Partha Dasgupta [a] and Paul A. David [b]
Toward a new economics of science
1. Introduction and motivation: Science, economics and politics
2. The old economics of basic research, and the emergence of a new economics of science
3. Knowledge: Codified or tacit? Public or private?
3.1. Knowledge, information, and the endogeneity of tacitness
3.2. Science and Technology: Public and private knowledge
4. Priority and adherence to the norm of disclosure in the reward
system of science
4.1. Priority and the science reward system
4.2. Priority and secrecy in Science: Public virtue and private vices
4.3. Culture: The enforcement of cooperative rivalry and collective regulation in Science
5. Resource allocation within scientific fields and programs
6. The timing of research programs within Science
7. Policy challenges: Maintaining Science and Technology in dynamic
7.1. Capturing the training and screening externalities generated by open science
7.2. Managing competition for scientists between complementary research activities
7.3. Promoting greater ‘industrial transferrability’ of university research findings
The thrust of the analysis in the two immediately preceding sections has been to show that there are numerous features of the reward system and characteristic institutional structures of open science in the modern West that give rise to resource misallocations and static inefficiencies in the conduct of basic and applied research. Correspondingly, there may be a wide field here for economists specializing in contract theory and institutional mechanism design to familiarize themselves sufficiently well with the detailed internal workings of Science, as they have lately begun to do with regard to the realities of public regulatory bodies and procurement agencies. The problems in these two areas of public economics are not the same, of course, and the sources of the allocative inefficiencies to which we have pointed lie so close to the core of the collegiate reputation-based reward system of the open science system that, in a sense, they may be said to be intrinsic to it. Nevertheless, it is premature to declare that it is beyond the wit of economists to devise modifications of existing institutional procedures that would ameliorate some of the problems identified here. 
Science, however, is not a self-contained system - and indeed, could not survive as such. Rather than risk suggesting that the agenda of the new economics of science is concerned exclusively, or even primarily, with the more static resource allocation issues internal to publicly supported research activities, it is now time for us to recognize that many of the most important challenges facing science policy-makers concern the dynamics of science-technology interactions - the disposition of research resources and the flows of information between the open science and proprietary science communities, and the consequences these will have for the improvement of economic welfare. 
That the open conduct of research in Science offers continual benefits to firms operating in the realm of Technology by making available complementary information - typically basic scientific knowledge - free of charge is, of course, the familiar point with which we began our analysis. Basic research may, of course, yield unexpected discoveries that have immediate practical uses, some of which will be extremely valuable (such as lasers, and enzyme restriction techniques for recombinant DNA research). These are the rare exceptions, however. More typically, the important economic payoffs to society from basic research come in the form of higher rates of return on expenditures allocated to applied research, in both the private business R & D sector and in publicly funded, mission-oriented or ‘applied: R&D. 
61 See, for example, Laffont and Tirole (1993) on the distinction between regulation and procurement (briefly, that the latter is a principal-agent relationship in which the principal is also the buyer of the commodity supplied, whereas regulation refers to situations where a firm acts as an agent of the government in supplying commodities to third-party purchasers). In its attention to the realities of the institutional environment, and the informational, contractual and political and administrative procedural constraints upon the public regulator (the principal), the ‘new regulatory economics’ exhibits many points of kinship with the spirit of the analysis explored here. The key additional features with which the new economics of science has to deal are that the agents in question (the researchers) are supplying information products rather than conventional tangible goods and services, and have been assigned collective responsibilities for regulating many aspects of their activities.
62. On science-technology interactions and interdependences, see, e.g., the empirical studies in Grupp (1992), the survey in Freeman (1992) of modern formal institutions supporting science-based innovative activity, and the treatment of technological change as a dynamic system involving feedbacks between basic and applied research activities given in David (1993b).
63. To the extent that basic research funding is devoted to fundamental scientific inquires, the latter have been likened (by David, 1993b p. 230) to providers of “maps to guide mission-oriented researchers, directing explorers on the applied science frontier to the more fruitful areas, and sparing them the wastage of time and resources in searching barren regions or trying to cross unbridgeable chasms”. The role of basic research instrumentation advances in creating spillovers to applied R&D is also discussed by David (1993b, pp. 222-225).
While it has been customary for economists to emphasize the spillovers of information about the material world, it seems no less important to notice another informational channel through which the existence of open science institutions conveys benefits to R & D activities that are being carried on with immediate commercial goals in mind: the educational and evaluative activities that are closely coupled with academic research.
Open science discloses information about research methods and findings, and in the process about the abilities of the researchers themselves, which can be captured by private producers who transfer scientific personnel from the academy to their domain. Disclosure and peer evaluation mechanisms make available, at very low cost to managers of company R & D laboratories, a great deal of information about the qualities of scientists who they might wish to recruit as employees.  Thus, were there no institutional structures corresponding to those of Science (as we have defined it), Technology’s knowledge of the ability and experience of its scientific research personnel would be far less complete than it is. This would add substantially to the expected costs and the uncertainties involved in company financed research projects - even if the distribution of scientific abilities and training in the pool of potential recruits were unchanged. On average, their value to prospective employers would be lower on account of the greater uncertainty surrounding their individual qualities and the nature of the knowledge that they had acquired in the course of their training and research experience. Thus, the modern research universities’ productivity in training and evaluating the work of many more researchers than they collectively can permanently absorb ought not ipso facto to expose them to being castigated (as they sometime have been) for lacking social responsibility and a capability for manpower planning, or for disappointing the career aspirations instilled in many of their graduates. Quite the contrary; the export of scientists and engineers from the academy into industrial research is potentially the most important and salutary among the mechanisms available for effecting knowledge transfers that bring economically valuable ‘spillovers’ to the commercial R&D sector, and for creating informational networks that help impart industrially relevant direction to academic researchers and teachers..
Proper policy measures undertaken by government agencies, educational institutions and business corporations acting in concert are necessary, however, to assure that these potentialities will be exploited. There is nothing to guarantee it will happen spontaneously, and arrangements that evolve historically as legacies of responses to past problems are likely to drift far from the currently optimal state. Such a condition is illustrated by a brief glance at the prevailing arrangements governing the training of graduate scientists and engineers in the US. Historically, American research universities have adapted themselves rather readily to take advantage of whatever funding opportunities were created by federal and state government policies (or, should we say, by the collection of ad hoc legislative and administrative decisions that usually passes for ‘policy’ in this area), in this instance by organizing the subsidized education of science and engineering PhDs as a by-product of publicly supported research projects. From one point of view, this seems quite the most natural thing for these institutions to have done; it could be readily
64. Academic scientists, of course, form a pool of potential recruits upon which industrial research organizations regularly draw. Recent sectoral retention rates are very similar for US doctoral scientists and engineers employed in business and industry, on the one hand, and those employed in universities and colleges, on the other. Within the 1973-1987 interval, a 2-3% movement occurred in either direction between the industrial and academic sectors during selected 2-year periods (see National Science Board, 1987, p. 94; National Science Board, 1989, pp. 118-119, 325). According to unpublished National Science Foundation (NSF) survey data, 5% of doctoral scientists and engineers who were in industry in 1975 had moved to employment in colleges and universities by 1985; 8% of those in colleges and universities in 1975 had moved to industry by 1985. However, because the stock of researchers in the academic sector is much larger (more than twice that in industry), this similarity in rates of mobility implies that much greater numbers of academic researchers move to industry than vice versa. According to unpublished NSF survey data, in 1985, 16% of doctoral scientists and engineers in industry were employed in colleges and universities in 1975; only 2% of the stock employed in the academic sector had been in industry a decade earlier-(NSF, 1988, pp. 14-15).
construed to be consistent with the two-part structure of the ‘contract’ that we, have argued would have to be offered to qualified researchers under the patronage system. After all, is not graduate teaching, and especially training in research methods, the form of regular salaried employment that is most immediately compatible with the instructors’ ongoing engagement in research? There is little dispute over the contention that these two activities are mutually complementary: participation in research enhances the effectiveness of graduate teaching, while the use of graduate and. postdoctoral ‘trainees’ as research assistants, in turn, represents a significant form of subsidy for academic science.  Indirectly, of course, the arrangement also can convey important subsidies to the eventual non-academic employers of scientists and engineers, because it must reduce the costs of hiring PhDs when the latter have been trained under the auspices of publicly funded research projects. They did not have to pay the full costs of their acquisition of the information and experience which they will be expected to put at the disposal of their employers, and for which investment they otherwise would need to be compensated.
That is all very well, save for the one unfortunate ‘hitch’ that has developed in the operation of this ingenious machinery for subsidizing both open science and the transfers of knowledge (embodied in trained scientists) to the business sector. The growing demand for trainee research assistants in university laboratories was allowed to become a major factor, perhaps the major factor, driving the system and causing the population of academic scientists to reproduce like sunfish. It has been estimated that under the prevailing setup, the majority of PhD scientists in the US each train about 15 new doctorate-holding researchers, on average, over the course of their own academic research careers.  Evidently, this would be an unsustainable situation were it required that the benefits of university training externalities be captured somewhere in the nation’s economy - the supply of scientists and engineers that is being generated has, for some time, been outrunning the capacity of the academic and non-academic research sectors to absorb them in productive employment. That the dynamics of the market for new doctorate-holders in the sciences and engineering are characterized by lagged responses and, consequently, by periodic episodes of temporary excess supply or excess demand is well known, but the imbalance to which we are referring is a structural and persistent one, which manifests itself in the high and rising proportion of doctoral recipients in science and engineering who are foreign nationals holding temporary residence permits. By 1990, among the new PhDs in the physical sciences, mathematics, computer science and the life sciences, the proportion who were temporary residents had risen to 28.4%, and, among engineering PhDs it had reached 48.7%.  Educational and science policy makers in the US might well conclude that by thus subsidizing the growth of the international pool of scientists, it can cheaply provide itself with well-prepared and motivated trainee-research assistants and be in a position to select the most talented young researchers, thereby, maintaining at least cost the vitality of its basic science establishment. Reconsidering its immigration policies and encouraging the universities to prepare graduates for work in the R & D laboratories of the US corporations is an alternative course of action worth serious consideration under the rubric of improving university-industry knowledge transfers within the national system of innovation.
65. Whether it is the most efficient way to subsidize academic research is less clear. On the one hand, the support services may be costly, inasmuch as the research assistant staff is turned over rapidly, talented assistants cannot be retained for long, and it is necessary to train one cohort of students after another to perform routine laboratory tasks. On the other hand, the incentive structure of the trainees is such that they do not need to be paid highly to induce them to try to provide a quality of service that will bring favorable notice from their instructor-employers.
66. See estimates attributed to David Goodstein of Caltech in The Scientist, 20 September 1993, p. 5.
67. The cumulative shares of temporary residents among new PhDs during the period 1986-1990 were 25.7% and 46.8% in the case of the scientists and the engineers, respectively. Among the recipients of doctorates in the four science fields cited in the text, 63% were US citizens in 1990, compared with an average of 69.2% over the period 1986-1990; the corresponding figures among the doctorates in engineering were 42.9% in 1990 and 45.5% in 1986-1990. These percentages were calculated from the data for US citizens, temporary US residents and permanent US residents (excluding degree recipients whose nationality was not known) as reported by the National Science Board (1991), Appendix Table 2-24).
That course of ‘readjustment’ in the system seems much to be preferred to two others that may be contemplated, namely, cutting the level of indirect support for the university training of scientists until the intake of foreign students is reduced, or pushing the academic sector into using qualified and inexpensive foreign trainees to carry out a larger volume of applied research on behalf of national commercial enterprises. To support this argument, we shall examine the latter two potential science and technology policy thrusts in the following subsections, taking them in turn.
Academic science and industrial science are complementary activities when viewed from the societal perspective, but as professions they are distinct and offer contrasting mixes of monetary payments, peer recognition, and working conditions, on the basis of which the two realms compete for creative talent. A commonplace observation is that the material component of the rewards can be far greater in industry than in university research supported by public and private patronage. Thus it would seem rational for materially minded researchers to want to work in the open science environment, where others could be expected to assist them by sharing knowledge with them, but to plan to remain there just until they had made some discovery or invention that they could sell (either as a patented device, or as a trade secret) for commercial exploitation or further development in the hands of some proprietary research entity. Such attractive options, however, are rarely available. More often than not, both scientific research projects organized in academia and those organized in government laboratories demand pre-commitment from their participants. If the findings are proprietary and are not to be disclosed publicly (unless authorization is obtained), one has a project pre-committed to Technology, whereas pre-commitment to public disclosure is the hallmark of projects in the realm of Science. To the individiual researcher, then, the basis of choice between these alternative commitments is provided by his/her expectations of the pecuniary and non-pecuniary returns under each of these institutions. That being the case, it is not so apparent why any scientist with the goal of material success in mind would not immediately want to be signed up, at higher pay, to do proprietary research.
Aside from the obvious point that for some individuals the academic lifestyle may hold strong attractions, is there anything that enables the academic research sector to provide itself with young talent in the face of competition from industry? One consideration is suggested by the notice given a moment ago to the informational value to potential employers of being able to recruit from a pool of researchers who have worked in the open science sector. Even those who acquired a scientific training with a view to eventually putting it to use in the industrial R & D sector may have an incentive to enter academic science initially, and remain actively engaged in research there at least for a while. Doing so gives them greater leave to publish their findings, thereby signalling their innate abilities and acquired expertise to prospective employers in the other sector. Signalling in this way is quite compatible with preserving the option of continuing in Science, should they manage to win an attractive place there. In the extreme, one can imagine that embarking upon a research career in academic science is a form of investment undertaken purely for the purposes of this signalling. Let us see what this simplified way of accounting for the co-existence of the open and the proprietary research sectors implies about the requirements for maintaining a proper balance in the distribution of personnel between them.
Provided it is not overly costly to forego proprietary R & D employment by taking a postdoctoral research position, the young scientists who believe they have an exceptional talent for research would wish to join the Republic of Science, because it is they who benefit most from providing employers in Technology with a better quality signal as to their abilities. Once they have entered Science, however, the remaining lot represent a truncated distribution (with a lower average research aptitude). If they realize that employers would recognize that, the best among the remaining group would opt to go into Science as well, thereby making it necessary for the next best group to follow them, and so forth. In fact, in the thought-experiment we have set up, the process continues until they all embark on an initial so-
journ in Science - even at some material cost to themselves. This goes some way towards explaining why the academic sector initially retains so large a proportion of the current flow of newly trained graduate scientists and engineers, without supposing that the latter have been ‘socialized’ or otherwise infected by their professors with a desire for the lifestyle of university teaching and research. It aso accounts for the numerical preponderance of the outflow of postdoctoral scientists and engineers leaving university research for industry, without implying that those who move on are doing so because their ambitions for an academic science career have been frustrated. third point o note is the rough correspondence between this extreme, signalling model and the observation that the largest percentage of doctoral scientists and engineers in research who are going to leave their initial academic employments do so within a few years after entering; those who enter Science primarily for the purpose of signalling would not wish to tarry there overlong, especially if their training was being rendered obsolete (from the viewpoint of prospective employers in industry) by the rapid advance of the research frontier in their area of specialization.
The key qualification in the foregoing line of analysis is that the acquisition of credentials and signalling opportunities in Science should not be too costly for the individual researchers. (Otherwise they will directly enter research activities within the realm of Technology and try their luck there, or decide to abandon research as a career entirely.) his cost reflects, among other things, the foregone salaries in Technology and the length of time spent ‘queueing’ for access to interactions with senior established research leaders and specialized research facilities that may be required to produce results worthy of publication. If, over a period of time, the value of privately appropriable commercial profits from the production of information increases at a sharp rate, the foregone earnings would increase correspondingly. 
Should researchers be sufficiently myopic when weak public patronage of university research raises the penalties of deferring an industrial science career, we would expect to see a constriction of the inflow into Science and an ageing of the population of researchers occupied there. Following on from that would be a corresponding reduction in the benefit that Technology could derive from new additions to the stock of public knowledge, and from the opportunity to select from among researchers who had established a track record under the rules of open science. If prolonged, the constriction of these forms of spillovers would tend to have a substantial depressing effect upon the rate of technological progress, since technical enterprises would now have to conduct more duplicative research than they found necessary in earlier periods.
The foregoing analysis suggests that once material incentives have deteriorated to the point that it is recognized that very talented researchers are not remaining in university science to acquire a subsequently valuable signal, the signal value of starting off in academia itself will tend to be depressed - especially for those who may also wish to indicate that they are not unresponsive to prospects of material rewards. The signalling motive may be sufficient to compensate for some economic disadvantage of remaining in, the university sector, but it is not robust enough to stand on its own when weak public patronage of academic research causes the diversion of a substantial part of the distribution of newly trained scientific talent away from Science. Indeed, there may be a point at which it would evaporate completely, because failure previously to have been drawn out of the ivory tower and into the corporate laboratory would become a negative signal for prospective employers, a blot on the young researcher’s track record. To keep the balance on the right side of that critical point, Science is constantly in need of shoring up through public patronage so that it may initially command a substantial share of the most talented researchers in the face of competition from Technology.
In fact, the more closely the two communities resemble each other in terms of the actual research work that is being performed, the more vulnerable Science becomes. Unless young scientists are culturally conditioned to value scientific
68. Lower stipends for postdoctoral appointees, less well-equipped university laboratories, and senior academic scientists whose time was more and more occupied with proposal writing, administration and other tasks that rendered them less available to work with neophyte scientists would, likewise, raise the expected ‘cost’ of an individual’s investment in signalling his or her research capabilities.
inquiry for its own sake, or to desire fame and public recognition, or to derive satisfaction from teaching and the academic lifestyle - all of which may create considerable adjustment problems if and when they move into employment in industry - and unless the conditions supporting open scientific research are improved, the inflow of intellectual talent into Science eventually will be curtailed by the prospects that neither form of scientific research career is likely to remain economically rewarding. For it must be recognized that once the flow of scientific talent into open science is diminished, the profitability of firms’ investments in R&D in the future is likely to be affected adversely, which will reduce the future demands for scientists of proven research ability there, and undercut the signal-acquisition motivation for individuals to embark first upon a career in Science: In this way the complementarities between the open science and the proprietary R & D sectors can result in the dynamic system descending into a contractionary spiral, in which less and less investment is made in the production of new knowledge - public or private.
One implication of the foregoing dynamic analysis is that the repercussions of sharply curtailing support for training graduate and postdoctoral scientists and engineers may be far more destructive than linear extrapolation of observed responses of the system to modest funding cutbacks would suggest. It is important to recognize that the dependence of knowledge-based industrial development upon the science-technology nexus has made the stability of economic growth at high levels a hostage to rather fragile features of the cultural and institutional environment, features that require protection rather than assault from political and business leaders. It is the taste for the lifestyle of science, the compatibility of research with teaching, and the persistence of public authorities in subsidizing science at a level to which none of the constituents would appear willing to subscribe that has prevented the collapse of the economic structure erected upon a high level of open science activity. There is today a worrisome inclination to take all that has been achieved for granted. What can at best be politely described as a shocking lack of comprehension of the economics of science reveals itself all too frequently in the glibly confident pronouncements of faith in the workings of the market that continue to emanate from ‘conservative’ policy circles on both sides of the Atlantic: governments are being told, in effect, that if there is some research to be done that would be of immediate social benefit, the private sector is the natural place for it to be done, and - as a corollary proposition - public research support for science largely displaces corporate R&D funding that would have every incentive to accomplish the task more cheaply.  Under conditions approaching the state of ‘universally privatized. science’ that such ideologues call for, an unbalanced research regime might continue to generate economic growth through the exploitation of the scientific and technological knowledge base, but sooner or later, economic progress almost certainly would lose the sustained character that has been taken by many scholars to distinguish ours from previous historical epochs.
So long as university research supported by public and private patronage remains institutionally distinct from the world of profit-motivated corporate R&D, the general problems of exchanging information among learning entities will
69. Just as the assault on public funding for university-based science seemed to be abating in the US, it apparently gained adherents in the UK (see, for example, the account given by Adrian (1992 p. 528) of the views advanced by Terence Kealy in a pamphlet issued by the Centre for Policy Studies in London, a body often described ‘as ‘a right wing think-tank’. The (London) Times for 29 April, 1991 quoted a report for the Institute for Economic Affairs by Sir Douglas Hague as saying: “The best preparation for becoming a scholar is now not necessarily a post in a university but in a high-technology company and unless universities come to terms with this challenge they could face failure... People outside the universities will increasingly be working in similar ways with similar themes and with similar talents to those within; and they will often do so more innovatively and with greater vigour because they will come to what they do untrammelled with academic traditions, preconceptions and institutions.” Sir Douglas’s view of this new ‘competitive environment’ for universities appears to have been a welcoming one; his report is reported to have recommended that British universities’ monopoly of higher education be broken, and more organizations (including those from commerce and industry) “should be allowed to award degrees and compete for the finance available”.
manifest themselves most visibly at the boundaries between the two spheres. Lately, the difficulties that, appear to cause delays and failures in the process, of transferring basic research findings from university laboratories to corporate ,R & D organizations have emerged as a focal point for expressions of concern in science and technology policy circles in the US and western Europe. Some of the obstacles identified have their roots in the existence of divisions between the respective cultures of academic science and corporate R & D. This supposed cultural barrier to information dissemination might well be accepted as the downside of a state of affairs that is beneficial in other respects, for we have seen that the establishment of a distinctive, open science ‘culture’, identified with a set of prescriptive norms for universalistic, cooperative behavior, plays a valuable role in permitting the maintainance of effective informal networks of communication among university-based researchers.  Policies intended to promote greater transferrability of basic science findings by eradicating the open science culture in order to forge ‘a more perfect union’ between academic and corporate researchers may indeed be successful in capturing some immediate economic rents by more intensively exploiting the extant stock of basic scientific knowledge, but they risk fragmenting the networks in which tacit elements of that knowledge base resides, and so are likely to jeopardize not only the future growth of basic knowledge, but also the flow of economic benefits deriveable from the existing stock of knowledge.
Readjusting institutional norms to enlarge the social boundaries of the research community, as a way of facilitating the transfer of new findings from academic science to industrial laboratories, is only one among many proposed solutions on the table, or already moved onto the testing bench.  Much attention has recently been devoted to the promotion of university patenting and technology licensing initiatives, creating intellectual properties that offer profit-seeking firms an inducement to invest the complementary R&D that will be required to create commercially viable new products and processes based on the knowledge uncovered by academic researchers. Even though some delays and restrictions on the publication of findings are typically imposed to allow time for, the preparation and filing of patent applications (either by university authorities or corporate sponsors), such practices are seen as a compromise solution that is more compatible with the academic science community’s norms of disclosure than the alternative of protecting innovation rents by recourse to secrecy.  Here too, however, the task of the university as ‘technological information broker’ and ‘innovation entrepreneur’, seeking to transform the scientific discoveries and inventions of its faculty into intel-
70. Viewed from outside, however, open science culture(s) may be perceived to be (less benignly) preoccupied with promoting external reputational status - largely for the benefit of the participants, and even at the cost of jeopardizing the immediate interests of the organizations that employ them as researchers. David (1991) analyzes historical problems of principal-agent relations involving scientific networks in some detail. The problems of ‘culture clashes’ in a more modern setting are illustrated vividly by the reports of a recent internal security review conducted by officials of NASA at the Ames Research Center in Mountain View, California. According to the New York Times (22 November 1992, Sect. 1, p. 19):
the center had not properly handled ‘sensitive technology’ and was considered at ‘high risk for hostile intelligence operations’... NASA said it did not believe there were similar problems at other centers, noting that ‘the culture and environment’ at Ames ‘were found to be the underlying cause of NASA’s vulnerability’. Workers at Ames said the atmosphere there is more like that of a college campus than a Government laboratory, with people being more concerned with moving and talking freely than with following all security procedures. ‘The culture is strongly biased toward maintaining an academic reputation, rather than meeting US industry and national needs’, the agency said.
71. The scholarly literature on the subject is already extensive and is growing rapidly, as may be seen from Battaglini and Monaco (1991), Blume (1987), Blumenthal (1986), Fusfeld and Haklisch (1984), Kuhlmann (1991), Stankiewicz (1986). David and Steinmueller (1993) provide an overview of the issues raised by recent experiments with closer forms of university-industry research collaboration. See Hoke (1993) for a journalistic treatment of university administrators’ and scientists perceptions of the opportunities and challenges.
72. See Eisenberg (1987) for an extended discussion of the legal issues, which recognizes that the fit between the requirements of intellectual property protection under the patent system and the norms and reward system of academic researchers is far from perfect. David (1993a) examines points of congruence and non-congruence between the two modes of organizing research from an economic resource allocation standpoint.
lectual property that it can license to business firms, is not so easy. What complicates it, and restricts the university’s effectiveness as a scientific information transfer agent, are the informational asymmetries among the parties involved, and especially the difficulties of specifying and monitoring the content of the tacit knowledge transfers that often must accompany the transmission of codified knowledge - if the full commercial value of the latter is to be realized. As a rule, it is very awkward and costly, if not impossible, to write a precise contract for the purchase of tacit knowledge. To go to such lengths, however, really may not be necessary when the tacit and codified materials are strictly complementary, that is to say, when it is essential to enable the patent licensee to implement commercially the information disclosed by the patent. Arora (1991) shows that if the codified part can be owned, and its use licensed, the licensor will have sufficient incentives to provide (for compensation) the socially optimal amount of tacit information. 
Seen from this angle, a key structural problem impeding effective applied research transfers from universities is that even were the university ready to grant an exclusive license to a patent assigned to it by a faculty researcher, the university officers responsible for technology licensing do not possess the complementary tacit information (‘know-how’) that would make the patent really valuable to a licensee. The faculty researchers either have it, or are most likely to be in an advantaged position to develop what needs to be learned about the application of their work in concrete contexts other than the ones with which they already are familiar. But as their interests and those of their university’s technology management program will rarely be perfectly congruent, they may well decline to supply this knowledge, on the reasonable grounds that they have more intellectually interesting, or more socially useful, or even more financially remunerative things to do with their time. To address this awkwardness directly, without allowing them to capture all the university share of the economic ‘rents’ from the invention,  it would be necessary to alter the nature of the academic researchers’ relationship with their institution. Non-scientist administrators would have to be able to tell faculty researchers what they should work on, that is, to direct them to make best faith efforts to deploy their scientific expertise in furthering their employer’s legal interest - however the university chose to define those interests! Contractual ‘reforms’ of this sort, involving the loss of research scientists’ autonomy, and the supplanting of the open science reward system by another that would both loosen the nexus between teaching and research in the academy, and impede rapid public disclosure of discoveries and the cooperative sharing of novel research techniques and intermediate findings, would be tantamount to a complete removal of university-based research from the domain of the Republic of Science. Along with government-run research laboratories, university-based science would thus be dragged into the sphere of organizational and institutional structures that we associate with the Realm of Technology. To move towards altering the balance between open and ‘restricted’ science in this direction would be to jeopardize the fruitful symbiotic relationships between the two distinctively organized and functionally differentiated spheres of the modern system for generating scientific and technological knowledge. It hardly can commend itself as a sensible course of institutional readjustments and reforms intended to promote even the ostensibly worthy national goal of stimulating innovation and long-run economic growth, let alone the narrower purpose of relieving the public purse of some (small) part of the
75. Arora’s analytical and empirical studies detail the way tacit information is transferred between business firms that have differing technical capabilities, in conjunction with patent licensing agreements between them.
76. The university, presumably, would need to impose a fixed charge against the income derived from the licences to cover its out-of-pocket costs; it might also attempt to extract some rent from the faculty patentee, in exchange for future research support, or other conditions of employment. It should be apparent that these fantasies do not constitute recommendations of a course of action that is thought desireable; that they leave unaddressed the issues of equity that would arise among colleagues who believed that their efforts had contributed to the success of the patented discovery, and fail to consider the implications of such institutional arrangements for the management of conflicts of interest, and distortions of university procedures for internal resource management and academic advancement.
costs of supporting university science and engineering.
The broad message, emerging from recent advances in the economic analysis of science that have been reviewed here can be expressed in the following four propositions.
(1) Although the institutions and social norms governing the conduct of open science cannot be expected to yield an optimal allocation of research efforts, they are functionally quite well suited to the goal of maximizing the long-run growth of the stock of scientific knowledge - subject to the constraints on the resources that society at large is prepared to make available for that purpose.
(2) Those same institutions and. social norms, however, are most ill suited to securing a maximal flow of economic rents from the existing stock of scientific knowledge by commercially exploiting its potential for technological implementations. The distinctively different set of institutional arrangements, and different modes of conduct on the part of researchers, that accordingly have been contrived for the latter (technological) purposes unfortunately leave unsolved the problem of securing the right amount of resources for the conduct of open science. Here, adequate public patronage is critical and warranted.
(3) The organization of research under the distinct rules and reward systems governing university scientists, on the one hand, and industry scientists and engineers, on the other, historically has permitted the evolution of symbiotic relationships between those engaged in advancing science and those engaged in advancing technology. In the modern era, each community of knowledge seekers, and society at large, has benefited enormously thereby.
(4) The institutional machinery which has been performing these vital functions for our society is intricate, jerry-built in some parts, and possibly more fragile and sensitive to reductions in the level of funding for open science than often may be supposed. For all their importance to the modern economy and polity, the social mechanisms that allocate resources within the Republic of Science are still too little understood, and remain vulnerable to destabilizing and potentially damaging experiments undertaken too casually in the pursuit of faster national economic growth or greater military security.
The foregoing propositions provide basic tenets ‘to guide discussions of concrete problems and proposals that fall within the purview of decision-takers responsible for science and technology policies. Obviously, they are too general to have positive prescriptive value, and are meant to be largely cautionary. If they are found to have some utility, it will reside not in instructing us what to conclude about this or that policy question, but rather that the economics of science can help frame better science and technology policies only insofar as it comes to grips with the logic and the performance of the specialized institutional structures that organize the ‘production and distribution of that very peculiar asset: scientifically reliable knowledge.
The authors are grateful for comments and suggestions on earlier drafts received from Richard Nelson, Laurence Rosenberg, Peter Temin and Harriet Zuckerman, from Ashish Arora, Ed, Steinmueller, and other members of the (Fall Quarter, 1990) Technology, Organization and Productivity Workshop at Stanford University, from Chris Freeman, Keith Pavitt, and other participants in the SPRU Seminar at the University of Sussex (Spring 1991), and from Alfonso Gambardella and other members of the IEFFE Seminar at the University of Bocconi, Milan, in April 1992. Weston Headley and Philip Lim provided able research assistance in the early phases of this project. The present version has benefitted from the comments of two anonymous referees. This also is an appropriate place to acknowledge the financial support provided for this and related research by the Mellon Foundation Program on “Science and Society”, and (for P.A.D.) from the American Academy of Arts and Sciences, and the Information and Organization Program of the National Science Foundation (Division of Information, Robotics and Intelligent Systems, Grant IRI-8814179-02). The Center for Economic Policy Research (CEPR) of Stanford
University provided administrative and other support for the research funded by those grants.
Adrian, R. (Lord), 1992, Do We Need Science Policies? Proceedings of the American Philosophical Society 136 (4), pp. 526-532.
Arora, A., 1991, The Transfer of Technological Know-How to Developing Countries: Technology Licensing, Tacit Knowledge, and the Acquisition of Technological Capability, unpublished PhD Dissertation, Stanford University.
Arrow, K.J., 1962, Economic Welfare and the Allocation of Resources for Inventions, in: R.R. Nelson (Editor), The Rate and Direction of Inventive Activity: Economic and Social Factors (Princeton University Press, Princeton, NJ).
Arrow, KJ., 1971, Political and Economic Evaluation of Social Effects and Externalities, in: M. Intrigator (Editor), Frontiers of Quantitative Economics (Contributions to Economic Analysis Vol. 71, North-Hollancj, Amsterdam).
Arrow, KJ., 1974, The Limits of Organization (Norton, New York).
Arrow, K.J. and W.M. Capron, 1959, Dynamic Shortages and Price Rises: The Engineer-Scientist Case, Quarterly Journal of Economics 73, 292.
Axelrod, R., 1984, The Evolution of cooperation (Basic Books, New York).
Battaglini, A.O. and F.R. Monaco (Editors), 1991, The University within the Research System - An International Comparison Handbook of the Law of Science: Comparative Studies, Vol. 1 (Nomos Verlagsgeselischaft, BadenBaden).
Baumol, W,J., and W.E. Oates, 1975, The Theory of Environmental Policy: Externalities, Public Outlays, and the Quality of Life (Prentice-Hall, Englewood Cliffs, NJ).
Bernal, J.D., 1939, The Social Function of Science (Routledge & Kegan Paul, London).
Blank, D.M., and G.J. Stigler, 1957, The Demand and Supply of Scientific Personnel (National Bureau of Economic Research, New York),
Blume, S., 1974, Toward a Political Sociology of Science (Free Press, New York).
Blume, S., 1987, The Theoretical Significance of Co-operative Research, in: S. Blume et al. (Editors), The Social Direction of the Public Sciences, Sociology of the Sciences Yearbook, Vol. 7 (JAI Press, Greenwich, CT).
Blumenthal, D., et al., 1986, University-Industry Research Relationship in Biotechnology: Implications for the University, Science 232 (June).
Boorstin, D.J., 1984, The Discoverers: A History of Man’s Search to Know his World and Himself (Random House, New York).
Cohen, W.M., and D.A. Levinthal, 1989, Innovation and Learning: The Two Faces of R&D, Economic Journal 99 (397) 569-596.
Cole, J.R., and S. Cole, 1973, Social Stratification in Science, (University of Chicago Press, Chicago, IL).
Cole, S., 1978, Scientific Reward Systems: A Comparative Analysis. Research in Sociology of Knowledge, Sciences and Art., Vol. I, 167-190.
Cozzens, S.E., 1990, Autonomy and Power in Science, in: E. Cozzens and F.G. Fieryn (Editors), Theories of Science in Society (University of Indiana Press, Bloomington, In.) pp. 164-184.
Dasgupta, P. 1986, The Theory of Technological Competition, in: J.E. Stigliiz and F. Mathewson (Editors), New Developments in the Analysis of Market Structures, (MIT Press, Cambridge, MA).
Dasgupta, P., 1988, Patents, Priority and Imitation, or, The Economics of Races and Waiting Games, Economic Journal 98.
Dasgupta, P., and P.A. David, 1987, Information Disclosure and at the Economics of Science and Technology, in: G. Feiwel (Editor), Arrow and at the Ascent of Modern Economic Theory (New York University Press, New York).
Dasgupta, P. and P.A. David, 1988, Priority, Secrecy, Patents and the Economic Organization of Science and Technology, (Center for Economic Policy Research Publication No. 127, Stanford University, March) (forthcoming in Science in Context).
Dasgupta, P. and G. Heal, 1979, Economic Theory and Exhaustible Resources, (James Nisbet and Cambridge University Press, Cambridge).
Dasgupta, P.S., Marglin and A. Sen, 1972, Guidelines for Project Evaluation (United Nations, New York).
Dasgupta, P., and E, Maskin, 1987, The Simple Economics of Research Portfolios, Economic Journal 97.
Dasgupta, P., and J.E. Stiglitz, 1980a, Market Structure and the Nature of Innovative Activity, Economic Journal 90.
Dasgupta, P., and J.E. Stiglitz, 1980b, Uncertainty, Industrial Structure and the Speed of R&D, Bell Journal of Economics 11.
David, P.A. 1984, The Perilous Economics of Modern Science, (Technological Innovation Program (TIP) Working Paper, Center for Economic Policy Research, Stanford University, August.
David, P.A., 1991, Reputation and Agency in the Historical Emergence of the Institutions of ‘Open Science’, Center for Economic Policy Research Publication No. 261, Stanford University, April.
David, PA., 1993a, Intellectual Property Institutions and the Panda’s Thumb: Patents, Copyrights, and Trade Secrets in Economic Theory and History, in: Global Dimensions of Intellectual Property Rights in Science and Technology, M.B. Wallerstein, M.E. Mogee and R.A. Schoen (Editors) (National Academy Press, Washington, DC).
David, P.A., 1993b, Knowledge, Property and the System Dynamics of Technological Change, in: Proceedings of the World Bank Annual Conf. on Development Economics: 1992 L. Summes and S. Shah (Editors), (Washington, DC World Bank Press).
David, P.A., D.C. Mowery and W.E. Steinmueller, 1992, Analyzing the Economic Payoffs from Basic Research, Economics of Innovation and New Technology 2 (4), 73-90.
David, P.A., and W.E. Steinmuller, 1993, University Goals, Institutional Mechanisms and the ‘Industrial Transferability’ of Research. Center for Economic Policy Research, Proposal for an AAAS Study Project Conference, October.
Eisenberg, R.S., 1987, Proprietary rights and the Norms of Science in Biotechnology Research, The Yale Law Journal, 97 (2), 177-231.
Freeman, C., 1992, Formal Scientific and Technical Institutions in the National System of Innovation, in: National Systems of Innovation, B.-A. Lundvall (Editor) (Pinter, London)
Fudenberg, D., and E. Maskin, 1984, The Folk Theorem in Repeated Games with Discounting and with Incomplete Information, Econometrics 54, pp. 533-554.
.Fusfeld, H.I. and C.S. Haklisch (Editors), 1984, University-Industry Research Interactions, Pergamon Press, New York).
Gaston, J., 1970, The reward system in British Science, American Sociological Review, 35, 718-730.
Gaston, J., 1971, Secretiveness and Competition for Priority of Discovery in Physics, Minerva 9 (1).
Greif. A., 1989, Reputational and Coalitions in Medieval Trades, Journal of Economic History, 49 (3), 857-882.
Greif, A., 1992, Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies, (Working Paper, Department of Economics, Stanford University, November).
Griliches, Z., 1957, Hybrid Corn: An Exploration in the Economics of Technological Change, Econometrica 25.
Griliches, Z., 1960, Hybrid Corn and the Economics of Innovation, Science, July 29.
Grupp, H. (Editor), 1992, Dynamics of Science-Based Innovation, (Springer, Berlin).
Hagstrom, W.O., 1965, The Scientific Community, (Basic Books, New York).
Hempel, C.G., 1966, The Philosophy of Natural Science, (Prentice Hall, Englewood Cliffs, NJ).
Hoke, F., 1993, Technology Transfer Boom Offers Scientist Rewards-And Challenges, The Scientist, Vol. 7 (17) Universities Reassess Social Role as Tech Transfer Activity Mounts, The Scientist, 7, (18).
Kuhlmann, S., 1991, The Research-Industry and University-Industry Interface in Europe: Federal Republic of Germany, Research Memorandum and Frauhofer-Institut fur Systemtechnik und Innovationsforschung, Karlsruhe.
Kuhn, T.S., The Structure of Scientific Revolutions (University of Chicago Press, Chicago, IL).
Laffont. J.-L. and J. Tirole, 1993, A Theory of Incentives in Procurement and Regulation, (MIT Press, Cambridge, MA).
Lamb, D., and S.M. Easton, 1985, Multiple Discovery: The Pattern of Scientific Progress, Trowbridge, (Averbury Publishing, Trowbridge).
Latour, B., 1987, Science in Action: How to Follow Scientists and Engineers Through Society (Harvard University Press, Cambridge, MA).
Latour, B., and S. Woolgar, 1979, Laboratory Life, (Sage, Beverly Hills, CA).
Lind, R. (Editor), 1982, Discounting for Time and Risk in Energy Policy (Johns Hopkins University Press, Baltimore MD).
Little, I.M.D. and J.A. Mirrlees, 1974, Project Appraisal and Planning for Developing Countries, (Heinemann, London).
Luhmann, N. 1979, Trust and Power: Two Works (Wiley, New York).
Mansfield, E., 1991, Social Rate of Return from Academic Research, (University of Pennsylvania, Department of Economics Working Paper).
Marschak, J., 1971, The economics of information, in M. Intrilligator (Editor), Frontiers of Quantitative Economics (North-Holland, Amsterdam).
Meckling, W.H., 1962, Predictability of the Costs, Time and Success of Development, in: National Bureau of Economic Research, The Rate and Direction of Incentive Activity, (Princeton University Press, NJ).
Merton, R.K., 1957, Continuities in the Theory of Social Structure and Anomie, in: Social Theory and Social structure (Free Press, New York).
Merton, R.K., 1973, in: N.W. Storer (Editor), The Sociology of Science: Theoretical and Empirical Investigations (University of Chicago Press, Chicago, IL).
Milgrom, P.R., C. North and B.R. Weingast, 1990, The Role of Institutions in the Revival of Trade; the Law Merchant, Private Judges, and the Champagne Faires, Economics and Politics 2 (1) 1-23.
Moulton, P.J. Young and J.J. Eberhardt, 1990, Technological Advances in Instrumentation: Impacts on R&D Productivity, (Battelle Institute Working Paper, Seattle, WA).
Mowery, D., 1983, Economic Theory and Government Technology Policy, Policy Sciences 16.
Mulkay, M.J., 1972, The Social Process of Innovation: A Study in the Sociology of Science (Macmillan, London).
Mulkay, M.J., 1977, Sociology of the Scientific Research Community, in: I. Spiegel-Rbsing and D. de S. Price (Editors), Science,Technology and Society, (Sage, London, Chap. 2).
Musgrave, R.A., 1968, The Theory of Public Finance, 2nd edn. (McGraw-Hill, New York).
Musgrave, R.A. and A.T. Peacock, (Editors), 1958, Classics in the Theory of Public Finance (Macmilllan, London and New York).
Nagel, E., 1961, The Structure of Science, (Columbia University Press, New York).
National Academy of Sciences, 1989, Information Technology’ and the conduct of Research - The User’s View (Report of the Panel on Information Technology and the Conduct of Research), (National Academy Press, Washington, DC).
National Science Board, 1987, Science and engineering indicators - 1987, U.S.G.P.O. (NSB87-1), Washington, DC.
National Science Board, 1989, Science and engineering indicators - 1989, U.S.G.P.O. (NSB89-1), Washington, DC.
National Science Board, 1991, Science and Engineering Indicators - 1991, (Government Printing Office, NSB 91-1, Washington, DC).
Nelkin, D., 1987, Selling Science: How the Press Covers Science and Technology (W.H. Freeman, New York).
Nelson, R.R., 1959, The Simple Economics of Basic Scientific Research, Journal of Political Economy, 67.
Nelson, R., 1990, What is Public and What is Private About Technology? University of California Center for Research in Management, CCC Working Paper No. 90-9, Berkeley, CA, September.
Norris, K.P., 1971, The Accuracy of Project Cost and Duration Estimates in Industrial R&D, R&D Management 2 (1), pp. 25-36.
Pavitt, K., 1987, The Objectives of Technology Policy, Science and Public Policy 14, pp. 182-188.
Peters, L.S. and H.!. Fusfeld, 1983, Current U.S. University/Industry Research Connections, in: National Science Board, University-Industry Research Relationships, Se-
lected Studies (U.S. Government Printing Office, Washington, DC).
Pigou, A.C~, 1932, The Economics of Welfare (MacMillan, London).
Polanyi, M., 1951, The Logic of Liberty, (Routledge, London).
Polanyi, M., 1962, The Republic of Science: Its Political and Economic Theory, Minerva 1 (1), pp. 54-73.
Polanyi, M., 1966, The Tacit Dimension (Routledge & Kegan Paul, London).
Popper, K.R., 1968, The Logic of Scientific Discovery, 2nd ed. (Columbia University Press, New York).
Ravetz, J.R, 1971, Scientific Knowledge and its Social Problems (Clarendon Press, Oxford).
Romer, P.M., 1990, Endogenous Technological Change, Journal of Political Economy, 97, pp. 71-102.
Romer, P.M., 1993, Two Strategies for Economic Development: Using Ideas vs. Producing Ideas, in: Proceedings of the World Bank Annual Conference on Development Economics: 1992, L. summers and S. Shah (Editors, (World Bank Press, Washington, DC).
Rosenberg, N., 1990, Why Do Companies Do Basic Research With Their Own Money?, Research Policy, 19, 165-1740.
Rubinstein, A., 1979, Equilibrium in Supergames with Overtaking Criterion, Journal of Economic Theory 21, 1-9.
Rubinstein, A., 1980, Strong Perfect Equilibria in Supergames, International Journal of Game Theory 9.
Salomon, J.-J., 1973, Science and Politics, trans. Noel Lindsay, (Macmillan, London).
Samuelson, 1954, The Pure Theory of Public Expenditure, Review of Economics and Statistics 36 (November) 387-389.
Science, 1990, Data Sharing: A Declining Ethic? (New & Comment) 248, 25 May, 952-953.
Shannon, C.E. and W. Weaver, 1949, The Mechanical Theory of communication (Free Press, Urbana, IL).
Squire, L., and H. Van der Taak, 1975, Economic Analysis of Projects (John Hopkins University Press, Baltimore, MD).
Stankiewics, R., 1986, Academics and Entrepreneurs: Developing University-Industry Relations, (Frances Pinter, London).
Stiglitz, J.E., 1988, The Economics of the Public Sector (Norton, New York).
Storer, W., 1966, The Social System of Science, (Holt, Rinehart & Winston, New York).
Thompson, D., 1992, Science’s Big Shift, Time Magazine, 23 November, pp. 34ff.
Turner, S.P., 1990, Forms of patronage, in S.E. Cozzens and F. Gieryn (Editors), Theories of Science in Society (University of Indiana Press, Bloomington, IN), pp. 185—201.
Vincenti, W., 1990a, Engineering Knowledge, Type of Design, and Level of Hierarchy: Further Thoughts about What Engineers Know..., Paper for the Eindhoven Conference on the History of Technology, Eindhoven, The Netherlands, November.
Vincenti, W., 1990b, What Engieers Know and How they Know It: Analytical Studies from Aeronautical History (John Hopkins University Press, Baltimore, MD).
Von Hippel, E., 1990, The Sources of Innovation (Oxford University Press, New York).
Wade, N.. 1978, Guilleman and Schally, Science, 200, pp. 279-282, 411-415, 510-513.
Whitley, R., 1984, The Intellectual and Social Organization of the Sciences, (Clarendon Press, Oxford).
Ziman, J.M., 1968, Public Knowledge (Cambridge University Press, London).
Zuckerman, H.A. and R.K. Merton, 1971, Patterns of evaluation in science: Institutionalization, structure and functions of the referee system, Minerva, 9, 66-100.
Zuckerman, H.A. and R.K. Merton, 1972, Age, ageing and age structure in science, in M.W. Riley, M. Johnson and A. Foner (Editors), Ageing and Society, Vol. 3 (Russell Sage, New York).