Conceptualizing the Breeding Exception to Patent Rights: A Legal and Economic Appraisal

This type of exception is seen as positive for innovation since it avoids the transaction costs of licensing for research purposes on the subject matter. In addition, it enhances innovation by increasing the difference between the return from the new innovation and that from the old innovation. As a consequence, prices for consumers lower. Research on protected inventions seems to be more desirable when research is conducted for academic or non-commercial purposes. In this case, the low willingness of researchers to pay licensing fees, issues related with avoiding patent enforcement in academia and the large spillovers of academic research seem to justify the adoption of the patent exception.²⁴ Considering the importance of academic research for the progress of science, it appears advisable to adopt a broader exception if patents on basic research are broad and vice versa.²⁵ Moreover, even if ‘the commercial versus the non-commercial distinction is dropped and the focus is returned to the underlying goals of the patent system, the analysis of experimentation “on” the subject matter of an invention shows that it is essentially a species of enabling disclosure. Such disclosure enhancement promises to have beneficial effects on the pace of follow-on innovation for non-self-disclosing inventions and minimal effects on the invention for self-disclosing inventions.’²⁶ Therefore, this type of exception has not been object of large controversy so far.

While there is general agreement between economists on patent exceptions adopted for conducting research on protected subject matter, they are more skeptical with respect to the second type of exception, which allows using patented inventions as a tool in the research process. The main controversial argument on the economic feasibility of an exception that makes use of inventions as research tools lies on the fact that ‘one man’s research tool is another man’s business’.²⁷ There is common belief among economists that the use of research tools by second innovators might deprive the first innovator of substantial profit. This may be especially true for inventions with a significant market among researchers, but exempting even purely academic research from patent infringement could also deprive patent holders of a portion of their expected profits. Consequently, their incentive to innovate will be undermined.²⁸ Those against the introduction of an exception on research tools further argue that there exists no barrier to licensing.²⁹ They see no reason for the patent holder to refuse bargaining with other innovators, thereby, an exception to patent rights is deemed unnecessary.

Other commentators fear that, on the contrary, the first inventor might be granted too much power in the innovation process. This stems from the different characteristics of innovativeness embedded in research tools. While for most innovations, the contribution that the invention makes is the inventive idea, for research tools, is the tool itself.³⁰ Therefore, the patentee might have more interest in ensuring a financial return from exclusive control of the research tools rather than from widespread commercial use of an embodiment of the inventive idea. If patentees control research tools in view of maximizing their own profits, technological progress will slower at the expense of the society.³¹ This might be particularly worrying when the research project is aimed at addressing important societal concerns, such as food or health issues. In this case, the societal harm resulting from a slower technological progress may be very severe while the private incentives to delay so as to keep a larger share of the monetary and nonmonetary benefits of the research may be stronger.³²

This hypothesis generates worries because it weakens the primary function of research tools: the facilitation of technological progress. Hence, the relevant question becomes: how to provide appropriate returns for incentivizing inventions while spurring technological progress? In this regard, Mueller proposes a limited exception for experimenting with research tools that compensates the patentee for use of the tool through a compulsory licensing requirement.³³ This proposal has been further elaborated by Strandburg, who subjects the exception to two conditions.³⁴ The first condition requires that there must be no close substitutes for the research tool, while the second prohibits close substitutes for the research projects that require the tool.³⁵ She suggests that only when the tool is of unique importance to a uniquely important problem, it will have an impact on the society. If researchers are relatively indifferent between problems requiring a patented tool and a whole host of interesting problems for which they do not need to use the tool, then the patentee will not exercise significant power over research progress.

These considerations allow inferring that economic theory on patent exceptions is centered around the incentive to innovate. Empirical studies, however, are not able to define the ‘right’ scope of patent exceptions given the lack of a direct correlation between R&D expenditures and the breadth of experimental use exception, and the heterogeneity of subject matter. Therefore, it is impossible to find a definitive answer in economic arguments. Nevertheless, economic reasoning provides useful guidelines for evaluating the effects and desirability of patent exceptions. This guidelines aim at increasing social welfare, which in turn is inspired by public policy issues.

‘Public policy’ is usually understood as ‘the governing policy within a community as embodied in its legislative and judicial enactments which serve as a basis for determining what acts are to be regarded as contrary to the public good’.³⁶ Therefore, in terms of patent law, public policy considerations should aim at guiding the legislator towards the public good, which itself is the main underlying concept of patent law. Indeed, the rationale of patent law is not to remunerate inventors, but to promote the disclosure of their innovations for the benefit of society. The disclosure of inventions avoids duplication of R&D investments and facilitates follow-on inventions. In turn, more knowledge accrues to society. This makes it apparent that patent law is conceived for the good of the society.

In these terms, some exceptions can be considered as necessary to the patent system.³⁷ For example, when inventions are not-self-disclosing, exceptions are considered necessary for the benefit of the public.³⁸ This necessity becomes even more imperative when patents are granted on basic research given its influence on subsequent innovations. It is often suggested in the literature that when patents on basic research are broad, then the research exception should also be broad.³⁹ The adoption of exceptions, in this case, recognizes the fact that the purposes of the patent system cannot be achieved due to the exclusive rights conferred upon the patentee. In any case, the exception should be granted in view of the public interest and the purpose of the exception.⁴⁰

This view implies a balance of countervailing interests. Whenever a patent conflicts with an important social goal, the latter should be given priority. Important social goals may be represented but not limited to ‘national security and emergencies, such as the exception in the Bangui Accord that permits exploitation by an administration or organization authorized by the Minister of the Member State concerned, for the purposes of “vital economic interest, public health, defense other country’s needs”, subject to remuneration’.⁴¹ Thus, public policy issues guide the adoption of exceptions to patent rights whenever there is a need to disclose the invention for the benefit of the public. In this respect, a WTO panel has also stated that the absence of exceptions to patent rights ‘would frustrate part of the purpose of the requirement that the nature of the invention be disclosed to the public’.⁴² The relevance of public policy issues in judicial decision-making will be further explained when analyzing national practices on research exceptions.

Here the focus will be on the US research exception in order to provide an example of exceptions developed through judicial decisions. The choice of the US exception is based on the fact that that the US courts offer a rich and interesting jurisprudence for the purpose of the analysis. The birth of the research exception in the US is attributed to Justice Story, who affirmed in the Whittemore v Cutter case that:

it could have never been the intention of the legislature to punish a man, who constructed such a machine merely for philosophical experiments, or for the purpose of ascertaining the sufficiency of the machine to produce its described effects (emphasizes added).⁴⁶

In the same year, he explained in Sawin v Guild that experimentation must not be with an intent to infringe patent rights and deprive the owner of the lawful rewards of his discovery.⁴⁷ These two cases seem to create the basis for differentiating between experiments for research and commercial purposes. Whereas the former have been deemed permissible by courts, the latter have been strictly prohibited. A landmark case—Roche Products v Bolar Pharmaceutical ⁴⁸ or so-called Bolar case has mitigated this view. The case will be briefly examined below in order to understand the circumstances that influenced the decision.

The defendant, Bolar Pharmaceutical, was a generic drugs manufacturer who aimed at marketing a drug covered by a patent of Roche Pharmaceuticals, the plaintiff. The marketing of the generic drug required a marketing permission from the health authorities. Since the process for obtaining this permission was time-consuming, Bolar decided to conduct clinical tests and apply for marketing authorization prior to the expiry of Roche’s patent. Roche filed suit. Bolar’s defense was based on the experimental use exception. But the appeal court said that it was not applicable to the concrete case since the exception is narrow and limited only to experiments which are not in any way motivated by commercial interests. This decision negatively affected the interests of drug manufactures which exercised pressure upon Congress to change the result of the decision. Less than 6 months after, the Congress emanated Section 271-e-1 of the Drug Price Competition and Patent Term Restoration Act, informally known as the ‘Hatch-Waxman Act’.⁴⁹ This solution was a compromise between the diverging interests of the generic and proprietary drugs manufactures.⁵⁰ Although it softened the categorical prohibition of experiments with commercial intent, it allowed a research exception solely for the purpose of obtaining regulatory approval. Patentees interests were safeguarded by enabling them extend patent rights beyond the statutory patent term of 17 years from the grant of the patent. In this way, they were compensated for their ‘loss’.

This decision is significant on an international level because it paved the way for other countries to adopt exceptions to patent rights for regulatory purposes.⁵¹ The compliance of such exception with the main international rules for IPRs has been clarified by a WTO panel.⁵² However, the exception is very narrow and limited only to experimental trials for obtaining regulatory approval for generic medicines. This was further reaffirmed in Embrex v Service Engineering, where the court of appeal indicated that they were extremely narrow in scope and could not be used to escape liability for infringement simply by cloaking infringing activities in the ‘guise of scientific inquiry’.⁵³

A more recent case that supports this view is Madey v Duke University.⁵⁴ Madey was a researcher at Duke University, who owned patents covering a new laser device, and left Duke University without licensing or assigning any rights to his inventions. After he left, the university continued to use his inventions without permission. The court found Duke’s research ineligible for the common law research exception because it ‘unmistakably furthers the institution’s legitimate business objectives, including educating and enlightening students and faculty participating in these projects’. This decision was justified by the fact that academic institutions have legitimate business objectives since the research they conduct increases the status of the institution and lures lucrative research grants. The court, however, reaffirmed that the research exception remains viable for experimental-use that is ‘for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry’. Such a restrictive interpretation of the common law research exception in the US has induced most commentators to argue that Madey v Duke signals an end to the common law research exception to patent infringement.⁵⁵

The restrictive approach of US judges is further confirmed with regard to the interpretation of the statutory exception. In Integra Lifesciences v Merck, the court argued that basic scientific research not aimed at obtaining regulatory approval is not exempted under Section 271-e-1 of the Hatchman-Waxman Act since the research is surely not ‘reasonably related to the development and submission of information’ to the Food and Drug Administration (FDA) authorities.⁵⁶

The research exception introduced in civil law countries is formulated in various ways. Although there are some differences in the language used, the practice followed by countries is to provide an exception for ‘acts done for experimental purposes’. In many EU States, for example, the statutory exception is identical or substantially similar to Article 27.b of the Agreement on a Unified Patent Court (ex art. 31 of the Community Patent Convention). The Czech Republic, Denmark, France, Germany, Ireland, Luxembourg, Spain, Sweden, and the UK all almost identically reproduce this provision relating to the research exception, namely ‘acts done for experimental purposes relating to the subject matter of the patented invention’.⁵⁷ Other countries have chosen different formulations. Italy, for instance, exempts non-commercial or experimental acts aimed at obtaining regulatory approval for pharmaceutical products.⁵⁸ Belgium, on the other hand, has adopted a broad exception by expressly permitting scientific acts done both on and with the patented invention.⁵⁹ The particularity of this exception is its impact on protected research tools. The Swiss Patent Act also distinguishes between research on the patented invention and research with the patented invention.⁶⁰ But only research on the patented invention is exempted since access to patented research tools is explicitly assured through non-exclusive licensing.⁶¹ Unlike most countries, Swiss law exempts research done for both non-commercial and commercial purposes, as long as the objective of the research is to generate new knowledge about the patented invention.⁶² Similarly, the Dutch exception relates solely to research on the patented subject matter. Article 53.3 of the Dutch Patent Act of 1995 provides as follows: ‘The exclusive right shall not extend to acts solely serving for research on the patented subject matter, including the product obtained directly as a result of using the patented process.’ The Dutch law, thus, clarifies that the scope of the research exceptions extends to the products of research.

This brief overview of statutory exceptions makes it apparent that civil law countries have adopted a liberal approach with respect to exceptions for research purposes compared to US practice. Civil law courts have further contributed to this approach by embracing a broad interpretation of the scope of research exceptions. Two German decisions, Clinical Trials I⁶³ and Clinical Trials II⁶⁴ are a cornerstone in this regard and they are often brought as an example in many EU countries when interpreting research exceptions.⁶⁵ These decisions will be examined below.