AN ECONOMIC ANALYSIS OF THE LAW GOVERNING PATENT SCOPE IN THE U.S. AND IN THE U.K. The Doctrine of Equivalents versus the Catnic/Improver Test

Transcription

1 AN ECONOMIC ANALYSIS OF THE LAW GOVERNING PATENT SCOPE IN THE U.S. AND IN THE U.K. The Doctrine of Equivalents versus the Catnic/Improver Test Master thesis ANNA ZAVAGNIN 2003 Erasmus Programme in Law & Economics Supervised by Prof. Andrew Griffiths 1

2 TABLE OF CONTENTS ABSTRACT... 4 AUTHORSHIP DECLARATION... 5 COPYRIGHT... 5 INTRODUCTION CLASSICAL ECONOMICS ON PATENT SCOPE The meanings of patent scope The allocation of the rights Equivalent inventions Patent justifications Economic rationale Philosophical perspectives Impact on scope PATENT DISTORTIONS IN THE MARKET The monopolistic effects The Tragedy of the Anticommons Additional costs THE DOCTRINE OF EQUIVALENTS Main principles and functions Evolution Winans v. Denmead Graver Tank v. Linde Air Prods. Co Warner Jenkinson v. Hilton Davies Economic foundations

3 3.4 The role of the prosecution history estoppel Recent modifications: the Festo case The Federal Circuit precedent The reverse decision of the Supreme Court Future perspectives and developments THE CATNIC / IMPROVER TEST Interpretation of claims Content and function The Catnic case The three-part test or Improver questions Consequences of the Epilady case The relationship with Article 69 of the European Patent Convention and the Protocol on its interpretation Role of the Protocol Interpretation effects on the Catnic/Improver Test Divergent comments AN ECONOMIC ANALYSIS OF EQUIVALENTS: A COMPARISON BETWEEN THE US AND THE UK Similarities and differences Tendencies toward harmonization Economic evaluation Cost Benefit analysis Empirical data Final implications on scope CONCLUSIONS REFERENCES Graph 1 (a b)..20 Flowchart Table

4 ABSTRACT This thesis deals with the optimal patent scope, which represents a fundamental key for balancing the trade off between the desirable and undesirable effects generated by patents. On one hand they alter competitive forces in the market by imposing a monopoly with a result of social inefficiency, on the other hand they promote innovation and R&D activities. The dilemma between a broad and a narrow scope is addressed by analysing infringements rules in the case of equivalents. In particular, the U.S. Doctrine of Equivalents and the English Catnic/Improver Test, as modified by the Protocol on the interpretation of Article 69 of the European Patent Convention, are compared from an economic and legal perspective. Both patent systems attempt to outweigh the interests of inventors and of society, although the Doctrine of Equivalents is quite a flexible theory, which is in favour of a broader scope and thus of a patentee s strong monopolistic position with the opportunity of some corrective measures to apply case-by-case. On the contrary, the English system entitles a quite narrow scope according to its tradition, partially extended by the EPC. Although concern is expressed with respect to an inadequate and too wide scope, the breadth granted by the U.S. Doctrine of Equivalents is more efficient than the Catnic/Improver Test, since the U.S. theory is characterised by high flexibility and a case by case approach, paying particularly attention to giving incentives to technological advance. The empirical data, the different solutions provided in some technological areas and some tendencies towards an international harmonization show the consistency and the equilibrium achieved by this theory. Consequently a quite broad scope, characterised by a flexible system that enables incentives to be outweighed in every case, may represent the socially desirable patent breadth. However, the American system ought obviously to be improved, especially on the grounds of legal certainty and predictability, while the English and European ones should accelerate the process of convergence towards the American model without losing the extraordinary experience matured in centuries of tradition. 4

5 AUTHORSHIP DECLARATION I hereby declare and confirm that this thesis is entirely the result of my own work except where otherwise indicated. I acknowledge Prof. Andrew Griffiths for his supervision, availability and friendly guidance during my work on this thesis. Manchester, 1 st August 2003 (Anna Zavagnin) COPYRIGHT Copyright in text of this thesis rests with the Author. Copies (by any process) either in full, or of extract, may be made only in accordance with instructions given by the Author and lodged in the John Rylands University Library of Manchester. Details may be obtained from the Librarian. This page must form part of any such copies made. Further copies (by any process) of copies made in accordance with such instructions may not be made without the permission (in writing) of the Author. The ownership of any intellectual property rights which may be described in this thesis is vested in the University of Manchester, subject to any prior agreement to the contrary, and may not be made available for use by third parties without the written permission of the University, which will prescribe the terms and conditions of any such agreement. Further information of the conditions under which disclosures and exploitation may take place is available from the Head of the Department of Social Sciences and Law. 5

6 INTRODUCTION Patents, which grant the inventor of a new product or process the exclusive rights to use, make and sell it for a certain period of time, establish a monopoly in his favour and thus create both desirable and undesirable effects. On one hand, benefits may be accrued by increasing the level of innovations through monopolistic reward, otherwise there may be an insufficient numbers of inventions. On the other hand, some disadvantages exist in terms of deadweight loss and high prices due to lack of substitutes. The delicate trade off between gains and losses produced by patent systems has been deeply analysed in all its features, especially with regard to the socially desirable scope. The importance of patent scope becomes increasingly clear when infringement cases and rules are addressed: legal principles may represent a key for balancing opposing incentives, particularly concerning equivalents. Indeed, equivalents raise several problems in terms of claims drafting and interpretation and insist on the outweigh activity that ought to be undertaken in every decision. Above all they underline the fundamental role played by economic arguments in providing legal guidelines. This study, divided into five chapters, directly addresses the comparison between the Doctrine of Equivalents and the Catnic/Improver Test from a legal and economic perspective. Hardly any economic and legal literature has been developed in the field of infringement by equivalency, especially with respect to these theories. According to Chisum s opinion 1, this might be due to the complexity of the topic, to the different factual applications and technologies involved, to the trade off between the patentees and competitors activities and finally to the effects in terms of distortions on competition. The first chapter provides an analysis of patent scope by highlighting its meaning and its effects, especially regarding equivalents, and explaining the main advantages of patent systems, as conceptualised from a theoretical point of view. The second chapter mentions the 1 See Chisum (2000, 940). 6

7 negative effects of patents in the competitive market: it specifies the losses due to the monopoly, the Tragedy of the Anticommons and the race to invent. The third chapter faces the Doctrine of Equivalents in the U.S. patent system by illustrating its evolution until the recent and debatable Festo decision. The economic rationale and the temperaments offered by the doctrine of prosecution history estoppel and other remedies, such as the reverse doctrine of equivalents, the all element rule and the foreseeable bar, are primarily referred to. The fourth chapter describes the fundamental features of the Catnic/Improver test and its problematic relationship with article 69 of the Protocol of the European Patent Convention. The final evolution and the strong debate concerning this harmonization are underlined, obviously with particular attention to the economic perspective. The last chapter concentrates on the comparison between the American and English systems and elaborates a normative analysis in order to establish the more efficient rule between the Doctrine of Equivalents and the Catnic/Improver Test. Again, some implications in terms of socially desirable patent scope are provided. 7

8 1 CLASSICAL ECONOMICS ON PATENT SCOPE 1.1 The meanings of patent scope This thesis deals with the optimal scope of patents, the granting of which entitles the inventor to the right to exclude others from using and selling an invention and consequently imposes a monopoly, which alters competitive forces in the market with a result of social inefficiency. These monopolistic powers enable the patentee to protect himself and his invention by suing those actors who infringe directly or indirectly his patent. For this reason legal rules that regulate infringement represent a key point for patent economic rationale, as they affect directly its scope. Not only do infringement rules influence and determine patent scope, but also exercise an indirect effect upon the systems of drafting patent application. Although this relationship may not appear evident, a deep analysis of the United States and the English systems will make it clear 2. The patent right has two main dimensions: the length, that is the number of years between a patent s registration and its expiration, and the breadth, which is the degree of similarity between two inventions allowed by the system without infringing the protected patent. The economic value of a patent, as commonly known, is strictly connected with its scope, that defines the extent to which its holder can exercise his rights and allows him to establish whether or not such right has been infringed, especially in the case of equivalent inventions. In general, the broader the patent scope, the stronger is the patentee s monopolistic power and the wider the market in which he can exercise it. A broad patent also increases the number of eventual infringements and the incentives for creating new ideas. On the contrary, it produces a weaker stimulation for disseminating information and applying new discoveries. This implies the existence of a delicate trade-off between these two opposing incentives, the balance of which should be sought. In fact, a double-sided problem arises in terms of incentives for innovation: on one hand a broad patent implies strong protection of pioneer discoveries and promotes innovation and technological research, yet on the other hand it may 2 See the 3 rd and 4 th chapters. 8

9 negatively influence the development of applications or improvements. Broadly speaking there are different approaches in case of broad or narrow patents: the first ones encourage fast and duplicative research, while the second ones promote slower and complementary development. 3 The scope is derived from the patent application 4, which is made up of claims defining what the inventor considers to be the invention, a specification that is a full description of it, and eventual drawings. 5 Despite this, the scope is mostly characterised by the claims contained and allowed in the patent application. In fact, they determine the degree of control and the power granted to the invention, as the patent breadth enhances the probability that other discoveries in quite close sectors might have of infringing the patent. The Patent Office and the Courts usually make decisions on patent scope: the first with respect to the claims allowed within a specific patent, the second during litigation. Several problems may emerge throughout the interpretation of claims, as discretion often delineates the large number of legal rules and principles. For instance, United States Courts refused to consider a patent claim [..] like a nose of wax, which may be turned and twisted in any direction, by merely referring to the specification, so as to make it include something more than, or something different from, what its words express [..], as stated in White v. Dunbar The allocation of the rights Although patent scope may be analysed in terms of rights allocation according to the Coase Theorem 7, a deeper analysis will immediately reveal the impracticality of this perspective. In fact, in a Coasian world with no or negligible transaction costs parties will bargain to achieve a Pareto superior solution and the initial entitlements will be absolutely irrelevant. Every allocation of rights will spontaneously result in allocative efficiency, without 3 When two inventions are similar but not identical, a broad rule entitles the first inventor of both discoveries, while a narrow one entitles every singular inventor of his discovery. 4 On the basis of the 35 U.S.C. 33, an applicant [..] shall particularly point out and distinctly claim the part, improvement, or combination which he claims as his invention or discovery [..]. 5 See 35 U.S.C. 33 and 112 and U.K. Patent Act (1977) 14. According to both, the disclosure element ought to be sufficient to allow a person skilled in the art to make, perform and use the invention. 6 See White v. Dunbar, 119 U.S. 47, 51, 7 S.Ct. 72, 74, 30 L.Ed See Coase (1960), Demsetz (1967), Cooter and Ulen (2000, 82), Ko (1992, 782) and Gutterman (1997, 64). 9

10 consequences regarding overall social benefits 8. According to this theory, patent scope will have no effect upon technological research and progress within a competitive market. Unfortunately these assumptions are completely unrealistic for this study because the real world is characterized by high transaction and administrative costs 9, especially in the case of equivalent inventions 10, so efficiency will not be reached. In fact, I believe that a negotiation between technological pioneer inventors and improvers is extremely improbable because it involves the agreement on the values of both the pioneer invention and its enhancement, coupled with an appropriate division of the exploitation rights. According to some commentators 11, the failure of the Coase Theorem can be corrected by means of two legal remedies: on one hand, bargaining among inventors can be increased 12, for example in the case of joint ventures, though antitrust law may represent a strong obstacle. On the other hand, patent rights may be allocated to the party who values them more, as designed in the Normative Hobbes Theorem 13. Hence the scope ought to be broad in the case of pioneer inventions, which have low commercial value alone, while narrow in the opposite case. This solution is not completely persuasive so it will be argued in detail in the last chapter. 14 However the Coase Theorem remains absolutely fundamental: while it stresses the need of patent systems in innovating and promoting progress and scientific investment, as the market mechanisms will fail if left alone, it simultaneously reveals the decisive function performed by patent scope to achieve economic efficiency and a proper technological output. 8 Individual gains of parties will only be affected by different initial entitlements because the final benefits will remain the same as a result of the bargains. 9 In particular, high informational costs due to asymmetric information and strategic behaviour will complicate the bargains by preventing an efficient solution, while licensing will cause enormously high transaction costs. Moreover it should be recognized that allocation of property rights between technological pioneers and improvers is not a zero-sum game. 10 See Cooter and Ulen (2000, 129). 11 See Cooter and Ulen (2000, 132) and Merges and Nelson (1990). 12 This remedy is also known as Normative Coase theorem. See supra Cooter and Ulen (2000, 132). 13 Ibidem. 14 See the 5 th chapter. 10

11 1.3 Equivalent inventions The theoretical analysis on patent scope becomes more and more complicated when it is applied to equivalent inventions. They can be defined as those discoveries, which are essentially the same, in terms of function and result, as other ones that have been already patented. 15 Equivalent inventions impose the basic distinction between pioneer inventions and their improvements or applications. In fact, every pioneer invention, which was formulated in a famous U.S. sentence 16 as [..] a patent covering a function never before performed, a wholly novel device, or one of such novelty and importance as to make a distinct step in the progress of the art [..] enables a broad range of equivalents. The decision whether or not to issue a broad patent may strongly influence its economic effects by imposing excessively high costs compared to the benefits accrued. However the dilemma between a broad and narrow patent cannot be simplified in this way because it involves too many factors whose trade off appears quite problematic. The difference between broad and narrow patents is mainly underlined in the field of R&D. Some pioneer inventions may not have an immediate commercial value but a high potential of developing in many applications and improvements. The criteria of efficiency are apparently in favour of broad patents when the social value of investment on fundamental research exceeds the social value of investment on developing applications, while narrow patents should be preferred in the opposite situation. 17 On the contrary, a study 18 demonstrated that fundamental research might be discouraged if pioneer inventors are not entitled to receive anything from the sale of commercial applications, especially if their value is enormously superior to the basic discovery. Therefore the dilemma reappears as soon as commercial applications and pioneer inventions are considered joint products of basic research. Many economists have studied the legal boundaries of the patent grant to figure out the socially desirable patent breadth and to draft efficient legal rules. My economic analysis will start by summarising the main contributions made by economic literature to clarify the 15 See infra 2 nd and 3 rd chapters for more details with regard to the USA and UK systems. In particular, see Graver Tank & Mfg co. v. Linde air Prods. Co. in 339 U.S. 605, 608 (1950). 16 See Westinghouse v. Boyden Power Brake Co., 170 U.S. 537, (1898). 17 See Cooter and Ulen (2000, 131). 18 See Chang (1991). 11

12 dominant perspectives and provide a basis on which develop my own arguments. Some studies 19 have focused on the static trade off between the incentives provided to potential inventors by the patent system and the deadweight losses in terms of social costs due to the monopoly granted and the under use of new discoveries 20. This fundamental point of view has been debated by many commentators 21, who have underlined the important role played by the eventual substitutes of patented inventions. In fact, Gilbert and Shapiro 22, together with Klemperer 23, recognize that broadening the patent scope may reduce potential substitutes and consequently increase social costs. Conversely, Merges and Nelson partly disagree with this view in one of their works of research 24, in which they have stressed the economic meaning of patent scope and its effects on R&D from a dynamic point of view. In fact, substitutes might include not only subsequent discoveries, which can surrogate the initial invention, but also improvements and enhancements that represent social gains. According to these economists, patent scope can be used as a means for balancing incentives in investments and benefits gained but a deep analysis ought to take into account additional elements, such as different actors involved, levels of technology, specific sectors of industry, the efficiency in the long run and costs derived from litigation and uncertainty. In addition, conditions for improving pioneer inventions should be preferred and fostered, even at the expense of preserving the domination of pioneer firms 25. They concluded that [..] without extensively reducing the pioneer s incentives, the law should attempt at the margin to favour a competitive environment for improvements, rather than an environment dominated by the pioneer firm [..] 26 because the benefits gained by the pioneer firm may be balanced by the loss of competition in the technological market. Moreover they pointed out that there is a proportional relationship between the speed of innovations and the number of inventions: an 19 See Nordhaus (1969, 70). 20 The value of a patent right is a function of the costs associated with establishing, maintaining and enforcing its boundaries. Uncertainty due to unclear boundaries might obviously increase in a relevant way these costs. 21 See Gilbert and Shapiro (1989), Klemperer (1988) and Merges and Nelson (1990). 22 Ibidem. 23 Ibidem. 24 See Merges and Nelson (1990, 868ss). 25 This is also consistent with goals of antitrust policies. 26 See Merges and Nelson (1990, 843-4). 12

13 increase in R&D activities will lead to more inventions, which will promote positive effects on the future growth of economic productivity. Therefore their analysis is quite comprehensive, as it stresses the post-invention process for development and subsequent improvements. In particular, incentives to invent ought to embody post-invention conditions favourable to the inventor, while the social costs of a patent should include the potential risks of reducing competition in the market of improvements within a specific technology. My analysis is based on the same assumptions, as I agree with them that the trade off between benefits and deadweight losses is not sufficient to determine the efficient scope of patents and it should be completed by considering additional factors. 1.4 Patent justifications The desirable effects produced by patents consist mostly of encouraging inventive activities, disclosing information and increasing the level of innovations and R&D. Although there are some negative effects too 27, patent systems and their scope have been the object of a wide theoretical debate, which was particularly strong in the second half of the 21 st century. It has been developed by providing several legal, economic and philosophical justifications in order to balance their anticompetitive effects Economic rationale One of the most important sets of economic arguments in favour of patents are called consequentialist theories 28 and attempt to justify intellectual property rights on the basis of the good consequences flowing from their recognition, which can be generally identified by the advancement in knowledge, industrial progress and economic efficiency. There are two main branches of it, the utilitarianism and the teleology, which both concentrate upon the incentives provided by the law but differ in the way these good consequences are identified. 27 See the 2 nd chapter. 28 See Machlup and Penrose (1950,17ss) and Gutterman (1997, 8). 13

14 While utilitarianism, also recognized in the modern theories of Law & Economics 29, measures the good in terms of satisfaction of human preferences, without any judgement on their content, teleological theories focus on the evaluation of what should be identified as good in the prospective of the proper purpose of human life 30. Nowadays utilitarianism represents the dominant perspective: intellectual property rights in general deserve special protection if the benefits to the society exceed their costs. 31 This view has been developed in the last century in different and specific theories, which coincide with the main advantages of patents: the incentive to invent, the incentive to disclose, the incentive to innovate and the prospect theory. Firstly, patents are necessary to give incentives for inventions and to encourage the socially desirable level of investment in R&D, because without them the society will under - invest in inventive activity as a result of the free rider behaviour of other competitors, whose imitation will prevent a proper rate of return for the inventors. 32 In particular, patents provide inventors with some assurance with regard to rewarding the social value of the invention and recovering the development costs, especially the high risks involved in research 33. In fact, a rational inventor will invest in research until the point where the expected marginal return from more research is equal to the marginal costs. Consequently patents allow the inventor to internalise the externality and to obtain benefits derived from the new discovery by imposing costs on potential imitators. These advantages are basically pointed out in the incentive to invent theory, also known as reward theory, which emphasize some debatable elements too. 34 Moreover society would suffer from reduction in R&D 29 See Landes and Posner (1989) and Gutterman (1997, 8). 30 In particular, they emphasize the intrinsic value of knowledge and aesthetic experience. 31 The utilitarian perspective can be seen as an expression of the cost-benefit analysis within the Law & Economics movement. 32 In fact, inventions consist of new information and thus are characterized by the public good character. Therefore the private rate of return to R&D is estimated to be less than the social one. Although imitations lower prices, it will also lead to reduce number of inventions. See Gutterman (1997, 38), Einsenberg (1989), Ko (1992, 791), Machlup and Penrose (1950, 17) and Carlton and Perloff (2000, 505). 33 The idea of patents as a means for promoting progress is pointed out in the Constitution of the United States as the Congress in entitled to the power [..] to promote the progress of science and useful arts, by securing for a limited times to authors and inventors the exclusive right to their respective writings and recoveries [..]. 34 In particular, empirical data on R&D activities in a normal competitive market showed that patents are determinant in selected sectors (for instance, in the pharmaceutical industry), for smaller firms and new market entrants, while they may not be especially significant in other industries (electrical engineering). In fact, other barriers to competition may perform the same function of patents: difficulty of acquiring technical knowledge, 14

15 activity because new inventions may lower production costs of existing devices or increase productivity. Secondly, patents foster and accelerate inventive activity by means of disclosure, otherwise there would be no public access to valuable and new information 35. Furthermore no patent systems would create high social costs, waste due to duplicative researches and losses as a consequence of reduction in wider distribution, alternative uses and price competition in the market 36. This positive effect refers to the incentive to disclose theory 37, which has often been concretely applied in judgements, even if many scholars are quite sceptic. 38 However, various sets of empirical data have demonstrated that the utility of disclosure clearly depends on the sector considered, as the predictability of discoveries is higher for mechanic industries than chemical ones, and on other factors such as national policies, the access system and the technological state of countries. Thirdly, patents encourage the investment in innovation by emphasizing post inventive activities and considering all of the steps required for the commercial exploitation of the new creation during the patent monopoly. 39 This patent advantage is expressed in the incentive to innovate theory 40, which was firstly developed by Schumpeter 41, whose distinction between invention, which produces [..] no economically relevant effect at all [..] 42 and innovation, which represent the key for economic modifications within a [..] process of production facilities, managerial experience and finally marketing strategies. See the research of the Department of applied economics at the University of Cambridge, as contained in Gutterman (1997, 43). 35 This requirement is satisfied when the specification in a patent application is fulfilled in a sufficiently complete and clear way that a person skilled in art could make or use the invention. 36 Disclosure may have positive consequences with respect to the investments in improvements and enhancements of existing inventions instead of duplicative researches and to the technical know how often connected to an invention. See Gutterman (1997, 47), Ko (1992, 795) and Machlup and Penrose (1950, 25). 37 This theory is also called the exchange for secrets or enablement theory, according to which the inventor is not entitled to receive a privilege from the State but he bargains with the society a kind of Rousseau s social contract. 38 Some commentators argue that inventions would have been disclosed in any case, even without patents, or that the information contained in patent applications is not sufficient to understand the discovery and to diffuse a proper level of knowledge. Besides, the introduction of patents should be required only in case of impracticability of other means, like inadequate trade secrets, easily discovery and diffusion of technology and high degree of misappropriation. See Gutterman (1997, 47). 39 This theory differs from the previous ones since it underlines that patentees incur costs for the investments in R&D before the patent grant but some additional expenses during the patent term as well (distribution of products, refinements, marketing and promotional activities). 40 It is named also the monopoly profit incentive theory. 41 See Schumpeter (1950, 81), Ko (1992, 799), Scherer (1984, 53) and Gutterman (1997, 52). 42 See Schumpeter (1950, 83-84). 15

16 creative destruction [..] 43, is famous but criticized at the same time. 44 According to this author, competition based on new commodities and technologies is the engine for promoting and fostering scientific progress. This implies that a patent monopoly may encourage and achieve more innovation and economic growth than a perfect competitive market within a dynamic model. 45 In conclusion, the grant of patents with a broad scope may perform a prospect function in terms of affording investments in further research, avoiding duplicative researches and coordinating his activity with other firms during the development of the discovery. Moreover, patents may simplify and stimulate bargains by eliminating the risks involved in the negotiations between the patentee and society. These benefits are emphasized in the prospect theory, developed by Edmund Kitch 46, according to which subsequent improvements, enhancements and refinements should be included in the patent scope, even more than the standard applied by United States Courts. This theory, which is quite consistent with the analysis of the allocation of property rights developed by Posner and Demsetz within the Law & Economics research 47, plays a decisive role in emphasizing the importance of proper incentives after the creation of a new invention but nevertheless some economists have shown scepticism from different points of view Philosophical perspectives Some fundamental arguments in favour of patents deal with philosophy and are commonly known as the natural law or the moral rights theory: the inventor is rewarded 43 Ibidem. 44 The essential distinction between invention and innovation is traditionally discussed, as it can be problematic in the case of a continuing process of innovation and also useless regarding routine improvements. Furthermore, Schumpeter s assumptions have been challenged on the ground that empirical data are open and not persuasive in demonstrating that patents provide enough incentives and innovation. See Kingston, as quoted in Gutterman (1997, 56), who refers his enquiries to the chemical and pharmaceutical area, in which high investments can not be made by individual inventors. 45 In fact, this monopolistic protection provides firms with time, space and resources for further researches. 46 See Kitch (1977, 256), Ko (1992, 799) and Gutterman (1997, 57). 47 See Demsetz (1967). 48 In particular, high rivalry between future patent holders will increase the risk of a race to invent, high transaction costs will prevent the achievement of an efficient licensing system and finally unpredictability may affect negatively the underlying technology, specifically in the pharmaceutical area. See Kitch (1977, 256) and Ko (1992, 800). 16

17 for the idea incorporated in his invention because it is an expression of his natural rights, belongs to him and could otherwise be stolen. Two different applications of this theory can be determined: the first, the Lockean labour theory 49 considers the inventor s property right as the reward for the difficulties and the efforts due to the labour, while the second, the Hegelian personality theory, connects the patent right to an expression of the human personality. Unfortunately both theories present some weaknesses in terms of policy implications and economic incentives; for instance, the Lockean theory does not clarify whether the reward refers to the past work of the inventor or to save future researches of others. Moreover, arguments in terms of fairness, justice and morality can be also read in the Constitution of the United States 50, even if the natural law position is not completely shared. These theories are obviously important from a historical point of view, though quite useless in terms of patent scope and social policies. Hence, my analysis will concentrate more on economic arguments Impact on scope The desirable effects produced by patents, wholly described in the previous theories, provide some implications concerning the appropriate scope. All of them are in favour of a broad scope, even though no concrete guidelines are suggested. The incentive to invent theory insists on the need of granting patent protection for inventions characterized by high costs in R&D and it approves of a wide scope, as judged necessary for recovering initial investments. Likewise, both the disclosure requirement and the incentive to innovate theory prefer a broad scope 51 since it can facilitate the process of divulgation and acceleration of innovation; the unique exception is represented by the case 49 This theory represents the expression of the Anglo-Saxon jurisprudence on intellectual property rights, as underlined in Braga (1990, 17). 50 See supra at footnote The scope should concern not only the information disclosed in the specification and in the claims, but also the eventual applications, in order to avoid free rider behaviours. See Merges and Nelson (1990, 844), Ko (1992, 800), Cohen (2001, 30), Einsemberg (1989). 17

18 of biotechnological innovations 52. In general, the prospect theory opts for an extremely broad scope by emphasizing the steps following the invention and required for its full exploitation and commercialisation. Although these perspectives can be easily found persuasive, it should be recognized that broad patents are dangerous because of their potential adverse effects upon R&D activities: they might inhibit the discovery of fundamental improvements and they might discourage research in some areas. The final result depends primarily on the nature of the technology, the sector involved and the influence of other factors. Consequently it remains an open question whether scope ought to be broad: this problem will be solved at the end of this analysis as a result of the legal and economic comparison between the American and English patent systems. 52 See Cohen (2001, 31), who stresses the importance of a narrower scope by protecting the product or process claimed and its predictable variations. 18

19 2 PATENT DISTORTIONS IN THE MARKET Criticism and scepticism on the maintenance and the excessive broadness of patent are still common among many scholars who believe that the costs of protection are too high and exceed the benefits obtained in exchange. In fact, patents produce not only positive but also negative effects. The disadvantages of patent systems refer not only to the increase of prices due to the monopoly, but also embody different kinds of distortions: the deadweight losses derived from the Tragedy of the Anticommons 53, the waste of resources caused by the race to invent, the high transaction and administrative costs generated by the definition of claims drafting. These costs will become particularly relevant within the economic analysis in order to make an optimal balance between losses and gains. 2.1 The monopolistic effects From a general point of view patents create monopolies in favour of the patentees and potentially distort competitive markets. In fact, the patent holder may increase prices as a consequence of the monopolistic power and also limit the use of the invention. Specifically these higher prices may be an expression not only of the mere costs incurred by the inventor for the developing activity but also a reflection of the value of the invention and the effort contributed by the patentee to achieve such an invention. This implies that a deadweight loss in the consumer surplus will be produced and it will be coupled with a reduction in terms of social welfare. In details, a monopoly can be defined as a market structure characterised by one only supplier of a product for which there is no close substitute. Consequently the monopolist faces a downward sloping demand curve and sets the price above the marginal cost. As a result society will suffer a deadweight loss because it is sold a lower quantity for a higher price compared to the quantities and prices sold in a competitive market. In the case of patents, the 53 See infra at

20 patentee usually reduces the output and increases the price above the marginal cost, so there will be a social deadweight loss represented by the triangle DWL in the graph 1a. This area refers to the social loss due to monopoly compared to a perfectly competitive market. Costs derived from patents do not only consist of the social losses but also cover a part of the patentee s benefits, as underlined by Posner and other economists 54. p (1 a) MR Monopoly profit D DWL MC p Q m Q r Q c Q (1 b) Revenue Profit Q Q m Q r Graph 1a b : Monopoly profit maximization, in Carlton and Perloff (2000, 90). 54 See Posner (1975) and Carlton and Perloff (2000, 90). 20

21 Although monopolistic profits should represent only a surplus transfer from consumers to the producer/patentee and thus they should not affect efficiency, they are considered to be a loss because they are incentives to use resources for obtaining patents. This means that the overall losses imposed by patents should include the costs incurred by the patentee in competing with other agents to get a patent (rent-seeking activities). Monopolistic benefits and the deadweight loss are influenced by elasticity of demand curve: if it is close to zero, overall losses will increase. On the contrary when the demand curve becomes less elastic in the equilibrium, consumers will not be willing to renounce the products in spite of the significantly increased price. Therefore patentees exploit the situation by increasing their profits and social costs, as their increase in price can affect and reduce the quantity demanded by consumers less than in the case of an elastic demand. 2.2 The Tragedy of the Anticommons Patents may generate a phenomenon commonly known as the Tragedy of the Anticommons 55, which occurs [..] when multiple owners hold rights to exclude others from a scarce resource and no one exercise an effective privilege on use [..] 56. This may create an under use of resources due to the fact that there are no exclusive rights. Consequently the proliferation of property rights and their fragmentation create conditions for a sub optimal use of resources as a result of double externalities, on the static and the dynamic ground. 57 In a world without transaction costs bargaining between parties may prevent this Tragedy because an efficient allocation of resources will be reached spontaneously, regardless the initial legal entitlement. In the real world, characterized by high transaction costs, the 55 See Depoorter, Parisi and Shulz (2000, 2001, 2002) and also Heller (1998). Patents represent the typical case of vertical Anticommons problem, which refers to a situation characterised by a vertical relationships between exclusion property rights in terms of their sequential exercis e. Anyway a real case is often a combination between horizontal and vertical features. 56 See supra at footnote Principally the different holders of property rights cannot completely internalise the costs derived from the enforcement of their exclusion rights. The sources of externalities consist of the actual cost created by the reduction or exclusion of value of the right (static point of view) and the future costs produced as a result of the under - use of resources (dynamic perspective). See supra at footnote

22 Anticommons problem affects patents since the patented invention risks to be under used with respect to an efficient level of technology, scientific progress and welfare gain. This might be the result of the failure of the negotiation between parties when a product requires the application of several patented products or processes. The systems of licences, which would solve this obstacle, might be complicated and prevented by the supra - mentioned high transaction costs. As pointed out by some commentators 58, patents are a clear example of the recent trend in intellectual property rights to shift from a public common property, with free availability of the information contained in the public domain, to a private property regime, characterised by privatisation in R&D activities. This tendency, coupled with an excessively broad scope, has generated a proliferation of property rights, which might increase the danger of Anticommons. Indeed, the sector of biotechnological industry is often characterised by both upstream and downstream research, which consists of developments and improvements on a pioneer invention. The failure of negotiations between parties on licences may lead to the conclusion that profits are unrealistic in specific areas so no investments will be provided; this can be specifically dangerous in the biomedical areas 59 because too many owners can block each other incurring negative consequences to the health of society. Therefore, some commentators 60 recommend privatising biomedical research in a careful way to sustain both basic research and developments. Otherwise, an increase in patents delay and reduce the number of new discoveries in the field of revolutionary medicines and life-saving processes See supra at footnote See Heller and Eisenberg (1998) for a deep analysis on the subject. 60 Ibidem. 61 This is due to the failure of the internalisation of the positive externalities created between patentees of pioneer inventions and developers of improvements, which produces an Anticommons vertical situation. 22

23 2.3 Additional costs Patents generate other undesirable consequences. Firstly, they might encourage a phenomenon known as the race to invent, according to which potential inventors will compete against each to discover the new product or process. A broad scope obviously will increase the negative effects in terms of duplicative researches and waste of resources with inefficient results as a consequence. Secondly, patents impose significant transaction and administrative costs with respect to the definition of their scope. Indeed, patent breadth affects either the patent prosecution or its enforcement because a narrow scope will encourage the tendency of drafting claims quite broadly in order to compensate this limitation, while a broad scope will imply the opposite. However in both cases the specification and the claims, contained in the patent application, ought to be drafted in an accurate and precise way in order to achieve the strongest protection available. Consequently the costs always remain quite relevant. Finally, the impact created by patents on R&D activities is vital, as it influences direction and dimension of scientific research. The areas that are more profitable tend to be developed at the expense of the less ones, which are equally important from a social point of view, so concerns have been expressed with respect to a too broad scope. Unfortunately the alternative measures to patents for promoting innovations, like prizes, research contracts and joint ventures, are difficulty implemented by the State. In fact, prizes and research contracts require that detailed information is available to the State, while joint ventures are exposed to the risk of antitrust law See Carlton and Perloff (2000, 513ss). 23

24 3 THE DOCTRINE OF EQUIVALENTS The 35 U.S.C., also known as the Patent Act 63, governs patent law within the United States; a patent is issued when the invention is characterized by novelty, utility and nonobviousness. 64 The scope is determined on the basis of claims, which should be integrated by the specification and drawings in order to interpret in an adequate way. There can be two kinds of infringement: a literal one which occurs when the accused device falls literally within the patent scope, and a second type of infringement, which causes more difficult problems and refers to equivalents. This chapter will address the famous Doctrine of Equivalents, which represents the American solution for infringement under equivalence. Its evolution will be analysed in detail, as the principles fixed in several decisions constitute its core until the recent decision on the Festo case. The analysis will be enhanced by the economic rationale, with particular attention to the prosecution history estoppel. 3.1 Main principles and functions The Doctrine of Equivalents, also know as DOE, is a remedy of equity according to which there may be infringement when an accused infringing device (or process) is an equivalent to that claimed in the patent. Specifically a patentee can invoke this theory against the producer of a device [..] if it performs substantially the same function in substantially the same way to obtain the same result [..] 65. This jurisprudential theory makes it possible to extend the patent scope beyond the literal language of the patent claims; it is considered to be an exemption 66 and it has repeatedly given rise to divisions and critiques not only within the Federal Circuit but also among jurists. 63 See 35 U.S.C. 64 Ibidem. 65 See Sanitary Refrigerator Co. v. Winters, 280 U.S. 30, 42, 50 S. Ct. 9, 13, 74 L.Ed The equivalency of function, way and result is also know as the triple identity test of FWR test. See infra. 66 Judge Learned Hand mentioned it as an anomaly, to which Courts [..] resort..to temper unsparing logic and prevent an infringement from stealing the benefit of the invention. No doubt, this is, strictly speaking, an 24

25 Its controversial nature is the result of the common dilemma with regard to patent policy. There is a trade off between the interest in a clear and certain definition of the patent scope on the one hand, and in a fair and equitable entitlement of the patent value on the other one. In fact, lack of clarity may impede the desirable level of investments in R&D, while an excessively strict and literal evaluation of the scope can decrease the value of the patent. A sort of mitigation is offered by the doctrine of the prosecution history estoppel. 67 The Doctrine of Equivalents can be applied not only to pioneer inventions, but also to secondary discoveries, meant as combinations of elements to produce a new result, even if the range available may vary. 68 At the beginning of its development it was used only for mechanical components but it was later extended also to chemical ingredients in different devices and formulas. 3.2 Evolution From an historical point of view, the concept of infringement by equivalency has been developed over almost 150 years of history, but some decisions performed a special role regarding it. Firstly, Winans v. Denmead 69 consists in the major pronouncement by the Court in the past, even if the doctrine was already established before. 70 Indeed the previous cases emphasized the extension of the scope in limited situations, such as in the case of imitation and addition only of a colourable variation. 71 Secondly, Graver Tank v. Linde Air Prods. Co. 72 represents one of the most leading sentences on the doctrine, since it has introduced the requirement of interchangeability. anomaly; but it is one of which courts have frankly faced and accepted almost from the beginning [..]. See Royal Typewriter Co. v. Remington Rand. Inc., 168 F. 2d 691, 692 (2d Cir. 1948), as quoted in Chisum (2001, 874). 67 See 3.4. infra. 68 See Imhaeuser v. Buerk, 101 U.S. 647, 655, 25 L.Ed. 945, Continental Paper Bag Co. v. Eastern Paper Bag. Co., 210 U.S. 405, , 28 S.Ct. 748, 749, 52 L.Ed. 1122, Seymour v. Osborne, 11 Wall. 516, 556, 20 L.Ed. 33 and Gould v. Rees, 15 Wall. 187, 192, 21 L.Ed See Winans v. Denmead, 56 U.S. (15 How) 330, 14 L. Ed. 717 (1853). 70 See Merges (2001, 917), who underlines this position, in spite of the contrary Court s opinion in Graver Tank. 71 See Gray v. James, 10 F. Cas. 1015, 1 Peters, C.C. 394, , 1 Robb Patent Cases 120, 124 (C.C.D. Pa. 1817) (No 1015) (Washington Circuit Judge). 72 See Graver Tank & Mfg v. Linde Air Prods. Co., 339 U.S. 605 (1950). 25