Patent examination priorities.

• Author: Meurer, Michael J.

TABLE OF CONTENTS INTRODUCTION I. A MODEL OF PATENT EXAMINATION ERRORS A. Introduction to the Model B. The Source of Errors C. The Social Cost of Examination Errors II. STRATEGIC PROSECUTION III. INCENTIVES AND MONITORING IV. EFFORT COST V. FUZZY BOUNDARIES VI. EXAMINATION REFORM CONCLUSION INTRODUCTION

The United States Patent and Trademark Office (PTO) faces a massive and growing backlog of patent applications. (1) The PTO cannot seem to increase its staffing fast enough to keep up with an explosion of applications, especially given problems with examiner morale and retention. (2)

At the same time, the PTO struggles to improve examination quality. (3) A decade or more of criticism pushed the agency to start experimenting with new examination methods. (4) One experiment, called the "second pair of eyes" review, required a second examiner to review an application that had been allowed by the original examiner. (5) The experiment was introduced in 2000 to improve the quality of certain kinds of business method patents. (6) A Federal Trade Commission report praised this program and endorsed its expansion. (7) The difficulty, of course, is that a second look at a patent application absorbs examiner time that could instead be used to address the application backlog.

Mark Lemley vividly commented on the resource constraint facing the PTO and its implications for patent quality. (8) He noted that patent examiners devote an average of eighteen hours to each patent they examine. (9) This compares unfavorably to what I expect of my students when they write research memos. Eighteen hours certainly is not much time, and thus, we should not be surprised if low quality patents sometimes issue.

Lemley defended "rational ignorance" at the PTO and argued against investing heavily in more examination hours as a method of improving patent quality. (10) Few patents are ever asserted, licensed, or litigated. (11) Thus, it would be bad policy to exert much effort to perfect the bulk of patents because they go unused. (12) Lemley contended the better choice is to continue the current practice of a relatively cursory review and save a more resource-intensive review for litigation, if and when litigation arises. (13)

The PTO has tentatively embraced two other paths out of this dilemma: shift some examination responsibilities to other parties, and stem the explosion of patent applications. In 2007 the agency initiated Community Patent Review, also known as the Peer to Patent program. (14) The goal of this experiment is to improve the quality of technical information available to patent examiners by enlisting volunteers to search for and share relevant "prior art" with examiners. (15) Prior art is patent jargon that refers to patents, publications and other information that can be used to limit or invalidate patent claims because, for example, they lack novelty or are obvious. (16) Commentators and patent officials are also considering outsourcing patent searches to for-profit companies and relying more heavily on searches and other examination activities by the Japanese and European Patent Offices. (17) All of these programs have potential to improve patent quality without consuming additional examiner time.

The PTO recently proposed a complementary set of regulatory changes that could have reduced the workload on examiners. (18) One proposed change would have limited the number of continuing applications that an inventor could submit. (19) There is evidence that one cause of the explosion of patent applications is a surge in the use of a string of related "continuing" applications that are linked through a shared disclosure contained in a single initial application. (20) This regulatory change was successfully challenged on the grounds that the reform was substantive and the PTO does not have substantive rule-making authority. (21) Other proposed changes would directly or indirectly limit the set of claims contained in a patent. (22) Reforms that successfully reduce workload could free up resources that could be used to improve examination quality.

Measures that discourage excessive patenting and claiming, propose shared examination responsibilities, and increase staffing all have potential to raise examination quality and alleviate the patent application backlog. So far these measures have been too limited to have much impact, and there is insufficient evidence to reliably judge their effectiveness. In this Article, I consider a different approach to examination reform. I take as given a significant scarcity of examiner time, and I ask how the PTO should set examination priorities. In other words, how much of their eighteen hours should examiners devote to the various tasks they are expected to conduct before allowing a patent to issue? (23)

It is vital to recognize that examiners will make mistakes given the time constraints that they face. (24) Thus, good patent policy requires much thought be given to minimizing the expected social cost from those mistakes. Part I of this Article sketches a model of patent examination errors and explores the social costs associated with various types of errors. The model supposes that patent examination leads to three possible outcomes: rejection of a patent application, allowance of a patent with narrow claims, or allowance of a patent with broad claims. Errors arise when the examination outcome for some reason departs from the proper outcome.

The remaining sections of the Article add complications to the basic model. Part II considers strategies that might be used by patent applicants in response to various choices of examination priorities. Good examination policy must anticipate how patent lawyers respond to examination practice. Part III discusses human relations management within the PTO. Any employer who wants to successfully implement a reform in an organization's core processes needs to think carefully about how to monitor and reward employees for advancing reform--the PTO is no different in this regard. Part IV notes that examination priorities should depend on the cost of performing various examination tasks. Part V replaces the assumption of Part I that claim scope is determinate with the assumption that claim scope is fuzzy. Finally, Part VI takes a quick look at policy issues presented by various examination reforms.

The goal of this Article is to identify policy issues that must be addressed by analysts who want to set examination priorities and to help guide empirical research that will be needed to make good policy judgments. At present, patent scholars know very little about patent examination errors.

1. A MODEL OF PATENT EXAMINATION ERRORS

   1. Introduction to the Model

      Inventors who want to patent their inventions normally hire a patent agent or attorney who drafts a patent application and submits it to the PTO. (Hereafter, I will call the patent agent or attorney the "prosecutor" in keeping with patent law jargon.) The key features of a regular utility application are the written description of the invention and the claims. (25) At a minimum, the written description teaches people skilled in the art how to make and use the invention. (26) The claims articulate patent-based exclusionary rights over the invention. Patent law commentators often speak of a hypothetical bargain between the inventor and the public in which the inventor discloses technical information to the public in exchange for exclusionary rights. (27) The scope of the rights corresponds to the quality of the disclosure; more valuable disclosures entitle the inventor to broader claim scope.

      Patent prosecution and examination is an ex parte administrative proceeding that is often described as a bargaining session. (28) Initially, the prosecutor typically asks for broad rights, broader than justified in light of the disclosure. The examiner responds by objecting to or rejecting aspects of the application, (29) and the prosecutor comes back with arguments or changes to the application that are responsive to the concerns of the examiner. (30) The examiner is supposed to serve the needs of the inventor-applicant, but also assure that the application complies with the requirements of the Patent Act. (31) When this bargain reaches an impasse and the examiner will not allow the patent to issue despite arguments or amendments by the prosecutor, then some inventors will abandon their application, and others will appeal within the agency to the Board of Patent Appeals and Interferences (BPAI). (32)

      I will model certain aspects of the prosecution and examination process to uncover the role of various examination tasks in generating examination errors that degrade examination quality. I assume that there are three possible outcomes of examination: no patent issues, a patent with narrow claim scope issues, or a patent with broad claim scope issues. For the time being I will ignore appeal and litigation. I will assume that the standards of patentability are well-conceived, and therefore, I will speak about the proper outcome of examination as an outcome in which an applicant gets the broadest rights it is entitled to. Table One displays the proper outcome of an examination in the three rows. The columns of the table display the outcome chosen by the examiner. Naturally, there are three choices, and thus nine entries in the table.

      Correct examination outcomes fall on the diagonal where the proper outcome and the examination outcome match. The off-diagonal entries represent different kinds of mistakes. The three italicized entries in the southwest part of the table represent false negatives--the examiner should have allowed a patent to issue, but did not, or should have allowed a broader patent. The three underlined entries in the northeast part of the table represent false positives--the examiner should have rejected or narrowed the application.

      I have further divided the errors in a natural fashion. Perhaps the...