Innovation Incentives
1
Innovation Incentives
THIS CHAPTER DISCUSSES the role of intellectual property rights (IPRs) as one type of incentive to promote creative research and development (R&D), taking into consideration such factors as competition, products, markets and industrial characteristics. These factors determine the environment in which incentives for creative R&D operate. Discussing the role of IPRs alone and in the abstract, as though they produced the same effects in different market conditions and for different industries, may risk defining the problem poorly, and retarding its solutions. Analyses undertaken by economists show how and in what varying degrees intellectual property protection needs differ with different industrial sectors. The cumulative (sequential) nature of innovation also requires consideration of the scope of patent protection and licensing policies to enable innovation to occur continuously. Many analyses have also been made of market structure and patent race. These analyses indicate why controversies over IPRs are likely to occur, particularly in the pharmaceutical and biotechnological sectors. By the same token, the analyses suggest that IPR protection should be integrated into R&D policies, to improve efficiencies of investment and increase possibilities for innovation. This means that coordinating and implementing various policies is a prerequisite to a balanced approach to IPRs.
I VARIETIES OF INCENTIVES
Innovation generally refers to new ideas which contribute to the creation of new technologies and products, improvements in product quality or the reduction of production costs. Innovation can be the result of the industrial application of new scientific knowledge and information, but the processes by which innovation occurs are not completely known, and attempts to innovate can be risky.
Mansfield in 1986 found among 100 randomly sampled US manufacturing firms that the patent system had a very small effect on innovation in most industries (primary metals, electrical equipment, motor vehicles, etc), but had substantial effects in a few industries, particularly pharmaceuticals and chemicals.1 According to Burk and Lemley, patent protection must have leverage (mostly achieved in the courts) to meet the needs of all new and existing technologies, to be applied with sensitivity to the industry-specific nature of innovation.2 Within industries, firm size, incumbency, corporate culture and financing have been considered to be some of the factors influencing innovation. However, no theory seems to be definitive.3 The ways in which large companies innovate are probably very different from the ways that small start-ups or mid-size companies do. Lemley, observing these findings, remarks that ‘part of the problem is that we may never be able to know exactly what sparks a thought or a creative idea in somebody’s mind’.4
Thus the causes of innovation are circumstantial, but industries have often used IPRs to ensure that their investments in R&D are recouped.
IPRs comprise such rights as copyrights, trademarks, patents, industrial design rights and trade secrets. IPR protection is considered to be a system of awards to intellectual creators such as artists and inventors. This approach is reflected in the US Constitution, which stipulates that: ‘to promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries’.5 According to Ladas, literary and artistic property is based on the right created by the law for authors of all forms of literary and artistic creation, whereas trademarks, patents, industrial design rights, utility models and trade secrets constitute industrial property,6 which is an aggregate of rights referring to the industrial or commercial activity of a person. Patents, utility models and industrial design rights are granted to inventors or creators to exclusively use or to exploit their respective creations (or to allow those to whom the creators have given consent to use or exploit their inventions) for a limited period of time.
Trademarks, trade names, appellations of origin, by contrast, are ‘vehicles of advantageous business relations . . . used to protect, maintain, and extend their activities and their association with the public’.7 A trademark is a sign which appropriately distinguishes goods or services provided by a company from those of other companies. It also serves to maintain business confidence of persons using the goods or service. It thus contributes to the development of industry and to the protection of consumer interests.
These industrial property rights over different intangible assets have historical origins which date back to very different times in history: the origin of trademarks can be traced back to the Middle Ages and even ancient Greece, and the first patent law to fifteenth-century Venice,8 but others, such as industrial designs and utility models, date only from the nineteenth century.9 Protection of trade secrets dates back to old times when there was no patent protection, but know-how is a notion that is a relatively recent development.10 Trade secrets (‘undisclosed information’, in the TRIPS Agreement) include a broad range of confidential information relating to business, production or marketing operations or processes that provide competitive advantage to the right holder. Know-how is confidential, technical information necessary for designing, manufacturing the product or information needed for any other technical or commercial operations. Under domestic laws, normally, trade secrets and know-how are defined as secret information that gives economic advantage, which the company takes reasonable measures to protect11 through secrecy of documents and contracts with employees, for example (see chapters 3, 4, 5, 13 and 14).
A patent is a legal title granting its holder the right to prevent third parties from commercially exploiting an invention without authorisation.12 The use of patents evolved over time, responding to changing technologies and different industries in various national and international contexts. Patent protection supported the marketing of new technologies and products and helped industries grow through innovation. However, filing and maintaining a patent may have multiple purposes, and, increasingly, patenting has become a business in itself.
Today, the idea that IPR protection is an incentive for innovation is disputed in various contexts. Lessig argues that new technologies make copyright protection unnecessary, and that the protection also burdens creativity.13 Jaffe and Lerner14 described situations where patent rights retarded the improvement of technology, despite licensing agreements. In a similar vein, the advent of biotechnology inventions and their patenting has reinforced the argument that patent protection of basic life-science information raises the cost of life-saving research. Concerns have been expressed by many developing countries that IPR protection hinders transfer of technology and economic development (see chapters 2, 4, 10 and 12).
In what follows, we will describe some typical incentives for undertaking R&D to innovate, with a view to exploring the potential sources of controversies over international rules relating to IPRs.
A Patents
A patent is granted by the state to an inventor, to an invention (and under certain conditions identical to those for inventions, to a discovery, in some countries) which fulfils the criteria of novelty, inventive step (non-obviousness in the US) and industrial applicability (utility in the US).
One of the important factors making up an invention (the subject matter specified by claims) is that it has not been publicly known. The novelty is tested by assessing the prior art, the state of knowledge and technology, of the country where the patent application is made, or anywhere in the world (absolute novelty).15 In this context, the medium (such as exhibitions,16 publications or goods which incorporate said ‘technical solutions’, etc) and the date of past disclosures are examined. In some countries such as the US, a period of grace is permitted between disclosure and the filing for patents. The criteria for judging ‘novelty’, therefore, are not harmonised internationally.
An invention must have achieved an ‘inventive step’, which is normally tested by the standard of knowledge of a person having an ordinary skill and knowledge in the art working in the relevant field (average expert). Patent offices and courts in the world have applied criteria with varying levels of stringency in examining the existence of ‘inventive step’. Additionally, an invention is not an abstract, theoretical construct, but must be capable of being carried out in practice. Whether or not an invention fulfils this requirement is closely related to the test of enablement, ie whether the specification sufficiently discloses the mode of manufacturing the invention in the patent specification.
Patent protection benefits the public, as the invention is disclosed, instead of being hidden. An invention must, therefore, be sufficiently disclosed in the ‘abstract’ (ie a brief abstract for the Patent Office and the public generally to determine quickly the nature and gist of the technical disclosure) and in the claims specification (a written description of the invention), as a matter of contract in exchange with the legal exclusivity. In return for the protection bestowed by the patent, the holder has to disclose the details of the invention. The disclosure of the invention in exchange for patent protection is also known as the ‘patent bargain’.17
Patents confer exclusive rights over the claimed invention, granted for a certain period of time in return for disclosure of the invention in a patent specification. The description of the invention in the specification should therefore be sufficient to allow others skilled in the art to read the specification and work on the invention after the patent expires. The disclosure of patented invention increases the likelihood that the invention will be used without restriction after the patent expires.
The scope of exclusive rights is defined in the part of the patent application known as the ‘claims’. During the period of protection, the right holder has the exclusive right to use or licence the invention and to prohibit third parties from using it without his or her authorisation. Third parties carrying out activities which fall within the claims are deemed to infringe the patent.18
Claims constitute the core of an invention and are set out as the scope of patent claims in a specification, the primary document for a patent application. A specification of an invention must be sufficiently accurate for such an invention to be carried out repeatedly by a person skilled in the art (an average expert in the same field). An inventive step means a remarkable technical level which a person skilled in the art could not have conceived easily based on prior art. The interpretation of industrial applicability includes whether the invention can be used repeatedly in production, or whether the invention relates to value creation in a market.
Markets in which new technologies are protected encourage firms to invest in R&D and, sooner or later, reduce production costs and product prices (known as dynamic effects of innovation). Patents normally encourage investment before a product is marketed, but market exclusivity does not always generate welfare. Exclusivity could diminish the use of technology in the short run, because of the difficulties of obtaining licences or the high licensing fees thereof. This, of course, depends on the licensing policies of individual firms, as well as the field of technology in question. If the term of patent protection is too long, or the scope of exclusivity too wide, the lost production due to the existence of a monopoly – often referred to as deadweight loss – causes a loss of social welfare, ie consumers lose more than the monopoly gains. It may give rise also to an inefficient allocation of resources (to the extent that resources are shifted to the monopolist from its customers). Much of the debate about patent policy has focused on this trade-off between the dynamic benefits of innovation and the static costs of market exclusivity.
The exclusive ownership of information in early research processes has become controversial, as it could impede knowledge diffusion. In such circumstances, various other incentives are sought to move towards better collaboration. However, each incentive has its own inefficiencies, and the respective pros and cons of each type of incentive should be weighted for the attainment of efficiency and welfare gains.
B Open Source
Today, the sharing of information as an incentive for innovation is considered to be a possible alternative to the exclusivity model in certain fields of technology. In the early stages of scientific research or software development, there are competing incentive systems. There are systems that use copyright protection, trade secrets or know-how protection or secrecy, and those that use more or less open-source methods.
For software, open source generally refers to a licence (e.g. terms and conditions of a software licence) which protects the rights of the software copyright holder and also allows the source code of the software to be publicly disclosed, distributed under the same terms as the licensing of the original software and modified under certain conditions. It is generally considered to be an incentive for innovation which is the antithesis of exclusivity. Although this is not unrelated to ‘freeware’, ie software which can be freely improved, open source is a corporate strategy which does not deny software developer copyrights. Open source software (OSS) is a method elaborated on for the purpose of developing software, and requires compliance with prescribed conditions such as the Open Source Initiative (OSI).19
‘Open source’ could be an efficient incentive for innovation, particularly when technological innovation is assured by the network effects of the efforts made by individuals participating in an innovation scheme, and is therefore adopted by certain businesses. There have been variations of business models relying partly on the ‘open source’ model of innovation. For example, hardware manufacturers such as IBM and SONY are presently forming the Open Invention Network (OIN)20 to enhance Linux. This system is said to be efficiently increasing the number of users by using network effects, resulting in increased sales of hardware manufactured by these developers at reduced cost to consumers. Companies can earn income from ‘open source’ innovation by selling complementary goods or services. For example, by offering consulting services relating to open source software (IBM), selling hardware incorporating the open source software (Hewlett-Packard), or offering free access to e-mail software (Microsoft).
Liebowitz and Watt find, however, that the staying power of open source is as yet unknown, and its underlying business model is not fully understood.21 According to these authors, a new business model in software (the open source software movement) has achieved a great deal of attention ‘far in excess of its current market share’. They also find that such a model may not be considered generally as an alternative to copyright protection, because ‘[e]ven the claim that creators of open source do it for reasons contrary to normal markets is questionable since many of the individuals providing the lion’s share of the work are apparently paid to do so by large corporations’.22 Whatever incentive schemes are used, what is important for all firms is to maximise profits, and at least to recoup R&D and other costs.
In the biological and life sciences, certain ‘open source’ models have been pursued, borrowing from the software model. Some scientists and researchers have advocated free sharing of genetic information, a ‘science common’ for society’s free use for interpreting and developing new products. For them, the current patent system orients scientists and universities to focus on patenting technologies that might otherwise have been placed in the public domain, and slows down the scientific understanding of the role of genetic variations in human disease and other traits.23 The relative success of freely sharing genetic information is expected in agricultural biotechnology, where the number of potential licensees is numerous and the reduction of transaction cost could therefore be a factor for consideration. Furthermore, innovation lifecycles in agricultural biotechnology are relatively short without involving costly clinical trials.24 Biological Innovation for Open Society (BiOS), for example, has shared information on the beta-glucuronidase gene (GUS) and the know-how to use it with hundreds of laboratories around the world, including Monsanto,25 to develop products. The interesting focus of BiOS is sharing patents to generate continuous innovation for the next generation of technology which, according to BiOS, would seldom occur if multinationals were the owners of lucrative patents. The BiOS-compatible licensing agreement model for patented technologies and know-how provides, inter alia, that those who use royalty-free BiOS technologies, derivatives and know-how must agree not to assert any IPRs,26 including patents, pending patent claims, or bailments against any others that have agreed to these conditions.27
In the life sciences, however, the value and cost of inventions are generally greater than in other fields of technology, and innovation cycles take longer than in software development. There is no network effect in the life science development processes. The cost of copying production of the resulting products, on the other hand, is not high. In these conditions, it may require much time and ingenuity for an open source scheme in this field to generate income to recoup investment and to become economically viable. Patent protection, which obliges the right holder to disclose information but assures them of recouping the investments already sunk at the time of patenting, may be more realistic in these circumstances.
C Subsidies to R&D
According to Scotchmer, given that the length and breadth of patent protection cannot depend on the expected costs of an R&D project, the only way to ensure that firms undertake every research project that is efficient is to let the firms collect as revenue all the social value they create. If an innovation leads to a reduction in the cost of producing a good, then the social value is the saved cost. If the innovation is an improvement to a product, the social value is the difference in consumers’ willingness to pay for the improved and unimproved products. When research firms collect all the social value as profit, shareholders rather than consumers benefit.28 However, there are situations where neither exclusivity nor open source incentives work well enough to ensure companies can earn money to invest in R&D for subsequent innovations.
Some projects which are socially desirable may not be undertaken under patent protection if the potential market is too small, because it is a decentralised, market-based economic incentive for R&D, under which the innovator’s incentive to invest in research is likely to be the economic rewards he may earn by taking risks.
When the scale of the market for a particular technology or product is not large enough to provide returns on R&D investments,29 government assistance or incentive awards may be useful. In this case, taxpayers, instead of the market and consumers, will bear specific R&D expenses in advance. Under certain circumstances, direct, public intervention in the form of contracts for research services or prizes may be more welfare-enhancing, even in market economies.
Wright assumed that patent revenues incur a higher deadweight loss than an equivalent amount of public funds financed by government contracts, and analysed the advantages of public intervention for research, such as research contracts or prizes, if administered by a social welfare maximising administrator in a competitive economy.30 His study paid explicit attention to differences in the informational roles of these alternative incentives. The system of patent protection is advantageous because it incorporates researchers’ information about the value of best inventions in the allocative processes. Both patents and prizes utilise private information about research costs and the probability of success, whereas a public administrator may not. Wright concludes that the allocative advantages of contractual research and of prizes may be undervalued if researchers are highly responsive to incentives. He indicates, however, that the choice of incentives depends on the nature of the information search process, and he suggests that further studies should take into account researcher attributes.
Market size is insufficient when either (i) potential beneficiaries are few, or (ii) each beneficiary (or his insurer/government) has a low willingness to pay. In either case, the value of the innovation (relative to other R&D projects) could actually be low, in which case government intervention may not necessarily be justified. At least from an economic point of view, if the cost of R&D (plus the cost of production) is greater than the utility that consumers derive from product, then innovation is socially wasteful. Government R&D subsidies should therefore be awarded with care: only when a specific market failure would prevent efficient innovation, or if they are necessary to achieve specific policy objectives. R&D on new medicines for neglected diseases is an example of a socially desirable policy objective which might require government subsidy.
II R&D INVESTMENTS INCENTIVES IN DIFFERENT INDUSTRIAL SECTORS
A The Role of Complementary Intellectual Assets (Trade Secrets and Know-How)
The ease and speed with which innovative products or processes can be imitated by a third party vary in different industrial sectors, depending on the ways in which the invention is technically or commercially exploited. This in turn depends in part on how easily know-how necessary for designing, manufacturing and distributing the patented products can be learned.