"Organizational rights in knowledge-intensive firms"

Transcription

1 Ugo Pagano Dipartimento di Economia Politica, Università di Siena and CEU. "Organizational rights in knowledge-intensive firms" Very Early Draft. Paper written for the European Network on the Economics of the Firm (ENEF) 2008 meeting. Laboratory of Economics and Management, Scuola Superiore S. Anna Pisa. September, 11th-12th 2008.

2 1. Introduction. The literature considering the relative merits of capitalist firms and cooperatives has emphasized how capitalist firms may be advantageous in capital-intensive sectors especially when capital is specific and difficult to monitor. Labour-hiring-capital is considerably onerous in terms of agency costs when one has to borrow large amounts of money. Capital-hiring-labour is considered to be the obvious solution in these circumstances. One may argue that in a knowledge-intensive economy organizational rights should move from capital to labour. Knowledge is often embodied in human beings: for this reason it is more likely that, in such an economy, knowledge-intensive labour should hire capital. As far as knowledge disembodied from human beings is concerned, one should consider that knowledge is a non-rival good and its marginal cost should be negligible. A firm using disembodied knowledge is not using resources that cannot be simultaneously consumed by other firms. Firms, using intensively this type of capital, could be rather small and be owned by workers who hire at a very low cost non-rival knowledge capital. Knowledge-intensive firms should be characterised by free entry and by a regime of pronounced workers rights. However, while the labour-hiring solution can be found in some knowledge-intensive firms, this picture does not to seem to fit modern economies. If disembodied-knowledge-capital ends up hiring labour, other aspects of knowledge may be more important. Indeed, while non-rivalry makes knowledge close to the economic ideal of a public good, the possibility of excluding from its use may have opposite consequences. The acquisition of knowledge, even in the form of imitation can be very costly. Secrecy can prevent or limit access. Finally, intellectual property rights can grant some parties a monopoly of access, which can give cumulative advantages on the acquisition of additional property rights. Because of these ambiguities, knowledge-intensive firms can show much diversity and be much influenced by the prevailing legal and institutional frameworks.

3 In a first best world of zero-cost market institutions, the producers of knowledge should be rewarded by the community and the knowledge should be made available to everybody for free because the marginal cost of additional users is zero. In real life this solution is impossible because it is impossible to evaluate the benefits of innovations. To overcome secrecy and to favour the diffusion of knowledge, two main systems operate. One is based on prizes, reputation and career advancements for people disclosing interesting and useful knowledge (universities, open-software communities etc.). The other is based on the acquisition of private intellectual property. In a knowledge intensive economy the typologies of firms, which is going to be selected, is going to be influenced by the prevailing regime of intellectual property rights. In this paper we will argue that the present institutional framework of the world economy is biased in favour of private intellectual property regime and, as a consequences, firms typologies tend to be selected in an environment favouring solutions where non-human intellectual capital hires human intellectual capital. The paper is structured in three sections. The first compares the "capital-hires-labour" versus the "labour-hires-capital" solutions in the context of a knowledge-intensive economy. The second considers the ambiguous nature of knowledge and compares the origins and the relative advantages of the disclosure prize-driven system and private intellectual property and the possible criteria for finding a boundary between the two systems in a closed economy. We argue that, in the emerging global economy, the boundaries between the two systems are inefficiently biased in favour of private intellectual property regime. The third section argues that in the present institutional context of the world economy, the "capitalhires-labour" solution tends to prevail on the "labour-hires-capital" solution. This tends to generate a particular type of organizational equilibrium where large non-human capital intensive firms hold monopoly positions similar to those that the old corporations used to have for geographically limited areas in the age of colonialism. In this situation, the organizational rights in the corporation and its duty to behave take them into account can be a "third best" solution to the governance of knowledgeintensive firms.

4 In the final section, we conclude arguing that the "third best" solution to the governance of Knowledge intensive firm should go together with a "second best" attempt to reform the world economy. 2. Who hires whom in the knowledge-intensive economy? Neoclassical theory a-critically assumes a double neutrality: technology does not influence rights and rights do not influence technology. A famous dictum by Samuelson (1957, p. 894) states this dual neutrality of the neoclassical theory very clearly : In a competitive economy it really doesn t matter who hires whom. Samuelson thus asserts the irrelevance of firm ownership in so far as it does not matter whether it is the owner of the machines who hires the workers, or whether instead it is the workers who hire the machines. Neutrality thus holds in a twofold sense: whilst the various technologies and productive forces have no influence on the efficiency of the various kinds of ownership, the various kinds of ownership (for example control of the firm by workers or capitalists) have no influence on the nature of the productive forces and the type of technology used. The orthodox theory assumes a market with nil transaction costs and complete and perfectly enforceable contracts. Therefore, workers can always clarify their conditions of employment with maximum precision and not feel disadvantaged because they do not control the firm. A market characterized by zero transaction costs and complete contracts will protect agents whether or not they have property rights over the means of production. From this point of view, the characteristics of the productive forces are not important: a competitive equilibrium with complete contracts will in any case entail the efficient organization of production, both when the capitalist employs the workers, and when the contract provides for the workers to rent the means of production from the owner. The types of resources and technologies used will have no effect on the efficiency of the controlling actor. It will always be possible to stipulate contracts guaranteeing productive efficiency with other individuals in possession of physical and human capital. The nature of the resources employed in production does not tend to favour particular property rights, and accordingly are neutral.

5 In a world of complete contracts (in which other individuals are completely protected by those contracts), the firm s owners will not be able to alter the nature of the resources in their favour. They will find that they can only formally manage a production process, which in reality is exclusively organized by the constraints imposed by the enforcement of complete contracts. Upon realization that they are completely protected by their contracts, workers who have no rights to the means of production will not invest any less in human capital. For the same reason, the owners of the means of production will feel themselves equally protected if they rent their means of production to others rather than utilize them directly. In other words, the ownership form does not influence the types of resources and the nature of the technologies employed. The ownership structure, therefore, does not tend to favour particular productive forces, and consequently it too is neutral. Discarding the hypothesis of nil transaction costs has profound effects on the neoclassical edifice. It eliminates the twofold neutrality of rights with respect to technologies, and of technologies with respect to rights. The mechanisms identified by neo-institutionalists Oliver Williamson (1985) for example have cast serious doubt on the hypothesis that technologies are neutral in regard to the nature of property rights and types of control over firms. When it is impossible to write complete contracts, the characteristics of the productive forces influence the attribution of control rights. In the presence of contractual incompleteness, those in possession of relatively specific resources (i.e. resources which cannot be put to other uses without losing some of their value) find themselves in a very difficult situation. The fact that resources are specific does not constitute a problem in the neoclassical world of zero transaction costs and complete contracts. In this case, when a resource is specific (and therefore does not have uses alternative to its current one), it is always possible to protect oneself against opportunism by the counterparties with a complete contract (Hart 1995). The specificity of resources becomes a problem if it is not possible to obtain adequate guarantees by means of a sufficiently complete contract. In this case, those with control over the firm have greater guarantees than do other individuals. In these circumstances, because those who invest in specific resources are made vulnerable by the absence of alternative uses, they will seek to obtain such guarantees. Samuelson s proposition no longer holds, because in this situation who hires whom becomes important.

6 However, the reasoning can be reversed. Though it is true that those who possess resources specific to a firm seek to acquire control over it, the opposite is the case as well: in situations of contractual incompleteness, those with control rights over a firm have relatively fewer inhibitions about developing resources specific to that organization. 1 The same reasoning applies to information asymmetries. By virtue of the latter, some agents may possess hidden private information, which makes complete contracts impossible to stipulate. If some agents possess concealed information their monitoring becomes difficult, or even impossible; and this distances us considerably from a world of complete contracts. In this situation, because the technologies employed determine the distribution of information among agents, certain attributions of property rights tend to prevail because of their greater efficiency (relatively to the technologies employed). In particular, of greater efficiency will be the rights attribution that allocates the rights deriving from ownership to agents difficult to control (and/or controllable at very high costs) because they possess a greater amount of concealed private information. 2 Also this neo-institutionalist argument induces us to reject the neutrality of technologies: changes in the technological characteristics of the productive forces alter the nature of the information asymmetries and distribution of information among agents, thereby influencing the nature and attribution of property rights. However, the argument can be again reversed, and made to follow the reasoning of the radical economists. Those who own or control organizations will not perceive their accumulation of private information as problematic (because some of their characteristics and actions cannot be monitored by other agents). Indeed, they will endeavour to ensure that their concealed information is not used against them. But these actors will perceive an increase in the concealed information possessed by other agents as disadvantageous, for this inevitably increases the costs of monitoring them. The attribution of rights to a certain agent will therefore tend to shift information asymmetries in that agent s favour. Once again, the attribution of rights is by no means neutral, and it influences the nature of the productive forces employed. 1 Initially developed by Marx in the first book of Capital, this point has been taken up by numerous radical economists : for instance Braverman (1974), Marglin (1974) and Rowthorn (1974), Edwards (1979), Bowles and Gintis (1999). 2 This point was first made by Alchian and Demsetz (1972). See also Alchian (1987).

7 Summing up: while who-hires-whom does not matter in a world of zero transaction costs, it matters in a world of positive transaction costs. Even if technology is partially endogenous, one could argue that developments that go in the direction of making it convenient a high intensity of non-human capital should also make the capital-hiring-labour solution more appealing, especially when capital is specific and difficult-to-monitor (in the sense that its user induced depreciation cannot be easily assessed by observing the state of a machine and a relevant informational asymmetry exists between users and absentee owners of machines). In these conditions, the labourrenting-capital solution become prohibitively costly and also other solutions (such as labour borrowing capital and buy machines) involve very high agency costs. Perhaps a limited multiplicity of alternative technologies is available (and more could have been developed in the case of widespread labour-owned firms). However, one can argue that the intense use of physical capital has acted in favour of the capital-hiring labour solution. The central question whether the intense use of knowledge in production should imply a change in these tendencies. At the first glance, one could argue that unlike capital-intensive technologies, knowledge-intensive technologies should have two going in favour of the labour-hiring-capital solution: In the first place, knowledge is often embodied in human beings and in a knowledge-intensive economy one should expect for this reason that the labour-hiring-capital solutions should be often advantageous. In the second place, the knowledge that it is not embodied in human beings can be made available to additional members of society without depriving the former users of its avaibility. Unlike a piece of physical capital, an idea can be used simultaneously by many people without being warn out (indeed the opposite is true: the use of ideas helps the memory and the improvement of ideas). Since the marginal cost of using disembodied knowledge is zero, these firms should not face the renting and borrowing agency problems, which do usually upset the labour-hiring-physical-capital firms. From these two reasons, the advent of the knowledge economy should coincide with an increasing comparative advantage of labour-hiring-capital firms with respect to the capital-hiring-labour firms, and a radical departure from the organizational forms which have characterised the age of intensive use of physical capital.

8 However, while the trends of some sectors seem to confirm this prediction, on the whole modern capitalism does not seem to show a great deal of discontinuity with its recent past. Indeed a very early insight from Karl Marx seems to provide a better representation of the prevailing trends: It is a result of the division of labour in manufactures, that the labourer is brought face to face with the intellectual potencies of the material process of production, as the property of another, and as a ruling power. This separation begins in simple co-operation, where the capitalist represents to the single workman, the oneness and the will of the associated labour. It is developed in manufacture which cuts down the labourer into a detail labourer. It is completed in modern industry, which makes science a productive force distinct from labour and presses it into the service of capital. (Marx 1967, ch. 14, section 5) The monopolization of knowledge in the forms of IPR, which characterises many important sectors of the economy, seems to be more consistent with this early analysis of Karl Marx than with the potential discontinuities that we have considered. One could "save" the predicted discontinuities by arguing that the knowledge-intensive economy is not yet sufficiently intensive to produce widespread institutional changes. The other possibility is that the analysis of production of knowledge considered in this section is not catching important features of this complex good. This is the perspective that we are going to pursue in the following section of this paper. 3. Knowledge between private property and global common.

9 The hypothesis that the knowledge-intensive economy may involve a fundamental discontinuity in the organization of the economy relies on the fact that, unlike physical goods, knowledge is a public good in the sense that there is no cost in additional use by other consumers. However, pure public goods are a mix of two ingredients: non-rivalry in consumption (which characterizes knowledge) and impossibility of exclusion from consumption (which does not necessarily characterize knowledge). While pure private goods involve both the possibility of exclusion of others and non-rivalry in consumption, the two circumstances do not necessarily arise together. Indeed, goods characterised by rival consumption and the impossibility (or very high cost) of exclusion provide examples opposite to the case of knowledge. There is however a link between non-rivalry and costs of exclusion. In the case of well-defined physical objects the definition and the enforcement of exclusion can be done locally by verifying that others do not interfere with the rights defined over that particular object. The nature of the property of a computer, a car or a house is such that legal relations can be defined at local level. Only when individuals need to interfere with the local space occupied by the physical objects owned by other people, the respect of the property rights of others must limit their liberties. And, as long the objects that are not visibly taken away or altered by others, an owner can safely assume that his ownership rights are respected. The related legal positions have a local domain geographically limited by the position in space that, at a certain moment in time, is occupied by the material object over which the rights are defined. The material character of the good and its defined location imply a possible overcrowding by potential consumers and are a source of rivalry in consumption. When an individual uses the good, others cannot consume it. In this case, as long as individual keeps under control the good in a given physical location, he can be sure that the other individuals are not consuming it and are not violating its private property. Both the definition and enforcement of private property are specified at local level and they are unlikely to have any relevant implications in distant locations. The legal positions, defining the private ownership of knowledge, have wider implications and have often a global nature. They may involve restrictions for many individuals at various country locations and potentially for all the individual of the world. Intellectual property rights, such as they are currently defined by the TRIPS agreements and enforced by the WTO, have this nature. Their ownership by some individuals involves restrictions for all the other individuals. The ownership of a

10 house, a car or a field involves some duties for the surrounding individuals who should not interfere with the property rights of the owner and are, only for this reason, limited in the exercise of their liberty. By contrast, the ownership of a piece of intellectual property implies that all the individuals in the world have duty not to interfere with that legal position. They have to comply with the rights that it creates by limiting their actions in their daily life in multiple ways, irrespective of the country in which they operate. If some individuals happen to produce (or in a relevant cases they have already produced) 3 the same knowledge on which the right is granted, their liberty to use the results of their efforts is limited by the monopoly on knowledge that has been already acquired by others. The reinforcement and the extension of intellectual property have been compared to the enclosure of lands that preceded the industrial revolution. 4 Also in this case, commons were turned into exclusive private property. There is, however, a fundamental difference. In the case of land, the object of privatization was a local common that involved the legal positions of few individuals. By contrast, the privatization of intellectual property changes the legal positions of many individuals and has major implications for the international standings of the different countries. Here we have a paradox. Because of its non-rival nature, unlike land, knowledge can be used by many individuals without decreasing its value. However, the "public good nature" of knowledge makes its privatization much more limiting for the liberty of other individuals. Privatization turns the ownership of a piece of public knowledge into a private property right that involves duties for all the other individuals and has little to do with the traditional rights of exclusive consumption of the owners of material objects. To use Jefferson s vivid image, 5 knowledge is like the flame of a candle that can light many other candles without decreasing its own flame. The exclusive ownership of the flame can only mean that others are deprived of the liberty lightening their own flames. The rival nature of land implies that its private ownership restricts the liberty of non-owners only in the few cases in which it interferes with the (necessarily local) private uses a piece of land. The private appropriation of knowledge cannot imply that the liberty of the non-owners should be only limited when it interferes with the consumption 3 An account of cases in which traditional knowledge is stolen by multinationals is given by Shiva (2001). 4 For instance, see Shiva (2001: pp ). 5 He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me. Thomas Jefferson, Letter to Issac McPherson, No Patents on Ideas, 13 August Sometimes paraphrased as Knowledge is like a candle. Even as it lights a new candle, the strength of the original flame is not diminished.

11 of the owners: because of the public nature of knowledge, this never happens. Because of non-rivalry, one flame is never decreasing the intensity of another flame. The nature of ownership is here, necessarily, much more restrictive: it means that non-owners have no liberty to lighten their taper and use their own flame without the permission of the owner. This is more restrictive than simply nondecreasing the flame of the owner as the analogy with land would imply. The limitation of the costless liberty to use knowledge is inefficient. It is a well-known piece of economic theory that the non-rival nature of a good should not be the cause of an excessive restriction of liberty but rather a reason to grant to all the individuals the liberty to lighten their own flames. There is, however, also a well-known argument that can support this restriction: if the person that has borne the cost of lightening the first candle is not compensated for this effort, perhaps the overall flame of knowledge would be weaker. An appropriate incentive for the inventor requires that she becomes the owner of the knowledge that she has discovered and that the liberty of access of others is restricted. However, this restriction is always costly: after the first discovery, many other candles could have been lightened, in some cases also independently, without decreasing the flame of the first candle. The cost of depriving other candles of the flame increases when the knowledge is basic in the sense that it comes upstream in the production of other knowledge or it is complementary to other pieces of knowledge. For this reason, it is undesirable to subject more upstream or basic knowledge to the restrictions of private property. Since early times, public institutions like Universities have provided alternative systems of compensating the producers of open-access science. Publications, based on peer reviews, and careers and prizes that are based on these publications, are the most typical types of incentives offered by Universities to promote effort and universal disclosure of knowledge. Unsurprisingly, a great deal of the funding of these institutions has traditionally come from public sources. Where should one draw the line between more upstream knowledge produced and freely transmitted by Universities and the more downstream knowledge that can be privately owned by its discoverers? There is no precise answer to this question but, wherever the line lies, it will change when we move from a closed economy, ruled by one single state, to an open economy with many independent states. A World Government (or, in similar way, a State isolated from the World Economy) could try to draw the line between the production of open access knowledge (funded by tax revenue) and the

12 production of closed access knowledge (that is left to the profit-motive of private firms) in such a way to maximize the benefits accruing to its citizens. However, the real economy, which we face today, is different. No National State can be isolated from the World Economy and no World Governments exists. In this framework, each National State will realize that its citizens get only a fraction of the benefits of the investments in public knowledge while some of them (and all through national taxation) can gain the full benefit of the investments in privately owned knowledge because the benefits from the latter are not shared with the citizens of the other countries. Thus, in an integrated world economy, characterized by internationally enforced IPR, National States have an incentive to increase the number of closed access science research projects over which private property rights are defined and move upstream the line that separates them from the open access science research projects. The over-privatization of knowledge would simply follow even from the behaviour of benevolent world governments acting in the interests of their citizens in the present institutional framework of the world economy. It is well likely that the tendency to over-privatize is even more vigorous because national states are easy targets for the rent-seeking activities of global companies while international organizations are, in turn, easy targets for both national states and global companies. Institutions, producing and diffusing public knowledge, are increasingly seen as a waste of money and there is a widespread tendency to decrease their funding. For the same reason, the same institutions (Universities in the first place) are also under severe pressure to betray their nature of institutions mainly dedicated to the production and diffusion of public open-access knowledge and are pushed towards the production of private intellectual property. The (over)-upstreaming of private property rights on knowledge can easily create inefficient situations where, because of the monopolization of knowledge, an "anti-commons" tragedy occurs. Whereas in the Schumpeterian framework innovation created temporary monopolies that should be quickly exploited before imitation took place, here IPR creates long-term legal monopolies, which are sometimes not used for production but to block the innovations of others. In this way one may dry out the common pool of new basic knowledge which has been one of the factors which, since the industrial revolution, the epistemic basis of technological innovation. As Mokyr (2002) has outlined, the industrial revolution was made possible by a preceding industrial enlightment. Much "why" knowledge had been accumulated and made easily accessible. A large pool

13 of "why" knowledge made it possible the development of "how" knowledge in a form such that it was possible to move easily from one technology to another or from one "how" question to another. In situations where the epistemic basis of technology were underdeveloped, technological innovations were isolated answers to particular "how" questions and it was not possible to generate a continuous process of technological innovations. By contrast, once a rich basis of "why" knowledge was made available, it was possible to start a self-feeding interaction between why-knowledge and howknowledge. The growth of a common pool of basic knowledge is a necessary condition for the continuous growth of technical knowledge. A delicate balance between the global common basic knowledge and the proprietary technical knowledge must exist: the up-streaming of the proprietary arrangements to basic knowledge may imply the self-destruction of the conditions of its growth. While a great deal of the literature has concentrated on the arrangements and the incentives which stimulate the growth of technical knowledge, a more challenging puzzle is how to stimulate the growth of basic knowledge. It is true that intellectual curiosity may provide an intrinsic motive to the growth of knowledge but to share knowledge and find common methodologies to falsify wrong hypothesis requires very complex institutions. According to Paul David some development of the institutions of open science has predated the industrial revolution and has been a precondition for the development of capitalism. Pushing the frontiers of private appropriation capitalism may destroy the conditions that fuel its own development. In the words of Paul David (2004): Rather than emerging and surviving as robust epiphenomena of a new organum of intellectual inquiry, the institutions of open science are independent, and in some measure fortuitous, social and political constructs. They are in reality intricate cultural legacies of a long past epoch of European history, which through them continues to profoundly influence the systemic efficacy of the modern scientific research process. In this respect, the foundations of Universities marked an important change in the production of knowledge. As Berman (1985, ) has pointed out: What has been especially characteristic of Western Science, including legal science, since the twelfth century is its close historical connection with the institution of university; science was born in university and the university bestowed upon it its precarious heritage of freedom of teaching and research.. It takes, more than the progressive translation of the works of Aristotele to explain

14 why in the year 1150, possibly ten thousands students from all over Europe could be found in the town of Bologna in northern Italy studying legal science. They were there because society made it possible - indeed - made it urgent that they were there Interestingly enough, Universities did not only emerge well before the advent of capitalism but also well before the emergence of nation-states. They came about in a period when knowledge was conceived as global common good which should also been used in the government of the global institutions which characterized the Western Europe Middle Age. These global institutions had fierce contrasts but they recognized that each one of them had a proper sphere of influence. Good rules should specify the appropriate domain of their power. In other words the Middle Age was characterized by legal pluralism, that is a common legal order containing diverse legal systems (church vs. crown, crown vs. town, town vs. lord, lord vs. merchant). Legal Pluralism was a source of freedom and of legal sophistication and was a decisive factor in the foundation of Universities and the origin of Western Science. The typical questions, which emerged in this framework of overlapping legal systems were: Which Court had jurisdiction? Which law was applicable? How were different legal differences going to be reconciled? The independence of scholars became a precious asset to solve these disputes. Universities emerged as the locus both independence and fair adjudication of scholarly disputes should emerge. They were different from the classical academies where only one school of thought existed. Universities were based on the idea that there was some way to adjudicate truth that would emerge through proper debates and reference to the appropriate texts. For this reason, in the Western legal tradition, law was conceived to be a coherent whole within which all the disputes among the various authorities could be solved. In the formative era of the Western legal tradition, natural-law theory predominated. It was generally believed that human law derived ultimately from, and was ultimately to be tested by, reason and conscience. This theory had a basis in Christian Theology as well as Aristotelian philosophy. But it had also a basis in the history of the struggle between ecclesiastical and secular authorities, and the politics of pluralism. From this point view the legal doctrines developed in the first European Universities were much more than a simple rediscovery of Roman Law. Roman Law had not the same pretence and the same need to constitue a coherent whole within which the conflicts among authorities could be solved. Indeed, the phrase corpus juris Romani was not used by the Romans but by the twelfth - and thirteenth - century European canonists. It was the twelfth-century

15 scholastic technique of reconciling contradictions and deriving general concepts from rules and cases that first made it possible to coordinate and integrate the Roman Law of Justinian. The same methodology was then applied to other disciplines and paved the way to the emergence of Western Science. In some respects, the institutions under which Western Science emerged are polar to those of the contemporary economy. They came about in an era dominated by global political powers such as the Church and the Empire, which were supposed to represent the global interests of human-kind and accepted only limits due to their specific function. Local powers were weak, markets were not integrated, private property had still to see the emergence of the enclosure. The why-questions dominated the interests of people and offered a fertile ground to the growth of Universities and, in general, to emerging institutions of open science. The present world is characterised by an absence of global political powers and by integrated markets, which include knowledge. In this world, which open science has so dramatically helped to create, the institutions which make knowledge a global common are increasingly weaker than those that make it a private good. This does not mean that open science is not giving us amazing fruits but that its progress is slower than the one that could be achieved under more friendly institutions. 4. Biased selections and Organizational Rights. Much economic literature assumes that the institutions of open science, which an early stage of history gave as a present to capitalism, work for free. Typically, the Solow growth model assumes that the growth of technological knowledge is exogenous, costs nothing and is available to all. While the models of endogenous growth have usefully removed some of these assumptions, few contributions have explicitly dealt with the role of the institutions of open science. The typologies of firms, which tend to prevail in the knowledge economy will crucially depend on the role and the relative weights of the institutions of open science and the degree of private intellectual property rights protection. The absence of an adequate provision of public knowledge and

16 the over-up-streaming of private intellectual property is a likely result of the absence of adequate global institutions for funding public research and the rent-seeking activities of the firms which seek monopoly power on their products and the technology that they use. In this environment the selection of firms will be biased in favour of the capital-hiring-labour solution and it is very doubtful that the advent of the knowledge intensive economy will involve a substantial move to the labour-hiring-capital organizational form. At the end of the second section, we advanced the hypothesis that the intense use of non-rival disembodied knowledge capital would imply a substantial decrease of the agency costs in comparison to firms making an intensive use of physical capital. We can now qualify better the conditions under which this statement may be true: if much knowledge is produced using the institutions of open science, then the labour-hiring-capital solution is widely available. However, under the present global economic regime this tendency may be rather weak and may well be overcome by other more powerful factors. If agents can hold exclusive monopoly rights on knowledge, the use of the latter is going to be rather expensive and it is likely to increase the agency costs of labour-hiring-capital firms even in comparison to those which make an intensive use of physical capital. Moreover, under these conditions also another factor, favouring the labour-hiring-capital solution is going to be seriously inhibited. We have seen that the knowledge economy should also imply a growth of knowledge embodied in human labour and that in turn this is may increase the relative agency cost advantages of the labour-hiringcapital solution. However, when the knowledge-intensive economy turns out to be very intensive in terms of private IPR, then many individuals may find very little incentive to invest in embodied human capital. As the New Property Rights approach has shown (Hart, 1995), private property of the means of production has important incentive effects. A frictionless market for the means of production should imply that this property goes to the most capable individuals. However, the market is far from being frictionless and individuals are usually wealth-constrained. For this reason, causation may well work, in a self-reinforcing manner, also in the opposite direction: the owners of the means of production have a greater incentive to develop their capabilities and, for this reason, tend to become the best owners. This incentive effect of ownership is much stronger for intellectual property because the right to

17 exclude involves a restriction of the liberty of all the other individuals to replicate similar means of production (Pagano and Rossi, 2004). In the case of a machine, an individual, who has learnt to work and possibly to innovate with skills that are partially specific to the machine, is only partially damaged if he is deprived of its use. He keeps the liberty to work with other machines or to build identical machines. The damage is more relevant in the case in which an individual has acquired skills that are specific to a piece of intellectual property and he is denied the access to this piece. The nature of intellectual property implies that he does not keep the liberty to work with or to re-discover a similar piece of knowledge. IPRs involve a global right to limit the access of all individuals to the use of all the similar pieces of knowledge, including those that are independently developed. Turning a public good like knowledge into a private good transforms a universal unlimited liberty into an asymmetric legal position limiting non-owners freedom well beyond the restrictions that stem from the property rights defined on traditional rival goods. For some individuals the monopolistic ownership of intellectual property encourages the investment in the skills necessary to improve the knowledge that one owns and the skills that are developed make it even more convenient to acquire and produce more private knowledge. By contrast, other individuals may be trapped in vicious circle of (under-)investment in human capital where the lack of intellectual property discourages the acquisition of skills and the lack of skills discourages the acquisition of intellectual property. 5. Conclusion. In a knowledge-intensive economy the non-rival nature of disembodied knowledge and the increasing embodiment of knowledge in the human capital of the individuals should and could potentially imply a major shift from the capital-hiring-labour to labour-hiring-capital. However, this prediction assumes that much basic knowledge is produced as public good by institutions of open science. If much knowledge is privatised, the knowledge-intensive economy may turn out to be even more unfriendly than the physical intensive economy to stronger organizational rights of most individuals working in knowledge-intensive firms. Unions' protection and regulations of the employment relation are not outdated in modern knowledge-intensive economies.

18 References. Braverman H. (1974) Labour and Monopoly Capital, Monthly Review Press, New York Berman, H. J. (1985). Law and Revolution, Harvard: Harvard University Press. David P. (2004) Understanding the emergence of 'open science' institutions: functionalist economics in historical context. Industrial and Corporate Change V. 13, N. 4 pp Earle J, Pagano U., Lesi M. (2006) Information Technology, Organizational Form, and Transition to the Market. Journal of Economic Behavior and Organization. Vol. 60 pp Hart, Oliver D. (1995). Firms, Contracts and Financial Structure. Oxford: Clarendon Press. Marglin S. (1974) "What Do Bosses Do?" Review of Radical Political Economy vol. 6 pp Marx K. (1967) Capital. Vol. I. International Publishers. New York. Nokyr J. (2002) The Gifts of Athena. Historical Origins of the Knowledge Economy. Princeton University Press. Princeton and Oxford. Pagano U. (2007) Cultural Globalization, Institutional Diversity and the Unequal Accumulation of Intellectual Capital. Cambridge Journal of Economics, pp Pagano U. (2007) Karl Marx after New Institutionalism. Evolutionary and Institutional Economic Review. pp Pagano U., Rossi M. A. (2004) Incomplete Contracts, Intellectual Property and Institutional Complementarities'European Journal of Law and Economics V. 18 n 1 pp Pagano U., Rowthorn R.(1994) Ownership, Technology and Institutional Stability. Structural Change and Economic Dynamics. Vol.5 No 2 pp Rowthorn R.(1974) Neo-classicism, Neo-Ricardianism and Marxism. New Left Review vol. 86 pp Samuelson P. (1957) "Wage and Interest: A Modern Dissection of Marxian Economic Models", American Economic Review No 47 pp

19 Shiva, Vandana (2001), Protect or Plunder? Understanding Intellectual Property Rights. London and New York: Zed Books. Williamson O. E. (1985) The Economic Institutions of Capitalism. The Free Press, New York.