Rules for patents.

AuthorBurstein, Michael J.


There is widespread agreement that the patent system in the United States is in need of reform. Most of the proposals for patent reform that have proliferated in recent years share two common assumptions: first, that patent policy is best made through case-by-case adjudication of the validity of individual patents; and, second, that the existing allocation of authority over patent policy, in which the courts are primarily responsible for interpreting and applying the broad language of the Patent Act, ought not to be disturbed. This Article challenges both assumptions. I approach the problem of patent reform primarily as a problem of sound administration rather than innovation policy and argue that Congress should grant the Patent and Trademark Office (PTO) substantive rule-making authority.

The administrative structure of the patent system has been largely unchanged since 1836. But the administrative tasks that a well-performing patent system must carry out have changed markedly since that time. Most importantly, technology in the early- to mid-nineteenth century was relatively uniform. Today, by contrast, the process of innovation varies widely among different technologies and different industries. If the patent system is to meet its goal of providing incentives for innovation, it must self-consciously tailor the elements of patentability--both rules and standards--to those diverse circumstances.

Optimal patent policymaking requires forward-looking deliberation and cost-benefit analysis based on technological and economic expertise; clarity and predictability so that entities making investment choices based on the property-like aspects of patents can be confident in the legal regime governing those rights; and transparency and accountability to ensure that the public interest--which is often distinct from the interests of patent holders--is taken into account. Unlike courts, agencies acting through rulemaking can gather and expertly analyze all of the relevant information to make express policy judgments based on costs and benefits, can decide issues prospectively and avoid piecemeal decision making, and can systematically engage the public in the policy-making process. Although agencies are subject to certain well-understood institutional pathologies, such as capture by powerful interests, on balance they are more likely to make effective patent policy than courts.

Granting the PTO substantive rule-making authority would require significant changes to the structure and function of that agency, and to the role of the courts. The PTO would require the addition of a policy-making capacity separate but capable of drawing insights from the examination process. The courts in turn would play a constructive secondary role by surfacing issues that require attention in the interstices of agency rules and engaging in judicial review of those rules under traditional standards of administrative law.

TABLE OF CONTENTS INTRODUCTION I. THE CHANGING TASKS OF PATENT ADMINISTRATION A. Patent Administration in Historical Perspective B. Administrative Tasks Confronting the Modern Patent System II. WHY PATENT POLICY SHOULD BE MADE THROUGH AGENCY RULEMAKING A. A Normative Theory of Patent Administration 1. Reasoned Decision Making and Expertise 2. Clarity and Predictability 3. Transparency and Accountability B. Allocating Policy-Making Authority 1. Reasoned Decision Making and Expertise 2. Clarity and Predictability 3. Transparency and Accountability C. Evaluation III. HOW RULEMAKING WILL CHANGE PATENT ADMINISTRATION A. The PTO B. The Courts 1. Judicial Review of Agency Rulemaking 2. Litigation Over Patent Validity or Application Denials CONCLUSION INTRODUCTION

There is widespread agreement that the patent system in the United States is broken. A flood of critical commentary argues that the system is in "crisis," (1) or that it is a "failure." (2) At the foundation of these complaints is a belief that the system established in order "[t]o promote the Progress of Science and useful Arts" (3) no longer provides the optimal incentives for innovation. (4) A patent provides an incentive to invest in costly invention--or, more controversially, to develop or commercialize an inventions--by providing the inventor with a mechanism by which she may potentially recoup her investment in producing otherwise freely appropriable knowledge. (6) The standards for granting a patent are critical to ensuring that this incentive works. If patents are too few or too narrow, they do not provide sufficient protection for costly research and development (R&D). If they are too numerous or too broad, they may over protect, inhibiting access to critical inputs by downstream inventors. (7)

The standards of patentability are defined by statute. Patents may issue for "any new and useful process, machine, manufacture, or composition of matter" (8) that is not "obvious at the time the invention was made to a person having ordinary skill in the art" (9) and that is described "in such full, clear, concise, and exact terms as to enable any person skilled in the art ... to make and use the same." (10) Like any broadly written statute, the Patent Act requires interpretation to have any concrete meaning. Indeed, the patent statute requires policymaking to fill in the gaps that Congress left through its choice of broad terms. (11) That task has been delegated largely to the courts. (12) This Article argues that assigning the primary responsibility for articulating patent policy to the courts is suboptimal. Instead, Congress should grant the Patent and Trademark Office (PTO) the authority to make substantive rules implementing the Patent Act.

Most critics of the patent system make two assertions, either explicitly or implicitly, about the institutional structure of the patent system: first, that patent policy is best made through case-by-case adjudication of the validity of individual patents; and, second, that the existing allocation of authority over the standards for determining patent validity--in which the courts are primarily responsible for interpreting the broad language of the Patent Act--ought not to be disturbed. This Article challenges both contentions.

Most critics generally approach patent reform through the lens of innovation policy. They ask what the standards of patentability ought to be. I ask instead how the standards of patentability ought to be determined. My central claim is that our current institutional structure fails to produce optimal policy outcomes and should be reformed. Consider one recent example that illustrates the dysfunctional nature of patent policymaking: In 1982, the PTO issued the first patent on a human gene. (13) Almost thirty years later, a district court held that human genes were not patentable subject matter. (14) Human genes, in the court's view, were not a patentable "process, machine, manufacture, or composition of matter" to which the Patent Act applied. (15) In the time between the issuance of the first gene patent and the district court's decision, many issues surrounding the validity of patents on genes or DNA were litigated, including whether certain types of DNA molecules were sufficiently useful for patent protection, (16) whether a patentee seeking to patent DNA must disclose the entire genetic sequence he seeks to patent, (17) and whether isolated DNA sequences were "obvious" and therefore unpatentable. (18) Yet the fundamental question whether patents on human genes even constitute eligible subject matter was left unaddressed. (19) In the meantime, thousands of patents issued for DNA sequences. (20) An entire industry based on recombinant DNA technology grew up with the understanding that patents on DNA sequences were valid. The district court's decision cast the settled expectations of many in the biotechnology industry into doubt and threatened widespread chaos. (21)

This story illustrates at least two ways in which the system fails. The first concerns the substance of the court's analysis. The Patent Act does not answer directly the question whether human genes constitute patentable subject matter. (22) In the absence of legislative guidance, the goal of fostering innovation would be best served by an evaluation of the extent to which patents on genes would promote or inhibit incentives for research and development in biotechnology. (23) But such reasoning is absent from the court's opinion. The court instead drew primarily upon precedent to hold that genes are "products of nature" and therefore not the proper subjects of patent protection. (24) Of course, the "products of nature" exception itself is a judicially created limitation on patentable subject matter that developed not from economic analysis, but from the slow accretion of precedent. (25) The court's failure to focus on the economic factors that would shape a policy-based appraisal of gene patents' desirability is consistent with the judiciary's general approach to the problem of patentable subject matter. (26) That approach has drawn criticism from scholars who believe that the application of formalist legal reasoning is producing poor patent outcomes. (27)

The second failure arises from the uncertainty created by judicial decision making. In the case of gene patents, the vagaries of litigation prevented a core question of patentability from being litigated before the industry assumed significant reliance costs. Even after the question of patentable subject matter in the case was decided, there was little guidance concerning the status of gene patents in future related cases. Bessen and Meurer, among others, have criticized the patent system's failure to provide adequate notice to entrepreneurs and investors about the scope of patent rights. (28) Judicial uncertainty contributes mightily to this problem. (29)

The central problems that the patent system faces are not all that different from the institutional design problems...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT