthrochn Ameica atichtre the relativee utut
U to sequen ehods todifeeneoertioni etls hist aain Wwty o. xprineeful ehnain abrde 1969 idin fac 83adsa onts e~ia o. he ew ottn reuchfed in t ool innt the coaches

d~~menta.1 ~ thhnloian inct cotton Ibiducio ppnt 87-8.f
f6 n~c*i~atonfarmor esilthn dd wolproucinii for reasonsii
h~ri ntur ofthe wo ateial. AsLanes as ptl

(C~ottn len itsef tecnologcally o mecanisaion fr mor
Oa,, 47,. Osjiwool. I is a plnt fibr, tough nd relaively hoogeneou in is 6araterstic, werewoo is rgaicficle, nd ubty vrie

ip itj) viar. n theearl year of rdimetary achies, akwar
Ir -as de i
4 4'erk in hei movmens, te Tsistnceof ctto

Olf* Wllitote intenhcetuy lngate hetehiqe
94 j~ohnicl eginerin ha 'Mch iproedthee cotined o b

&iiiiiiiiiiiiiiiii betweenitheii i o ofiiiiiiiiinsiiiiiiiiiit IIIIi in ustry and their adaptationii t.................................... d == even........... so, there hasii~i~iiiiiii~iii~ii~i
reaiie n leet f rt-f ouhinwol auftuetht h

loom!$, up,
*heat mine so much. tbai thm, pic. mg. J,
9 t-V
the dema nd'foi ndiVidW iftv;6900"
liarvesting of, 7rh,*t: *j6 m*islly *bttAtr4iaed, W
WA that the"oth6b t it
p*6A Mg pjO#p? TegIGA
cotton- Sdath'
ncp 42L
j. ntive to bu4r Vinig'
Od toa verp poija 041
constra An Were tfy whet thOdjackrtWiW'Iyi t
ti f6witm thdW j
ury. $ th : p i1h,
S thil. the6 OOtad6I'*e'c
out radkal altOrAtibir Ai4 iiiioiro*ownt, v 2- techno logy. ai ing I
aryes ing- ho
for f Orae, *8:9 n Oqoli *1drief the
al the field as the,
er i ovor
p;ok t.
*ftllq hOdly fte *e nd' xcep 0,
,a ee c" anlb atti
110, acciftt that jhI 'It talb 60
:Ing-acti6nsotthe iWM+_q
A: I ro pAit'bf -flib 1,6, ,i6,hu Iii
loi 6?ifibwc6t"itd iWdti itt ,Ifl
r i1e6-1 e ti
66ift fii dliiid Jilgh I' c ta
i1i'volIV41-th6- lb:ijplo* Pii* g nd'ifi
iiiiis w of ifiAf
rnastdr. 6veir use
6 th
st el).,Thcse teeh became ith, ]Qo fun
boy- ]a
t*0118 to 1
06k OP AU
irw- k, 1 **i

a It* c nl",pptiwise
coll liot
th en
8iiA -ih &tal W01 g m e c nt
prcopce an.." ti

rTq J
was, Irl r o a
Wy Val

'n"f1Jiea~aa meas whidcs ban col o success fully dra ...............24 idM6 hit has onio with teorsurespdcts t oh
4 th fis alfom ted&ealteetn ckr. Athtg ifhfa
InIP0i I Wembdi siiica.tnt diernt a78 pomdw~~~~~ mngsweentitrodluced Intouchearlgik.I, vesorl exti
haj lgft A*T betur ly ad bnsiek of: sroncho expenives.
to klvidl hk'h ulthably otuhave bile evote to d sole
-Oine~iohk~kz~abage fad cottevon ticin macinfeamuchs sooth OJVQ' a tr -salo mcomplete.icroaue se tamachiet woul as havebn den A lrofv dvyfriA heB 1tnica prbem hchf had to e ovelret ftr 8 0 hiklgehious hnica probem ad the li relative tch0 kh~ogy -d hgfty to de mand he conseations asean expetilt io-ftnetieatiiy.T eefr,
d~e,61 thi'e ao rira n n fiues is seiuldfcient y
onsvenes of teholg an mvnt
P. sym thacienthe statledgof scie:y tol, fie ot hee" te "s iupyotvnin
do at th aep e m a bi dstrs itdibl to get asmany mvention asndtk.
ve esuresave b scenfallctdbt 4kk"-t4x6 *me to rdit, as) i en whresponse to the truc 4ik'Ath6Ikkodf~ion~fe ehctao.thti fco 'price haef raisedfisctant invernton a reci
ftlA* fifb *ld ~av' ep trs. Thsi bease ther Iso am
'eMa Jabimmen ofth sat o

thdistory esof siense a?4 moents ic tlye/pp deLiu.4Vpq ht
tlng aouie ingeP stve AWg 1 wi dierbily lte 4e o epnt q tdcnvenie hae 'id o ppkd: thvnpriq.o Letsm sueslotcoine or vtQey of tgsrt- ,M ttentto pltrbe thiretosLAJk ventiveriistoredeepy, inie arti~ l 1 whatgscieniicandy' isecnbog ive "otdisw f *Wf Y4AUY thonstrs o h ruceso WjDeRP c ut deoel)eat3Ot4
percived toic be redy. Iade th. repti ihyrly~ t,4,Y til tok. ao dentemi iatd thUa."dNhJ-, p W recilM- s lohea invntie procties. A ftr al th ,pw
dtiseaindy th e uton ofdy pam~ atrs.,f ea-h1 R axsd ineniel iot n t e t e utn frsns~ fsledsly in comrativey n t ett poswe that~ b the xpla ations bt ee cec'eh1
considerations It is unlkel that iventive activity inr 180ecold havmut4eanmcl oe," o
whae prciedfi an steclle capabl eswe ofdfrke;"0 certai clss f inventions o nwedenR isat .aj
I! ........r

zeouptris aorthi lreels of pricaer uh rmistd f sugestfu agumnts o inreutiomehn foandc M V the rvdt el deadorIntd aspo teatd tereia oa'dtermigion of osbt ie oeo1upysievra!P1 thanvzerotiv lroess. Anifnty !. edmn o ihrlvlp w
The pesetv whic I am sup
cinmtionoweg grwtey liiai~~fWO
gven, sceneae ivento ecl:
cisaseo and invenrution hic i oa dsofrhaepeW stae ofl knowledge y dow en t prog Itsesre~bp,,P
conidt ison met i niey that ur lacktfmoe dvW44 aecntribtd, an mortan was to suso oeyanhttia~ Ihaoe preodcbat it cspbe we oriigth1on:Tesp fceorncles o scf invention est soetmscmptlyilaimted withl na ssulouitcoe prahvrulyasmshP
0 -;V iiw
awy h neetn cio-i iutossieyleiIm..
temdae einofpsiiiteihmspl lstcteiaeiap thnzrobtlssta nfnt!' 0Yf'

Thiesetv hc msgegig hrfr~ttsA ~

ciinifi nweiy rw h cs fscesflyudrt!6 .
ivesinebsi neto ecixsfo iniieyM
cae fanivetonwic i otly nttiabe ihi,4G ,pet
stteo kolig,,on opigiiiei ovran fqe

0 t lif !..
i ti ofmetoighr hrtor-ak6inttsi hsto ffl
iiwi tpn

i 4 i -c e n tr====== ib======ii ii iit= .. .....:: :: ::::::::.............i .... .. ...... l m o x a a t........ t ,= :.i ... .... ..... ..... ....... i..
o ur::::::::::::::::::::::::::::::: .................... w............. = i t h s u c c e ss== .......... r l e ............................ b s t a n ...
r e s o u r c e s t o fms te sft in v e ff o r ts & A d r e ga 4 6 4 rd y th o s e w l h

suippl11yMuve of inventor grdal nedan
#4~ ~ 1 h 0o1at essent thy flattent ot is fcus.,a(mi
*&IJ~k W ohwhich: 96hokler'has adope theabitrar andim
sibe etreme assumption of perfect elasticity.) T the rowth
k4mleage mheans a gradual reucio t i the cost of
ft~btt iesof science-based inventios.l The timing o fin
thed~i6medi'to be understood in trsofhrich shifting sp
't 0"61aiiM ich'gradu ally reduce the cost df acivngcan 'Classes
vklvfii~ More precisely, we need to think indem of a nmher i ''' 11 tes for individual industiies, depending upton the ka vbt W per which inventive activity itn that industry dan drawer,
0 to understand more clearly the etent, to whieh different owvleage are potential substitutes 'in the itiventive prockl~htiklerls hypothesis states, in effect, that there is one supply
,6-1 iOL ndustrzes and that the extent of substitutiontrenders it
='tryto look at supply conditions in individual industries. It
'W that a clear articulation of the relations between scenicd,
iA6had ecenoi grwth requires a ertical exeamination of this6 Wertpp.,The basic economic question, of course, is not an "either
pt poisitibn telling us whether a particular technology bac4hievreiljwbiisme~ posexble ataparticular point in time. T re econoaic
4"Abi i-.-Given the state of sciences, aiat ot eaha techno16
'60'bo~beattainedI How does the state of individual siences difP~~nftastructure the cent of seeiety's 'techtolOgidal options ? 2* JA* eV~ tothese questions will carry us a long way towards a deeper klftdeg of both the nature of inventive activity adthe ptkes
atntjnuc growth by providing further insight into t ie eak
Aaugihg .capacity to respond to economic needs.

A r. By Morton I. JKamien and Nancy 8.Scwat
From the founset of Ecoiothio LIteratre, elrc 195p.1
J awoaes, we all recite, deals with allocation of limited resoures toward-s satisfaction of unlimited wants. Resources are typically identiidAs tnd, labor, abd capital plus a; techhology. that detendbas
O~~r,, ~ Imfr~i nto consumer good& aisparity between the
oY*1,* ,gw648 and services and the desired gives rise to searcity and hpquestion of what, how, and for whoma to pr-oduc. The focus then shife44description and. evaluation of alternative resource allocation Pagahonisms for, making the choices. The Pareto criterion, by which ani allqjstion of, resources is deemed efficient if any reallocation impqjingflhe position of some individual worsens the position of others, Rt *ncmmonly- employed gauge o~f a mechanisms promne. Is the ajano of externalities, increasing eun osae and uncertainty .00;ecty: compettive. market system yields a, Pareto optimal alloca tW p psurces; this. underlies the, view that individual self-interest
thamiy emphasis upon supply and cost considerations does not Imply any wort of Scinlko technological determinism. More costly inventions can always precede less cost1w ones In +twme if dlmwnde conditionsar suf aficPienly strong.

40" -f2itk
Jr*x 1144 Rot,
41V 'All
..the 'cQe B Of ;

C'XjD;L6.diug. i4or*i -qg. 1W, k
4eyond, Us gC46
tw;a reg Wdr i-R 0ge, M 61str"' Ait by t4 qges 11,or fits:*44 S**' W i f
Cq4., InI
as: a, parWeWjikqj #j, A
allocatiqpA 4* #Wlf up V
ducti 4F keth wA-h ,a sn *1 4a4qp*Fp 37 1 R'fe"
indt*ry-, (,sqw, F:f ,)UdtWp 144,10 pi4y, i of th'.,
QAN- Piachnicat Mn
n, "wal to, Vqqj f, 44e 'ppr4'41y r*aiu its eMekncy'. dii nagJ--,,w auld ljo tprn4tiV -,tbp, queAia 4,-"
can. qoilauu, U, 1-91y,, fw.jin div;id f4p tereAt top uggrqg*b o-qtv Gounately,, ,
Qjoe+ x is n4,,
thepxf.ti e4
J Si4' iolt-O-Of ciwt
patent and
the course and rate of te6irdeal advan* dae JU4,w I
so is of interest to policy makernug w #-las to dicorists.
allocates resourcesfor technical HUvan eF9e ])Ute who contend, that; dn4m arfmtly iMNP4iw*' tu '
alternative =4 0 1P
pse Aie)Ry ta,
action in the form, of some degree of government Analic'i of and development is Superior"AmPRgAt'hom contend t t
competition is best, there is a wide range of opunon re*optimal; degim ',o'f) imO&*6Uoii the, o firm-ww

ni italy relatie&O 4hor wonomie, iMuky
grd*th ,..-whiel' theoiselvesat* iheowplkelyl; ZO8616& Att(std to to, beitttwitm
tl" r' d
in, toehiiicaf Amgt Tit6fitly;ifiir cyed ty, V-.-K
1AC &* *k rdg b1i tejeMjhOk)O*'ftWjji 0, der 1h

rpe rAeand -m
79ft iR9 I? nit M MAM

4evantpal usefulness. In the nextvestage, known pribcipe ar us aed Vretpoypes of. new productsdor, preb This invenion or repq cmponent of R&D is characterized by focus ona pei goal agg M ,uperaty- regarding its technical feasibility and econonmic cost.
e, ingstages, development ands commercialization (imarketing) of opWWIg pr proved item, dotivated by profitare mareked especi11jr J7 uqetai..ty about t he eaxtent ef the market aid othned ititrodueag, .Ates of rivals. In aptuality, a itehnical advane seldonm'follow., ,Orderly sequence described'. Instead, there ay Ie overlappn ,4r feedback among thd stages as described by E. Ames 83, 191
ifT"W. Ruttan [77, 1989] ridfltoweharted by Machhtp [51, 1962,
1so depcribes current practeinelassification. We shall not dis w between: invention and innovation or baisid reseaircli anid
ere, except to then extenttati an artice reviewed endissivbly
.4 r oneof. these facets et the process. Likewise, we shall not in
4 eg listingi between prpeds innovations, that redue theboi
,Of pr94ucing an existing -item, anproduct innovations. The distind.4ret in, part ext, whao uses-the innovation. A new or 'improved p9rW ipof one, industry nay be a proeses ihottion for the indutry appplivdi We shlhowever, attemnt to distinguish between the proe", whith. resource allocatioid affects tehnieal advance and the effects of market structure and flim size on the allocationi of resoures
Ug.,4".i -review, selected knowledge about the invehtivel process is fopsidared first. emphasizing the measurement of its inputs and outputo 4nd the role, of "technological opportunity." The resilth of inne'lYainve activity, particularly theirimpact on mharket strncture, are
-Rcted:*txt.* Arguments: about which insitilonal arrangemientes are
*aionducive to innovati-ve activity beginiwith the Shhwitteih afess *of monopolyrpowveredligness. A wide range of emplirical
latep iettis of the Schungetetian hypothesis hasled t6 at diveirsity
,Hypotheises frequently' tested involve relationshp between R&D setiintyr and firmasiepadlovariously meaiured. Iess frqen Iv db@ 'nisak ersthinteirelationships amongteeb lp
-i Aese~et-0iifini sieuror the technical 4inovtio prd iobbn
May4..,.onthe tranhsformabien of orera tifrtit
research output (inventions; innovations). Studies hve indicated thaturdany'developikent projects weuld- typiley be nore efidiently dhdd hila relatively sall*furm thanbirts large onef (sed 19e8 An 1I U lielowi)f Put anatorasthe hee studisindicate that bjr ad large thine a 4 no ecooeis of scale with iespedt td firms sire in the ine
4r li'hefa6ct, however, that many R&D poects'could be done more OOidibuly, by smarll firms than large ones does not tell us anything aboli whe-e the bulk of R&D is'donle or about relative R&D effort or
thk~chieemets of firms of varving sizes. A question reviewed in Section V. A. is whether larger firms spend 'more on R&D relative toridir siife than smaller fimiewhether the intensity of innovational effoi-t increases with firm size. As will be seen. the bulk of the @idehiee indicates that, among firms engraged in RD reatk er
14ni t inreae wth ize-i to a point and then decline, with middle

size fimeostindustrems, wefwould rexpet theit tie o' b8)i~ loc nsitaiy woul ocr tat, -d aeE liraied foha grEnt esit dut dereain. efltieny in thtine1r o i ;a imhrat eserhoutut itenydes end patction a .sia irse wit ceangg finrsaeh Hoer, moto veylreui d.
Ten noucoparion aef their Seaioeten r sizesadheserhoiti ,ofth inoputcean resrchntpu nensiceft ofn e-go ~li plitherbmajr efortnc relado teSchumed t e er intete.Rminigl sheainueceofbato shsarkeh Mrau
to~~~~~~ inetv acivt ane 'tineniitot klcra ientie proce plysn an qiteraoeiay sie'ad'eei i tputes.t Measuinraes anke thetidereao con ndo6' le tudie ndictoa r;esearo pt, g ait tn yns*slo inres th frsin au ptiul den ifi e r inrae '. i Wihnmste indstes oneuc epectin thttepa 'wiO putapensterul ocinventw actal irm Thsiz hntepekraa remet ntfptenstdetia deraiary afiin nd ealution evdnc t c tat keyselemntf outur untniydestnd tofices ai~
hs n thi dimprcion madte oftherizeel atic:mkn :rathion.u Finall hreare whapt menite ocalr. cnohermjrefr eatdt h cpeterian hypothesis, inovigasoi beendvarmnte flue o owrke ndtnnovtveo eu de ed invetion atviI.ado netieotu.Aai fiile andoureut.iew ss hihl focuse, slecte nd onl notov
ths tucomprenie reviwst ofehn wabt se knonstof
any facsudese i n additionseneo ptonilrly soure muhsrse
LD7] F.umer on. Scertie [82iity 1970]; or irca A >esrrisend J. Voeaie [64r, 73 isvrytio recentren*.
man e Oteremvaluableoreviesncldn P. thesrei
971ort (oth dieon atthe eneouetteicaevl ar I iwei
urt e),on Fandy J.t.herno [94, 19721. b ale idrete:o
npedarvantives cas suizes arpear ainnF.vartieatvt;te

19. ] Jl siedF5. Jewkes; D. Saes, and StiladE D alc

stucues and D. M. Marquis T9, 1969.2.1.

IR04 4veeviewed, and syn thesized at vast amount of empi ricalF
ate factors influencing tech-nological innovation: a sub-'
o f their references are to sources other than the acaenuea teons, yournals.

fldaWW and inventiveoupa
above, including the state of technology- among the varid**-Mowlcomomic system is relatiVely recent, still requiring some
'ZWe begin therefore, with a lookc at some evidence of a peatv reatonship between inventive activity and the level of resou" dste~d, to it.
e X6 alA tlhe abstract concept of, inputs into the innovation p roces~ ~ to '6nr~miia embodiment poses difficulties. Usual input riwatf hale R&D spending, R&D employees, and scientists andt e~b~ft~Played. EAch of these, as well as other indices of innovatia4ot 1hds, recognized de eiencies. For instance, technical impi6 mAi~snay not only be developed in the R&D department but
strag afid other divisions. Using measures of scientific
nmg itb6mmettates this phenomenon, but al so i includes employees
with'00 't -uncation or contribution. Modified indices of R&D emWalk include a disinction between total R&D personnel
filR&D Personnel. "Technical personnel" may be refftodVA6 :H-technical personnel, or all scientists, or scientists and
e wNerm~sres are susceptible t6 institutional distortions. David
1 tw o significant jumps in reported R&D spending withont countensurgte increases in R&D activity 18 Blair, 1972, pp. 2019 e 19.% change in the tax treatment of research expendit an incentive forfirms to classify additionalatvte
.S econd, Sputnik and its aftermath made R&D more
pft ticularl imltary and space related areas. The astpact: extends to otlier' industries, where R&D may be vikSwtArshly by stockholders. vl
LiuRIouitat has been measured alternatieybpant
a Majettant patents awarded, important invenions or lnovatogvd gSales of new products. Deficiencies of patents as an output
n I ologe the facts that the patent recipient need not have been rev~gibe for the invention, that patented inventions are of unequal ifprane;tQa oe important Iiventions are not patented.
Ne rtheless,: .systen'tfid stuidy of patenting' behavior hans Ted Sphrjouler, Scherer and others to conclude that the number of patents granted a firm is a, usable proxy *for inventive outputs.
Schmnookler found in his intensive study of patents, and patenting tWIghaatenting, is not a Serious problem before 1940, but post-1945 co gaty patecting failed to keep paewth invention [85, 1966].
Japittibuedto both the',lengtherng period required to obtaipa p 1 approval and a more hostile politicaltand legal attitude toward
n,epvcially in connection with exchiusive licensing.Cne
a nen i v no 9 an r 94 a e t statistics are not com parable.

vtyolm e reltoshesin iewnac ent ade fimstheh! thlesizerowt;hfa eaat ersino the marke and oteIioo 1870--s 190, in 3 The numerssio ciei fth aii~ pats 85, 1966].rt He aso fon p989astoe15--4 be syltdin 1953ye forl 18imajortat aet rivnipa evaute f the ge ine iuty variateeion th il,2nmolr thate vari so motn netions in R& expeditres Bffedrsml thrges 17.S.l indutialrs in 195 fild ofnl maueo net~ ttu catediatvery neallintear nrelaction161. M1 fte
phersioe ma1955 and othnmer ofs (thfor-ear foag reflec t ten ssdr he ero
patent [8 lcans5Y, 1965]. IlLlofudptns n & xedtmt
Shevrrtcon iare RDexeneasures. ofse isasmlefoj-t#, rgaeutical.firmndsramly sals volum5Sher' rersinaalss anfed mefuand. reive [1rm si 9.zefimszehl fxd

between the rate of R&D spending andttlnme fiprat 'inventions forthcoming [53. 1968]. Comrr ssuyo h hrucn
-tical industry reached a similar conclusio[1.96] aitndWl
-found across 13 U.S. industries a highlorlto ewe R'''-tensity (R&D funtds/sales, 1964) andraeoftcnaliovin measured by the expected annual rate ofitouto Acknowledging a' poitive relationsh:pbtenivnieotu n prior devotion, of resources to inventive ciiyrie ute 'e tion. Does the presence of basic knowleg.ascld ech-1ki'a opportunity, stimul ate inventive activity stes iuustepr~ potential of innovations that satisfy aneitgwa,-Dnalft6 latter possibility would preclude explangivete'ciiyasa, economic phenomenon. It would also dn n nliuee o re


ditivention. The evidence, however, appears to support both
lbof basic sernce'ie reflected in interindustry differences in
6t& SEef% 0 and innovational output. A Phillips is'a leading
'feir the" pitive role, of "technological opportunity" i.e., t V tth U9e csitile knoewledge in the field [ 70, 1966; 71, 1971]
fte that, extensive R&D programs and high innovation rates 808d ofly ev68 period of tis where there is related but exogenous
*0pietific progress. If progress in the science slows or moves in direeti~s W!8tg t feeropportitis technical progress in the market
0 simlar argument is advance independently by N.

Judflfy %nt 6 dstinguish differences in technological opportuht (.&eoeAn cross-sectional linear regrssions by industry of patS*iUitd.b s4les P?9, 1965]. Thterindustry differences accounted
'S ekc of th tothl variance in corporate patenting as did twi eces in sds volume. The bulk of interindustry variat, hht agMintelated to sales.) was attributed to differences in the I.,nce~lyngscence base, or techno icl opportunity. The 14 indusidg ~ d into, four broad clsses, based on the regrssion cofagik patnts:awarded, that accounted for most of te interind di~~reifc i patenting relative to sales. The groupings were
generall themnicals and stone, clay, glass, (3) moder6t -'perdq-,m rubber preduts, fabricated metal products, MachinOe Utng~iaion equipment) and (4) unprogressives (food and
4 fktles'and apparel, paper and allied products, miscellaneous
selaneous chemicals, primary metals). L. Philip' regi '%f ottimrchworkers on total workers for 12 Belgian industry kt~dj* ikei~ereated interindlustry differences accounting for about
#Afi~qh f otal variance as interfirnr differences [(72, 1971]. Moreover, the a ton, based on regression coeficients, of the 12 industries
geg psldonosgificant loss in explanatory power relaA! groups, Although Scherer used innovational output
(Vdq~fsj.anI U.S. data, while Philips used -research input (R&D P0,4661 61 Belgian data, the classification of industries was strik,%q 6mi i. Further evidence of the role of technological opportunity ij, po~dtd iy T. M. Kelly's investigation of 181 large multiple-prod.Jtritm Wthin six industry grups [44, 1970]. Measuring 1950 in-Veni4 e 't -by the ratio of R&D employees to total -emrpl oylees, he
'Ia logiA1 opportunities played a positive, signdicaint role
Idu"'i irovatiVe activity-of the chemical and petroleum industries
i~t~to otler industries studied.
ao -asocated "technological opportunity" with ease of
a 16Yiproduct differentiation [15, 1967]. Based on McGraw-Hill 1seawyd of the purposes of RAD and his own studies of the pharmaceutitalistry, he speculated that a, maj or goal of R&D is development otinsw. differ-entiated products that afford a protected market position litail4 14,. 1965] He then hypothesized that research: effort would
wogeater! ini industries wher the prospects for sucessfulprdc
di~teniaton rebetter. Coniano'r Identified consumer durables and
invetmet podsas indusries wIt hig potential for productdifr

little. R esqirqh d,Tort, w 4, A
research for the avera"ge'firm. size within keh size Zss.
research tv! ,
jii9rjoW W t
fereAti,*tion. Cdma, C0 JA4. 41P 01
research levejs: ten W,: tq be, f4r-grgaler, 14
VI_ -rod Tai
sume r durables, and iAy tj eiit gwd,,q t4# K4
sume.r nqndijr&WS 4)d U, 1.13,pU44 Th
p, q :-, ai#qng, arge
product,,, typ( was, C eciajly r om 941 WYM
25,000 Ovlployee,, )
Schmockler'...0i. t4o, eontraxy, found,,4 Q
99 "4 1 Q "0
pYOVIdW,,jittl( 8 5,49 W] Ch.04QI f
0 wv(
comically or eeh fogica ly impor n 11v tion
t no' ta d I en S Tal
t e Oimu ,w.,Was typic jjy alAecll wal Pr4le RP.,qP1P90*Y lic taro +CP
conem d: l+X geJy U c6ntmt
ye ,. P (Mop
im lixte stiin4lv4, 4,
which pitpdvv )WI46 _kQble", s, ; Ie; i'n" ec
diseovm-y W not d as ar
as. iiave tioft 14 1, ATIY A
comparjqp,of1iin x, ri' ,,9tjMPo..tAIitjI V e io,
in: four fj Ws.l'mt, uwiupp.ort:W the hypqthe sthpbt in
beget fui thex-jnvent'
Schmopkjer a, W41,haf inywitiqn. is:, largely, A ecoffloo
h, MT( _V ,
PII.rsuPd forgod.n.; -t at e;Kpeqt g-&,iii v gi es, with'. 4 -W 1 10 jpf)
goods. embogyln 'aie, ijl edition; And th proye4 par ods arehvrgely -basq ,4 )w pre, ,t,",pitaj go
94* 9 1
of ca f al good inye4t],Ons T-W'GRICI there, W to vary overtime, a] J di ly w4h
0_#m6ng industrips, qfwt 414* YPYQ4,*
to, eurreqat:., spinall. goods, sales.,Tim'e Opi'de 11 i 0 uiventioi s was compared v
w4h that-of iia, estp aent for
14 94, 49,
Vuildji g, and, pe.troleuui r6l i 4
t on regression analyses suggested v4t,: o4 41,1
investment of! an iudustr Im, s,.
PI-oportiQn-t nt=berbf.capita' goo, s jnvoi4 jon&,,
Tn, Schmookler's stu4,.I'nvemtioTis. Qttrili)uW .. t i m*tmf
e? pected, to: us the n; 1,4 con t 11tw*
tions- by: the industry expected, t6 -wppTy, th, e-T f di*441 0
affected.the cond .regye- I WA JT kil Qgq"il W q t folly'Al
oa, ernaten asij 9 VM
patents, Taljted:'in. 1,959 alt ip#V
usinz S T er's.(14ta declassified ind e, wvnti" 'I
-; q4ry SIZPPjYiTW 6 Ix
Si. P.Jifar rearession, mployA i"S attrl -ut'
Th e, coeffi ient of size wfks, significalit. M4, ejow to unity lik W41,
eMpqient., Hej inf ,rrecj' f r xn the,% ip sultp: 11# 0
the'maricet Servea'by an'indu*str'y teuds Q inc .r invented fGr p
it to. prqduqqJ*4 jwsome pr. POTU
prepsing.ylve.stj d ,Y. tQ IjIqr(qSO
Wit, in, an; Ih ustr to
0 S'.14veP '-W:
pit p. *Us f9'r A owo4% th
preciably'so,411 PO i6il a,.Va tOAQq* u*
cl,,wiAfd by.. squr, pe, pgges. Wu,,
ieved b' cc4pow
y a, yamty of t '. 1. 4 GgA al w*A tbi Mvsti Oftwo!
Us*1'y s
Y O'Pi '4'
W 04 pawuw $ 410
-n j'd isqu55jQ ti-4sjAr "G 4 r %4
'I'city v' IV (I ii li I a u I The hypothp I ie e as n in t oug unequally available

t. "1,pesiiliiesbecome depleted was also tested b
tye eries.of pants for railroad track was similar VCi all, otlher railroad, patients. The resemblane in
Vppon,,despite differences in underlying technology, lelpqgress: slowed as it becameless valuable, net 466jAexaustion. The volume of invention in differas anuaebirig was likewise smlroe ie
MXAR11 wesre cently rorted by .M. N. Baily investigating tro [5, 1972.H f ound the number of new drugs
44,% ear. was positivey related to R&D spending in the
uA stry inpreceding years (the development period) .'-.eated to. a seven-year moving average of past total
py dAnT pg introductions..

rce allocation influences technical advance, what is the naW 'latiotishigit Is the innovation process essentially a lottery
Ac 'O6fmore r esources improves the chance of winning, or
ik I *ifdbd'b*defiurministic rales q. In-depth case studies of the inteiwdbelow s hed some light on this question. eb allocation related to the quality of innovations, to
t aid eir speed of development ? Are these economies M16 9 se in any of these dimensions ? Alternatively, is
aau iliiat by concentration of ever large. numbers of
igt d'1&er' anid technical staff along with equipment, or is
iV4 kkii'bey~iwhich no, further advantages obtai n? The techinovation process may influence market structure in'
oway as th ehooy of production does, defining
11, le, 6pt dhat scale, number of firms in an industry, and ease q ~. J'11 Olivers, for example, demonstrates that the presence
'&' e,,o6nis of scale in the innovation process would
=,teoncedtration in a single unit for efficiency [66, 1973)1. The empriclstudieg reported below relate to these questions, beginning j *jfh"6 -d6t tihde-off in development, ,proceeding to the effect of
g~hW6i~he e ciency and quality of innovation, and ending with the
dm~ '' litff of isu'ccessf ul innovation.
tU Scherer the cost of development increases more than
Vl 11 a lt with contract on of the development period for several rM9jiiO Itie tine-cost trade-off is convex to the origin [80, 1967].
Involves learning through time, cmrssion of the
&W16 Men' rid curtails the learning process. Second, the uncert 04,~ditedwith R&D necessitate experimentation. Sequential
i tk es more tim e but results in fewer false starts and
fwi d~i iups Compressinz development time requires mome edst Alle0 peimentation. Third, time compression required more IC e~ipTd er untit timb leading to' classical diminishing retii W3itield, etal i-n particular, found support for this relationspby iimAtihig the time-60st trade-off functions for 29 completed
ibasttb iby 11fims in the: chemical. machinery, and electronc inutlaits, using inthriew data [.55, 1971]. 4&401heptt0aly there are two major "scales" that may affect eficienc hidim qaity Iofl inrlova~tion. First is. the effect of ftirm size, on the efficlency of a gi'vettsize RD flity. Second is the effect of scale of the

1%D facility for a given size firm. These two ga iqqg, titll distinct and their answers can have quit6- differnt 66 plications. For instance, economies of scale related tofu 3 s indite that concentration of sales. could enhanePD1 n'ti i6 hand, economies of scale related to the size of tie R&D support cooperative R&D or concentration of* polil' el i&i fewer but not necessarily larger firms. There is more evdeco fte first question than the second, perhaps due to theAl_3atiwe-~!II~ of data and sometimes tacit assumption of multicolmixf y piicld studies over the last 10 years have typically sh& i thwd *h there may be certain advantages of size in exploits 9 tI R&I. it is more eficiently done in small-to addinni sizts li h~i large ones. The extent to which eficiency varies with toe:i6,ftt R&D program itself, for given firm size, has been little nsinit These two issues are reviewed together below....
Coilanor [14, 1965] attempted to discern, within a; f 4
pharmaceutical firms' the relationship between research R&D employment 1955--0, and new product output (the,-hm chemical entities or of new products in firm sales). X0-Cno lpq~ faiind that the presence of a largesupport staff relative toPM staff substantially increases efficie-ncy of a pharmaceutiW facility. Rapid growth of the entire R&D facility did nogSIAM7 pair research efficiency. MAarginal productivity of professdia. r personnel appeared inversely related to firm Size. Estimated eq6.1 of research output with respect to research input suggested ecupa of scale in R&D at low firm sizes but diseconomies'as alkn.6OP, moderately large. A. S. Angilley reported constant ormoe J.7
creasing returns to scale in the innovation production fkdiatiox fon international sample of 20 pharmaceutical firms, based on Tre of innovative output (new product sales 1958-70 or new weighted by therapeutic significance) on 1969 R&I),expniui[
Bchinookler, reviewing the broad evidence, found that modest level, efficiency of inventive activity tends to vary vrW1. with firm size [86, 1972]. Large firms spent more on R&D pending in 1953 than did smaller ones. The lUrger cost. per p s i not attributable to a differential propensity to patenrtt bY9M T W; Schniookler cited independent findings indicating that small in ploy a greater proportion of their patents commercially tbI 4i finims. Mansfield found for 10, 8, and 11 major firms in the cp. petrtoleum, and steel industries respectively, that -the number -~ nikfit inventions per dollar of R&D spending was lower in ftielgs firms than in the small an d medium size firms (53, 1968]. His,s4tYy also indicated that if firm size is held constant, increases in R&D ex penditures results in more than proortional, increases in InVIv output in the chemical industry. No such advantage of increasingRD effort was evident in the other two industries, howeverr. Mansl* et al. found a ranking of R&D programs of major chemical firmsand of 9 petroleum firms on the basis of overall quality and effectiveness
-per research dollar expended varied. directly with the firm'stotal R&D budget and inversely with its sales [55, 19711. W. N.. Leonard. found.
foi 6toadtre ii idsr goptaIhl, iitfnne.

roioneat nenity was poitively seneiated with asubsequent growth,
ppotedR&D was not [49, 1971]. High federally financed
R&Wk,:*,airemift and electrical equipment industries resulted in
1productivity of sueh expenditures. Scherer,, through
Wfbind that patent intensity (patents per billion dollars
W'4at inversely with firm size and increased with R&D in&p~t (R&Demployment per sales dollar) but showed diminishing
16t twrtse1951
06owrinterviewed about 25S development managers within exAbt large and small com anies or in rapidly expanding
il "ptieatorganisations in either the electronics or chemical in04.He also obtained actual cost figures for a particular
rav~edewl~ ment efibrt by a large company and a small company.
eessstntestimates indiested a given predaet wouldcs
th flias as mch to develop by a large firm as by a small one.
Lkiik fi~se found,: seem to become enmeshed in bureaucracy and ftd t )reftingin aless hospitable atmosphere for creative contribuU6 o~er~in' personnel. Superior technical personnel tend to be
41tid wkfa 9e companies where greater latitude may be afforded W Tfiel r the firm, the more dificult it may be to recognize the
Pf~l~s~iieding solution. Finally there is evidence of greater cost
iSmaller firms.
b* ; invention a-nd development more costly to large firms, but
ifi fl ive Is suppressed them. Blair finds evidence of this in
Lrm, th~iesubber, automatic glass machinery, shoe machinery, cable,
m aces and golf clubs industries, among others
FS 01 K 1gt ishties of large firms in developing certain innovations
ibted by Blair to a desire to protect an investment in &ri~iltaM n logy, satisfaction with the status que, underestimation ftial demand for a new item, neglect of the inventor, and
Mi~ie~tn..of research as well as incompatibility of bureaucracy and
v-Ve_-ardh'quality or importance of inventive output, D. Hamberg's
**i*'f1hi own and others' findings led him to conclude that large iWOi1 ls tend to produce mainly minor inventions [31, 1966). He .61in'tat the fraction of total inventive output of these labs identiffible"signpotnt" is less than the comparable ratio for inventive outt thber sources. Mansfleld,' et al. found modified support for aftf* ;tesis in their study of 1964 R&D programs of 18 major
d~on~afand -8 major petroleum th=s [55, 1971]. Up to some pbi*1dyier f1rms tended to devote: a larger fraction of R&D expendiftes -6o basic research, to have -more technically progressive pirojecS, and to have longer expected completion time. However, there
T~lifte 'difference in these areas between' behavior of the largest firms and that of firms half as large. Lt.:L.'Duetsch has dissented, at least regarding the ethieal drug industr [18, 1973]. He queried physicians and developed a list of ipr tait4w druigs appearing d urin 1940-67. The contribution of ethical 4thg manufacturers to the discer of important drugs (90 Percent dfW.. toal) 'was comparable to their participation mi discovery of All ne w, drugs (87 percent 'of U. S. total). Jewkes, Sawers, and Stillerman fomni the sources of invention diversee and the large research

labs of industrial corporations, not respowab r rtt q, bMW:f0497
nificant inventions .[3( 19691 P avitti#u4; W41Af evidence developed during the.'1960) audvonciaAedQ1,at small firms play-essentiaj,- Plementary-i avd,,interde roAqq.
Ua the process of innovation [,6TJ9711-,:L&Xqer;AVWcWW contribute mqst to innovation m'arem:reqWrmg
i -lame WA"I*w P"pror,
auction, or ruarketing. Smaller firms tend, to cimoent ate oRMpWipA7* but sophisticated components and 11 equipment. They, havei roado
very major innovations. w1ma largezmis lot tbe 9PPCI SIIP by.
Finally, emphasison good m management throughout rm
excellent working relationships and: cmnmun catiom awrw 44o A4% production, and marketing departments isaxepwrm 'thpMq parative studies of success apdfailure: in. inwvg4tions. U 04
insist these interrelationship scaxmot be ov"emphasize.0456 IQZAIJ4*I sides these factors, Carter and. WilliauLs also:emphaoz ;a e TOIPIO' communications between supplierfirms and customer.firmm .1 1 09t lasting and being receptive to tecImical advances J2-J9Aft7],., 0 man studied 58 attempted, pairedinnovations ( 9 failures o,44 cesses) in chemicals and in scleutific instruments, P and success, as weJl. as similarity Ofined.i rMA
inpairs,..wexe 11 ternwf?
market; not technical characteristics-, Rmt attempts, to J41-1
evolved a formal R&ID structure, and patenting, ulthol *Q1 sylste4w, atic differences in R&D organization or incentives coW bf, AiW^ The distingiiishing characteristics were related tomar ,efing,, broaft interpreted. Successful innovation Lenerajly involved. greaOr Upr tion to educationof users, to publied y, o market. fo, ef* selling. Most significantly, succepsfulinnovation was mftT*ei understanding of user needs, throng- ha-at al I ,
firm, includinLy the R&D,_and productioA,depar4wo'., marketing. i Wl 10
In case studies of *novatiGnsia rWS OY JAPPA *Aft,
British companies were compared with companies in the TJIiit94 and continental,-Eurppe, [V 19Z21...Particular attention.vU, n the, role of qualified scientists and e neers at every rlmin mv 7 pgu , ta .
to successful innovation.. M&UIT. instances were.,fou d Wof 4 initial. invention in the R&D apartment did not lead tosncpmf,4 Wnovation because of failure, to carry thro gii with'skillfq. I ii64' planning and/or marketing. C66d conunpnication.betweej .t. O'eyel opme.ut and marketing depart 3ents was, found', essmtial. in' goods industries. Laagrisfi and others studied 8i IE=oVAtAQAs t, lat WqQueen's Awards, in the United Xj gaom inJ966 and 1967 [ T, 19
7 innovate U,& pstrie ,
Myers and Marquis revw ed-56
owl, U 'I V ti[597 1969]. Both stud ies foipid, fliat iIOX4 i t 6 'inn ; M I .0,
identi.6ation of a' n, eed. that':,OoWd be met.'"- S, MpOr ant, giore, oft4p, than t6 realization of the, potentiRl, 1W, f .of the digVe,
In sum, there, appeArs to be Ln.'- 'I
vpr$e, '191 VOM.. ra
gene. Ily apppa
in the novitions There,..' ral r to be WOC0nOPIJF*
of scale.M' the 0m produotioiR. fv*tjon, VID. In,
with.further sca,10 economilp*. in transfor- na441IGa of RWe&o novation4l*tput (quantityand quall be t
4), 4 xc p
studies. have b gua to i ;olate Sonle, ..&IM UO mana erIL chiwa*ri t*. associated with' suepeAs miii*vati

M .1 l j ai. assessL aF ineOVATIa lACI*aj
dedto. grease profits and growth for tkiia sk
dysr eket structues Minimum seal for economipr
athe nie or fail as a result. The oSuaeg
~Wa genderr to become a viable industry member may
*to entry. Innovatishk may provide a. means of entering
aoaeang an existing firm's market power. FinallY1
-k te diversification of firms, into similar industries majr
innovative actiiity monsubseuent economic growth
Ofl Prductivity has been extensively studied and recent
pee 1. Nadiri [60, 1970. C. Kennedy and A. D. ThirlwaW
i~i~naOriliches, [30, 1973], Ata 1972 colloquium it was condespite mthodological difficulties and debates all the tfheplevel of the firm, industry, and economy-indicates
wt~ uion of R&D to economic growth/productivity is posia
,t and high [61, 192]. Estimates of rates of return range
wwn40_erqent and 50 percent with a bias toward the higher vfjo p ;tive association between sustained R&D programs and/ 4K,9 Jan 4 aet groth and profitabiht at the firm
MY .1 b Mnsfied f5, 1968, Smuth and Creamer (89>,
v2 61.7] among others.
pxehenswvely reviewed the literature dealing with, the
0 gcal advance upon economies of -wale [8, 1972]. Hle
1-0 e mthe late 18th century through the fast third of the
F~~~ ,,e Ic change exerted, a powerful impetus toward conqt q eaese advances in steam power, in maeIl n ehd 14' (Awi gapa nd in, transportation (railroad) permitted an encourpenalwon. Since then newer technologies (electricity, maS Iitf ode of. fabrication, trucks) teoded to have the opposite
raiihts size and capital requirements for optimal aireached -similar codelusions [32, 1967]. The most
t Oni4 dances have not edy tended f6 permit economic proa4t4maller plants,, but have. also increased effective rivalry to
*VO tsby widening the range of .substitutes.
;.;Xechrical progress can be barriers to entry, particularly in Oc, ijjn utres with igh technologicall opportunity." Phillips has
-fMD and innovative behavior by exist' firms tend to
aong with other factors, limits to entry by new s [70, 1966: T, 1_1. The Jowver the, product price and the greater the exploitation
uies presented by exogeneous scientific and technical develit4h leas likely is entry. of additional. firms. He found support
hypothis na extensive study of the commercial aircraft
Effective etr inothis industry has been achieved only
whir ecnialadvanceqs afri substantial cost and pr
4eR advan. stages for carrier. Relativey low operating costs seem ve bga anecssarybu not sauffcient, condition for a. pdln to CAP- gtend lrger rke share. The effect of technial advance as
g t efluy in the. conupial aircraf industry hasbe" ~ e
smw:a Vby an apparent prclvty bsucpcesu ensacturers to-re-, U,' toplone with their original unna.These firms harn not. nlwxya


continued to be scimtfficallp progresnVe, and t4mby retAin th r %A ket position.
'-Ofh* hid ustfY gffid- 0 Cd1fliMkW*Wnd
R&Ms 6, ffi 4jor; iblefti6*t 64 4nWiffithi Avaftk 'fhtl* ph&rhn4 dftf *U, "*th$f6fk ar aidtk U &I-n'k tWitiooedf hift#kt
SUcce .3 ": 19641'. eotiw blitrY 119dally
!9 rN
'in 4 ry. bblhwfil f6uild'RM 41* A statute an en try b or to this' d A FJ
eiiitiy barribr"n thol bl gd ist* laidriiatidaid 61w6totA6 Apito, industry.[21, 1965]. Rivalry occurs mainly in tecbmeml inn
"" ". "A' g .. LiLO.A
6,mc .1'118&&11s6 bfth Oibdi*W! du 'jlluwzr ir
-Manu c ur .1iiiin, fl
eThp oys thiiiu-6h
greoftionts"An &,
kno w "Lho''w (d :
.. I .. p 1 1 I.. ni i' ]; ft q I k
tedh ic 66h the-rnpre dil it 6 4A
ments' on ?& 4hble 1 = t i;i6
do. 1. Ont dAd
ve opm Onggleering ap bilitY Wso 6 Ors. AIUM
h' Thig IR&D o
ai hfain a:,d6fe'n':sive nih ki 681t.16 Pbig -N 'iAt bW a J 4t aqi e
1) C, Mtf61104 J. F,'.- Til dh'&xib ddod: ti6 ftbtio df 4o
ft barro6r b ibt
as 1 0'0 ry *itmifa; Atio bf'dd"161YM6iiV dftwiJIT,
h y po th esp ifs [, 8 119 69 1'. Th-ei t+6ivfe W' 0 f. A Am- Aoin o 11 studies (sevalognductors phofo op iAjt), 6fih+' I 64 1"'A' durffi'g, "th 6.1nidus s hifan yf 11MOM, tre'iequi ite. to'R11V!entToiij- -6*6100imm 6 r t i age; wita. Many id flii-" idw- tiy 1'th '* IC, 436 Z rqi Aw d7'AAd f-heits6iich: spb4 w w il m
sta ntial'604s of biiilding. d ftmtal-ni gr Im'. t, JhT: it
constitute an Phtry' baiti6f avorin, 0. rVe di, f MW ma Y ecor
St-i9e7: birit, ft.' to, M Olt Iii 11iM
patehO e:ipiro. nd Ar, ifio t6dfii6, 69iiie: ftfidi=iA!
f La
a on It" ft6i lAg Oji. i HAA
entry a-re hoO L I
'On. 0 U m
Tavitt, nd Wil foAVd: bppo'ff 1f16$,;fbr fii the 6a *ea ai& 'Offhd 4kr-64 j "Ae Wh
relativel* iffilft'i
scale are i, oftAq!; tk6 VO
(Sf -& .- d 151
entry, 1"R f h5jilre hi, i h 9061 1' d
n-0- %) od
vib VsSo
vo, h tu bf -1 t -bIfifi*
,r6 M&
lp "i, I
MD 'Aft. lit,
P910'. blyq 11 bi j ,4t I'd a h 616 :
14r 1,
m Eat "01f 6A e 61 fouft., lsw4k "nf fitiA1,16VA
Su fl
C ,ag t 6,T" th6 16 f f, ult f6f f
't iw
011& I VA6uv Nclu
Co's $I h% ihe Z 6A Ali
achieved in several ways and that oligopolistic firms need not have

ogicllyprogressive. ife then undertook a detailed inves. A ,byU.S3. commef-cial aircraft market: during 1932-65. Ad9 nce and technology relate t0 aicraf auatr
rp;,qutsade the industry bnt provided, opportunities and inceamutad46turek's to develop new commercial aircraft. Hle con'6hhstiem. of. iovationls had the hypothesized effects.
nrn rof maniufacturers decreased with large shifts in market'
e gyfaiing (71, Phillips,1971].
e9,sis. pof success Wre& success" also is supported in H. G.
Study. of R&ID in tli chemical, drug, and petroleum 8, 968..Oe p thie ajo dterminants of firm research I was found to be an indeix of the. firrix's prior research pro, iatre by the number of patents received per scientist
er epoyed. ni he samplefirms with higher ptented outr dbltiicWorker in the pas w, dteris paribous, more irerit sive th inthir rivals. If Grabowski's findings are broadly
past 1&D subcess tends to lead to greater current R&D
if ti llrn cohbe extected to produce further innovational fi eeralt a wid bt'ening gap beieet the technologically
fi ms antd their rivals.
lltere is evidence of a positive statistical relationship be siiy aind diversification. Diversification is the 11ehfirm classified in one industry produce goods classi81111 h6 r.*Pattorns df htdfindustry diversification seems to be
r it proceeds by internal d'evelopm-ent or acquisition.
h1sidy coverla 19291954, found that industries entered
root otenby: diersifylag fms ad industries: in which diversifying
mi) -fr ^tqently-based were -industries characterized by
Anci personnel ratios 97',.1962). He also noted a positive wieen industries eitored and productivity increases. A.
Wbb f0"" A d iversificatiett over 1959-1962 and found that 'Mdt~1#sbffrequent entry or origin Were 'characterized by high
+0fs b IR&D expenditures to sales controlling for: growth
Yad other variables [99, 1971). He suggests that hiX
iWA~h,!ten itjy tends to encourage and facilitate diversification haso

4*6i~id~tamong those associated writh! the position that monopoly
poterosd alare sze purinvntive &etivity is Schumpeter [87,
eonvisoned an economy as: an orglamsm withclscntnl
&Rbeinrgreplaced by superior ones. By this process of "creattive
f~citoi. the organism grows and flourishes. In an economic, Sys, 1gnrationrr and growth are achieved through replacement of 1kistliiproducts, processes, and modes of: industrial organization by
ipbved'0aes Schumipeter'saw the'quet for extraordinary profits tfih fi'iixnovation as the motivational force propelling the process q) 'Ve destruction. By the very nature of this process.. the monopoffib ibi ahieved through innM'otonis' temporary. I1t'is suscep-

&4 A[ R A v
1-1-1 Lh
reso rces,411 'R'ctkrh1nMi1r4U, fu Id -1hdiu it Media lini.466fi' rMAS li*-- 4.0,
P.e ]jL6 on W6 a el m
its incentive WJ)ftx qytLtk,, Thug'&11 '.k flifii al*aiili
on U IAI lint
porary in 46 Vine n
y Welo,
ttr ve. FA& d %VvM 6 oi6 Wtw4 t
Mpe i Wa wi
V fbet co, t, f,
ndeed, he ViOrit ied tion won throu,& 1",Oviai8il wo-UM bt %Sft4
-that is the 'parwdaotint rblevant for"M 'ot,
III cap)h"t reality, d!stiii&sW
ture, it. isinot that,'"d of w.. floj! price) competition from: tlie Aeiv..cpwb ity, Source 0. Supp T ltis, Alr, Y ry TTO
UOU. of 01, YWe how haye.1p.iPb4d Ac ts. itof oj4, also, when It is an ; ver presout Oi eai Vlie b to be in a competitive si ual i6h even.4;iais, a 6nej-A Schumpeter, 1950, pp,, 13"51. ;4 t t
H. H. Villard refers'. to ithis Situatjm as, 1'e 19581. Philb obaorves thocoWAOI tWR in
the anal' Wi.. -t Priet
off 11MA wa ,to rqu 'PiRtry
price competition. [711, 19,7111-TO'bolAteIrlais argtmmat ,madntaine& that.4 resource. be gauged: through time 'v)jjp
The. promise : or posmWiOn. of moxl Oty PrOW a,044* W)t enough to b n JW
ri fwth-, Mor"04PPOVAUY0144 Yvi
hraith emphasizes t&: ini OUnolo ArmAze or M., gb 9
,contends that the era td cheap hav6 set in. Current for teelmioal persouniel, and th*
The needed rewurces am, ftrw
subst tial, d
"just as the prospect of monopolisticposition raises of the most risky Hmovations, so bigness makes t h 4
expensive" [65, 19*, y A24]*, the technical uncertainties associated wit innovation by un&* Ono mw "4
9 Xplson' 4Pdjb*ve---*r9u*4.
aize an IWA alwm
4"Mozkopoly, ipo qiuv#PsfiJ*,
through WhwpW-Pturw JjaprPRO time erkAtas: QW, i4."44040ftft
q4r4-46ri Oro
lossto, Society fr?414 t .4)" P,-r4pa
A- g*
thP Prosewe,

Mg Poll wmi a I
en m-or-e selecdv lj so as A, im'pe

tpu~noof-he~ypthens. A statietial retmmanip. betweenan ease ..ti is mostfrequently song t th.eploration of N. i iippqaqf irmso on both the amount nnevational effort. and
Mn~~tinalsucess(output), (as distinct from the impact of fir
finotae d isor sed in Section IL. C.
-A reltio btween monopoly power, measured by industry i~oew~witonandirentivenactivity has likewise been anaght. Finally, O~hwe ben indie tests af the hypothesis, seeking statistical asaod~an btwen ivativeth vt and various suppoed advantages OESWandmaret ower.R F.M i he and P. Temnh have recently brid a. reuc a iterpretation of teSchumpeterian hypothesis to WAURU~t~ive, estblformh and to- show how various alleged tests ore ~ekt&1201973. They emphasize that these tests are by no meas idieliog4 one either brade an ge, of uses, for the uncertain ,**Aow of nR&aroect thn will a sudglel product firm [ 6S 1959. Re iplis~tht~a sigleipreducb firm is unable to exploit an invention A004retlylink d t priary produtt through hecensing to others QV t inga ne prdnct line.
tit him. ralthlan A iinthal [5d, 1053] argue that industries
Ow.pmm v seera lrgefimeacah twith ameasure of monopoly
Onmr~il, egag inmore innovative activity than those sensisting of moymnl firs wih little market \power. To the challenge that fewvA~q fir inhbitsprice comspetitiom and alsodsifles innovative cornpe~ion Gabrath epliesthat innowationr pans of rivals are more Wovl~odetet ta pricing decision (a quetionable assertion in
G. JStigler and J. indshl [93 1970] and
a~tberfore o cptible to the same pr essures. [46, 1948] *s at he esre to avoid -price copttonwin anis try will 14(1a inovtio s tdo not increase inutyoutput, a view sared
'.i IJ~oec picecopetition (due to oligogpoly) need not affect smallI &Mrad lage frmsin the industry in the same way. Phillips reasons tw .. r icecopeition is tacitly inhibited, non-price competition arnng.thelarestfins in the industry is likewise cm tailed [69, 1965]. 'R, *rinariy ato1 of small firms seeking profit improvement by intr~uoin-new sititutes for the existing product. A modertely qqix rate indstr may have firms large enough to realize the ad v~iake o sie i imovation, while market power is diffuse enough :thti rivl etaiatonis not an overriding consideration. Phillips conJecu~rs hatthee ight exist a degree of rivalry intermediate beI n re:Ot cmpeition and monopoly that is most stimulating for
In biof Scumpter and 'his followers J. S. Worley found this
614OICt ignficnty. greater than unity in only 2 of the 8 industries r*-sened y 18 vry large fi rms [100, 19611. 'Interest ingly, Worley note a endncyforfirms near the middle of the Size distribution to lip ~lalvel Inr &I) persoinneltha~n the. largest and smallest firms
....................... .. ................... .................. e .i~ ~~ll

0-0 9 d.

Comanor fit log-IhVW vAt4 A5 ,And 1960 d f 3,97
firms in 21 groups 15, 1967f.WTbhaWe PeAstimated eluticity of oxr"' ,ploymeiit with, to &m Wr e
-ca'ntl (greater th", jMfty and! wawgk nifi&jitko 16amb the
been smaller thAU..HWhbergs beemse df, tbPIAtt4?946ftdO07 Classes *.9
the 0 PTI rri'
tibgenousifidustrimi s wl'.H
S6herer criticiz6&-theAjtt4w:wi&tiaoi hmom6 of -ffie
in tbAse.studim 1719 1965JYAt b*tj-he4bmtczUIcdq the
-relationship. between )resesr& iAte am el 4,r",wMh6*tAIj*ge
were grn6rally omitte&from tbe MMLPIOM I&
or nonimontonicityim,.timin'lacaobsbi -bitwe"in*bin I
size. To,-6-verzonm. these, iweakn= _1 Sahwirp lid 4m" A;L "I'L largestAWM7
power of 19 5 51 sites, and; agwiWb po*em df thd 1 9 .cubi e,,,, e4,:u 4 tij Du i nv olves h i& coH Was r ity4 o f the but-rermits., det6qtionvffdnflw0on *oisb txna4Ram xelationshiw Scher6r, 4knind tho, -i*Iationshm'' bettbak (W ment and firm size typically had an-iuffactAai.qni* MW ,PIC(YMOA irwteasiuw tt t U
the, L"iiili4t I
. . . . .
R&P -WtejAityy-,,appwwA tw tahk*19i; J )'I 4101
ns"r abdi* 1*4in6irAkhtk AD,
ovw 49547-1959 iAd firiak skka V*40 m4k**hni fift',14iehe
"4Heieiiof. firm -BV*A I di& pat shi f t. p)wk ticwlly tov4 t
R&D relative to sales than did somewhat smWW flxk ssas regressedmesw6h cxpeftd 4,1200,4962;(A aaesing
ja Pnaidl
Anwng the: drtig Ams,, tnikdiih ii*4mity-initiA y inor 4ocroased.ovar. nuot oil the Irblevaom firm, fifi zk%.,
*Ww oij thw datsi flaa',
an(t fir-JR-AiO ill-" tw MWAS I#itr*Ujbib%
A." Fb4ib AM
4u sti6w 61W
ihe quOs
lop UP 1 "I; Vq .9
n 4
correlated, -and the iwult9'&p6nd hniewhat on &6'va 90 eel e

disuses their. features' and relative merits [78, 1965]. He
M~fti-, sWe?- heanse: it is neutral regarding factor proportions and UR bafa waybte based on projected sales. The dedfator is typically
Athe numerstor when possible. Thus R&D spending
Is fte t~ted by sales or assets, R&D employees and scientific
-Pe~smiii4v otal employees.
'EaJ~r st essuggested at most a very weak positive association
input intensity and firm size. With 1947 and 1951-1952
L.4 Toowiz.found industry -ranking by .value added per estabPanitively, but weakly correlated with both the breadth
-QUF~rdM', i n~ m research by Airms and research expenditures per
8,196-2]. For 340 -of Fortune's 500 largest firms in 1960, ,L.Aropp12imindustries, Hamberg found the ratio of R&D employMeth ttal employment to be only weakly correlated with total PI yand total assets [31, 1966]. Log-linear regression revealed
altkint~potR&D effort; with respect to firm size exceeded
-UO-A-I iwoy & of the industries. ,4644,A"bC eamer analyzed National 'Science Foundation (NSF) d~~t ainabsi*iJBAD: during .1957-1965, elassified by industry and arw,(w medium-size firm has'1000-4990 employees) [89, 1968].
i.1pal .intensity, esue -by company financed R&D per, net xa~lpiAO averaged 1.4, 1.5, and -al cents, f or- the small, medium and lirff lvi reapectively. Small firm R&D input intensity was no less UA hat e5medium size frms it six. of the twelve industries and -xeded that of large companies in -three industries. Similar conclsos obtained when R&D input intensity was measured alternat~lve R&Dscientists and engineers per 1000 employees.
q sfbur equation econometric model of the firn, fit using vfi-ms over 1957-40, indicated that'research intensity
assoca'te with" frm size ideasutid by sales [57, 1967].
u ile Ngression of the ratio bf R&D employees against
Fotal sesadwihted' mrket thate failed to
V T~oniship'el h between R&D infensity and either variable in
119 im 4181 fi-ms [44, 1'970].
116 -se6%iokaal study of P301 Belgian firms, Philips found that 'A. I 6f tesoarch personnel apon powers of total employment
eld49'.8bie as the statistically best fitting equation [72, 19,711.
that the nuinber of research workers fix Bel"an firms
r thain total employment up to about 1000 enip oees, and i Af &6thg rate -thereafter. eearch intensity (research emi~wee to.1000 fotal employees) pesdkd at abbut 10.000 employees. r i% x64lnship' for flie acgr~egative economy was not replicated in1 Iidh'idiial industries. there eing great diversity in industry patterns. Iffkibt pdustries, however, the elasticity of res'earchemploympent was 1Islta one.,.
WE. A~nkscompared research activity in the UItdSasan Frisheid to test the influec.offr siz [1970'. Hiearudta
side.US.frisiar arger 'than Feconsifabsolute fiM size
wee bdiciet' R&D, then; R&D spending would be6 more hihl con rthtd 'among, th ver lags im n Fne -thant among the

firtns W
nited 4
R&D, activi#1 Adkn%14qgmpm md intensity ihFram"q8 MTe1i" tO4MV* !)I v 'S I I Feeft4 Cai
Thusl it seems ihat with the possibk, ..ei 16f 1*1
hi=rd 4 ny s pport-4*-tbi : wp6thwO x1le
int 'ty, of inhdVt effort'. lndroaSeg jVfth fi"., aijid previous litenkura,.. 5,.,W; VCftCi6&dt-hat
t;oii.ds.-W,'mcreawml'ore'thxttpmpoitiamtoly.withBitmlsW point that varies fwm indt6try to ijtdufkrY,[W, .19M] Md# M Y APPeAA tM* le':6jj8t*tw
size. Sul*6qwrrt
force that general' 6 d4111
Three relatzd-cavents.'i 4arth
try differences: iii the ; region betwwfi e*Ae ,And imWi 'Second, much of the evidence on the, afib*lof vite haMd for other factors effort. &ze:' [to bd eai*
explanatory vari" ond6theg&4tberlaabd rs d* into:account.i-Third thwevidemm:ohbomuLbon*liiuN Va1* C6 .Of 11 smrch participation vatea Tkl nw ty fb"f I
and the, vsam,;:m rit, y of wzR firms d6 uA huVc O su hwtuw progrIUM.
B. Fimm S ', e 4a I LV V
A j eAWq
the iunuei oi A
VIS a mnfiX lvt7. Afr
*Pi i
u .1
trade, perswnpj. j_$ XQX
ng 191 in
-duri 7t] ", 9 Uldum4w
f. ar J au4*roleum, *'n
4r*ea 01 mtho_ *x
to po respon.Obleloi 1 IS
pr A-%, W_ tAye
innovations than of 4 ts produ five,, 4 t th,
producem wvro mp9401P, fqT fewa Ularket share?..,of tb f J T ; sharp, of, t1le. i ac
144B Wthan Unw( J'
0 10%, w
% 14 n lraw .4p 04
,we :.Od Ae
Max, rltlyini o iona loutpufm'tens.' OccurMd at a 4t the sixth Is rgest-J! -m Oteel Wukry h6w*r "r rX
ity- For ho px"Mmd* dYXACM Q Ite a
-.wpig ted, IDTM iy',T k 1,

U till

140img 4W the number of yeen isatd ioei fim in419a asrx
_1q1Aas* iventive oUtputt 35 ear edialisaffe, 196}] Wthine% sa pkiW8Qfivas' doom the 1955: Forkme504i as voluee wa on
meconcentrated& amn g thie laet bfirs their ]RD emplymon, which in turn tended to be slightly more concentrated thau.paet. Smaller finns, in thersample, we6 respensible for a highii~elaive share of inventive activity than sales. A regression of potft O:6 firt three powers of A~les, fit for 448 firms, yielded a
inresin at at decressig rate up to a point of inflection
6C nsle~l Oly three f in the sample were large than
on wasrepenid forthe 14 indhhti-ies and also for4
C""'. industy groups. The essential fidings were unchanged;
blitpa generally inceased'lbss than proportionately with sales
1W 6borpofations
U, ,08y& J M. Samnuels, and J. Tzoanos made a similar stud firms int chemicals' electrical engineering an
*%atthk9,aM~aeh' tol [90, 1972]. T-refirms were found more lil j &s~jat ones to particrpate un patenting. Patents granted d=Mi, wa--166 Were regaressed on the first.two Powers of firms size
ih),profits, and cash flow. The number of patents awre, greased more than proportionatel- wtin techeia iod *Mokatt1 bit the largbt firms, wx~thin the electrical engtI608 fid leebronies industry.In contrast, pateniting decreased ;WA ~ wet he machibs.to61 idtr.
r~tWOM-portd 6n a recent sturdy of some 1200 post-1940 innovatieigin B40h 1[23, 19tl], that 8tnall firms _(under 200 employees) 94bV #d_4Gr- hdet 10 perceh't of industrial innovations -since 1945, eon4* *4tl -about 25 percent of employment and 21 permit of rkiWd~diaiderderag of induries by the share of'small firins
69sv~tions cortesponded fairly well with the ordering thl firmfs ih net output. Small firms costiributed more IT t ini proportionate share of innovations, it ind'bstries charac044 00 mp 'o~yin 12 industria peators 837, 1971]. Exepting toe
f~ii 1 e-t. & s,'large firms' Aare of total patent applications i
aiventive output) wpsas abaitntlly, legs than their saae
sin 1966 (firm size). Patent .applicastions inceasud
ixrm size among members of the chemical ihdustry.
de han'proportionatey with firm size :in engineering and bathufacturers.
tthe, conluion about. the effect of size on inovational effort t~i 0be supported. and reflected in evidence on size and innovational 04#4_. Ikond some magnitude size does not appear especially conSeither innovational effort or output in either this country or iii opein contries where studies have been conducted. However. Sithef dIffer by industry. It seems noteworthy that the chemical dWStry is ctd as an exception both for, the U3nited States and
e ost rnin aers: to he in the technology

bhn er fscartits wit aotraes and- .[2 I81fncso r pree
process Theovatnintyemonpoist.-hdipscvrdta w1h pce lever alindars eatrh. The y e tvr$ mliWTT

coc. He onfclarnudied the. whilect olve d ttrcux~4ino tincsar n o tecnkgiM ogre S.i instis-yhf;k enntreprn uri le9de-sh 0 and h a,"crppet igh_,ti mnrlizetial. ialan i lso emphys.54 s enginoaeerin the r sittfice. b ubro aet spipjqu f entist inthctrael. mnesec ruww fpw atified ithr ressbjetred t

poess h rnigb monopolizationtha cosdee sios: readil 44r ratni," mescuring th t worion om dege fmysyPWJ (8,c20)sarge for p 'ehnloiciatn plI
[82,re1970]. In nrticularenconYqutrat on antrprdrict larhp and ma-"copltiv prt emp.1 penterlg rary inuor prytifc ae aun ieiaton n VII he blw. l.Sm atrh j aiKhv~m, uAont ation andi resec effjectt ttsia nlsq $ ox The hynpothestisn that reseac inut rainteviabei ."on waih" coneasntpotion of industry y sales ar1O~1KP1t&resu0ltsaHrowt found. the fitor- fir cohsma ~xhv bisuseklsscaed withh; reeearh expBl dollrdc and eaey coru related t h xet f t~l lotabs ivelres 20 pernovtofgte firmdcs e ~e"V, fond aeak positieorrltinbewe sAle ollanratin indsreiali ncetto weklypuggess that mreac highly conenty'l'oii forth aonegreatn reerhfort.dusrthsaer, pasbe' 6td w&Ou mrewtdesreadt mon more oentrtned trto to"b t selar innteitienotcmerelthd reflh erctfind~ty iv lain Scherest eeminto of the hym]3 '6 .,Hi]*

ployent [80,a 1967 indepeiacnentrvariab[1 16] Te inexl 1960indstr etmomelyntcnrte sodkig ied.
smre, wdusra or nodrae. cerer'n strs h,9 mIncerersios. Coeffiin of the oncenta'Itkf4, ypohe, -' se
deedniaibeia esrd yvrosidcso l-M

plomet 80 1671 Tdeenen Vrlale iclde'a
indx 90iduty ipoye&t n uc ulttvefc sjl
tech ologcalopiiiiit.Ailianityp fiood i!
POTprucr-r on
Su erduabeornoduale. iiiii frs ts ivii
ve Ib+++arit++

6 d~nc~t m'tin inde repthine'd positive and statisticallyb gii
t Ae, n "emntl explanatory power* of "eoncentratin a
fi isohd~estresarch diffort intensity, measured by ehia
'Of~o~i~t''s :fati of total employment, was regressedoth C ent~r~lon'nde~an qualitive dinmmy variables. The on ta
4 Efit howedoodest significance. In a third testwtre
bf~techni apioyment intensity against concentrain o
thd W&'tditolo clses with the most observations, the aoetr
W 7vee poitive and significant. Scherer concela ha
dthols'f psitve association between concentrationanth fbfr1serchefor was supported. The fact that theice
&6W'iiWAh~tor vowe'of concentration fell sharply on intou 66W'OfPluhimyvarabes was attributed to the positive corrlto b6Ub&160~nfatin and tchnology class. Scherer suggstdhi
mightipport Phillips' hypothesis that teehoo ca
itii~tfi~n Akiin fom 6pportuinity has led to inrae
al tetedthkhypothesis that increases in conentratoar
=i to't6 ~chtcalVigor only in relatively atomistic indutis
t#66nifil~ipbrtnt nce a certain threshold is: crossed. ThesIr Wfthe cocentatio ilex was added in the tlird set of regrsin
140,d~~w"Thelktierf1%tween technical emiployinentinest
Ile bnce trai bnd* to be concave, but the cofien
i~f~wgq~~ed erm'asV ignifidant, in only one ce. In all fout ass *AW61r linp oyent per 1000 employ e reached a prdce M. M.Wt cocentrtieon levels between 50 and 5 percet h 01 ht A& 6-ve apeard to be above 10-14 percent.
tin Mftdftrlsstuy, stimated elasticities of research effort wr e
NOS~ lainstaveagefirm size and an eight-firm concentrate ai
1461. Te, oefint of average firm size was, reported pstv
MW Aln~cat, bt noeffect of concentration was apparent. (oao 6*etimd'tat snce>roduct differentiability may be an imtan lffam% IA f the rerh decision, the effect of concentrationuo r~iTli-ight deendon whether differentiability is high. ots thi inustieswer g>uped by whether the eight-firm concenrto r~ti -&Weded 0 rent. Research levels, adjusted for firmnieb
wtre roupe by both concentration and differertailt
'Comnor onclded that high concentration tended to b so I 66W#withmuch eseach in cases where it is not a major elento
Md-rt'bd~vir; tat s, where prospects for product differenito
Mos-'rcetly Kelys study disclosed that maximum reseact n
t~vi apear tooccr at a 50 percent to 60 percent concenrto ;;ti_&-in loseagremet with Scherer's results [44, 1970]. ikeie
",.ur-frtn oncetraion ratio and its square have significant efl
when ummyval-iables reflecting technological opportuitar
omitte; butare n t gificant otherwise.
pothesis of positive association between ele condntrai"Onand rerh activity by comparing R&D sperdn n

i 'unlikely to be a good proxy for the extent of active rivalry
6'04 ht ation and innovat ve output
'.,Sodie xelating concentration to productivity increase's have found 13 1 gh .6neentration alternatively harmful, neutral, and- helpful. In
4Mch'. 6ase thedegree has been moderate. Stigler [ 92, 1956] compared t e rate-of,'technical progress, measured by the decline in unit labor 1, quirement 1899-1937, in 14 industries of high concentration with
Idustrielsin which concentration was declining and 8 in which it
w, as low. The largest reduction in labor requirements was in industries I In. Which concentration fell substantially during the period and the
illest. in, industries of continued higl concentration. These statis61.641,re its reii forced Stigler's assessment of the broad facts, suggegtiing: that competition of new rivals in an industry spurs rap d technical advan e.' B. T. Allen [2, 19691, updating Stigler's study using data for 19 industries ovet various periods during 1939-1964, found-no. significant differences in productivity growth rates by inja dustry concentration class.
-Uai changes in productivity of labor and horsepower per employeei as indices of technical chan re Phillips found that in 28 U.S.
irAustries over 1899-1939 industries with high concentration or larize facto showed greater technical change [68, 1956]. Carter and Wil iAM1% [12 1957] followed the pattern of Phillips' study for 12 United
'industries for 1907-1948- there was some ositive correlation between the degree of concentration and the increase of output per employee-hour. Weiss found productivity growth in the United States positively related to output growth in both 1937-1948 and 1948IL94 but -Aosigiiificant association was found between average fourfirm concentration and productivity increase in either period [96, 1963] B. Bock and J. Farkas did find a positive association between
productivity and concentration in the United States for 1963 [ 9, 1969
Scherer tested the hypothesis that technological output tends to
Isewith industrial concentration [79, 1965]. The number of in4p#,ry-related patents issued in 1954 to the leading four firms in the industry was regressed against their sales and the four-firm concentration, raifo as well as dummy' variables for technological class. No
support for the hypothesis was found.
To explain his findings about the bituminous coal, petroleum refining, and steel firms (Section V. B. above), Mansfield developed a model hi-it predicts the largest four firms in an industry will tend to account for a relatively large share of the innovating in cases in which (a) the iiivestment, required to innovate is large relative to the size of potential users., (b) -the mmimilm size of firm required to use the innovation pr fiiably is relatively large, and (c) the average size of the largest r6ur firms is much greater than the average size of all potential users of the innovations [53 1968]. This model explained Mansfield's data in the three industries mentioned and in the railroad industry as well Apparently, it wasn't applied to, his 1971 pharmaceutical data. It is faw more elabo'rate than a concentration ratio and goes deeper into striletural aspects of. the industry to explain relative innovational


0. E. WINianWIL W'19651' hund *Siftpls I
Sistent with the data developed by Mansfield. lyeolc four firms' share of inn wations, xela4ve to thek against the concentration ind bdl Rueaf IIAd I die equOtion., Tht iltdimce df concl6vh%660 -Oft Oki tional Perforit=co Of &C 1=Vt font fifibs *39 tound to I* the relative Atre bd ijan coUtrihifted'by
feared to decrtase with tWr morkor ly p6wti- ROr A, ratio above 3"0 perwn! the ltrgest fii*
-their p'roportionzCe share of'imovatibns.
Th inepneInsimpm of st-u-dies 60 -fictutrR60ft Slia, effbit is rewfor'C4 by 0 es dis&&% inL thig Awtio'&* Work ruggiU i* evea "teelmological opporttwiq enough to SoA out the &A '- M xtidbsM Aou h
Of component '-j'.JTI
ta. of industrialstrupture requ;rNt
C. 0 thler O&MMU of SWV*& *$rUefWV T P
The ease of entry into, sh industry. is mvieft"Mat 'Gd:1NkLV*&I" iiii iiiiiiii
that might influence research intensity. Cb=&"r.jNj&rArjWjdWtOI") P riw-iP.Al goal of resoatek wfivity'is aTROm of caw bl"jeft 00ad'ih product differentiatift [15, I& *euA tend to -be low !!LW*
YS vdwe Iigk-ehtry beitriotiot- d6hw
weift PreawL To tek,016 hypo&*W4,averw j*kd formftm size., was ieigr6wedaobst dunmy maftm
OPPortunitieslor product di-fferentiatim and -W'.b4ver&W, dr: 16w entry bax-rier due to WWe GCONOMY1 OWWUW.F= si gnificant effect of the high- U&T or a jonc6ftrv bA*4&*, MWever, moderate entry bst;ersi appmrad to haviw & Padth* 1*41 cant i1apact, aft*r other factors wen: takm: i*& Accaut&jected his origibial hypotheisismid, revmedbis*je*s'. when entry barriers are aiOmm-quite low or:*w-yj WjohS"w research, may -be subetantially Ifts flum at Ntno ifttftVIW A4 W^L Industrial reeear& effort p-p:earg itraftgwt in ifiduattiegeatry barrier, cauming rapid iadhAion tw b& : - W t i" bmiew *low
strongest whert mtj y i"H: has novbmh efiwti.
space industry has become in rem-ihA
g .Y Wblol '11 Y
as the
fk 1.
ijIL91T. Pretfttivo Of ift&atit- 0 *-*
9 19M jbhs' -it
Lin"-mm studying 8*4". tduthi' :;;I%hft4 116 6ff%4 "614, Market Ame, on ift PhtmA Wwunt f 37,119711, N FW&ft]A P$ "d am failum iA iA&ftSktW ihjl tu
Of tilo: WOV114-bo: i4i d"t(If dw tj& Notl to J" ON ft attimpt-W hihov [Al 19"10 "Lx4 RP'Vmv
(Vf Competitive bft0fh6Ib$9 e W
mpts toinikd#ate;
Th Stijn4Wf'WW tff bet Ot Aftity Oft VUM
investigAt64 b* G. Gftbowski and I T iWer
ISAMW Idf : eiow Ch1wiftI fitiftt &A g "194 t 1 gwor.7 a,
Hypothesis that firm R&D expenditures respondP0SWV81*.tAi A A 4*9

ILWO;Utla I rultiongrmesearh intvgcent Thns int Arlp17 ;Xpniuse Indmitnge in its apndailsRD
'Oh, g initsc~hrfwevlkh and dores notue indaplym to eshighrerelns ofdeslieas aciity. Ahnelvn ;16*4d 4wected'al es :paisicalsoehibeith cmte amth h )= AM) outapcth ofmmare suctre n a(inndov &
e Cah flw apead mket oppleotnimpaty sepna toryvarible.ChaneA. Trelepntwsugesons inD the 1infcn iu four cses and ze asn hlarket stren elaemn thre X~tlw &m valua shopsnor te uite lie nolier.d Iniabl xplmatoy poe.Tu tur pe lem etuy berostde co eei domeof rsponivenscsogfs, wt eacxtreers val e viviy

smakig furhercofirlationhip. Iteistenothcea htheri thetrvar
WD wil be trone, ase msilwor k olgpls hinry h otodthatas oncetain inaerfir rivalr & itnite3

Wles~~~~a si negh-imcncetain frmstmay bey funillsign fi
-sipdntet fiancedeeltopmewenent ratine seathecoeficint enevriating af substantni cahuflo cncet tim doe appxthue firmsrwith high liquipditre amo finis. J~t s~is.bA cslneate4,hwv hdoeypoteispl sugg Mg 6*m~~tation ~ sed ourch evl of iqidiyare ncessay f ,t=ebr current prmsofishvnbe iw

F, rA.wv~ cean othsi: an firmaenoym s ust selmnsidcae thb 4nibi aft reuiopilm quft R ikeD inoniead neee

fipostive inf lunost lielit sore1%&D.cladac beuete
Geterabwsinsestudyo selce armwse n ft resipngii frolem ndprusre inouinnternll Totsahs ojctrs svrl vtiaoshvinlddmeasures of divquiidaiio and/o prft ind thei reesiplanaysvfies. on per samp doll fS 195 jirm nera 21US.idutisn goHae offegr-e h ai fR& esne ottal depreoieis addedt ro all dercatd n sas.uThe ofgiqesuid) esfdrlRDdhtd
1go allregesstncmcent s and the esarleifRD l efad yg oieditastn31 of 6] the Indstr xua oo.eddretonofbelindustry (asd higheas ofo ae ardrosg inustmet aida
-thes more liimty antrie deeisatin hbaditeapaet try, RDthentr The. erest onheinashtps cl fI
-toemidsre nlddinternally generated funds, onon resarc an
M ueller o n iesructd And stiate, us: io esachp overx1957a1960,varfour-equin ecrsonoels~nRDepniue perlan Rdocial inves6].]tmen adverteremaue-bhu [57 1967].aK Inrhefitt dequecation for R&]tini tep o deriatio was positivesiift for al ea b nuty h jajt oitrateiaton eidsrD invoavrec eilo eti h~taml coefficientdforgthat year the u investry rbw~ ifrv-ta Muelloe ifta of resource s s rom apitiv taeyintesd rtrn tog the ormet may he ideinenvaibehXl betexpalaignaed f s R&Deac intensity.Futespor i J.i W.ntelliott investigaed inthe detrmbwianBxtrSuyde 53cfrsin 16ou indstie ove919196.
prelmarily srte and"xetinl reia, asincevain f Am orai whether cit insourent fliudvin andidedp through 1966, I he riteresedono R&D pn estthiofieetO

cuet andplaged m &D sen inten fir shar Elot marketiad fe Rdextemnant fRDs o