Introduction to the Economics of Patents

Introduction

One of the key questions in evaluating potential firms is whether their competitive position is sustainable. Firms can protect these positions in many ways, including by developing a favorable reputation ("brand equity") and economies of scale in manufacturing that other firms cannot duplicate. In most high-technology industries, however, two methods of protecting discoveries are most critical: patents and trade secrets.

Volumes can—and have!—been written about each of these ways to protect intellectual property. These notes on the economics of patent and trade secret protection can do little more than provide an introduction to these sources of protection. I also include a bibliography of several articles and books, however, that provide much more detailed information.

Patent Protection

A patent application to the U.S. Patent and Trademark Office (USPTO) essentially consists of a series of claims and supporting documentation. [Much of this section is based on Klitzman, 1984; Schwartz, 1988; and U.S. Department of Commerce, 1992.] Some of the claims in a patent application will be cast in very specific terms; others may be sweeping. A supervising primary examiner reviews each incoming patent application, and assigns it to one of the over 120,000 U.S. patent subclasses. This classification determines which examining group reviews the application.

A patent examiner in the assigned group then evaluates the proposed patent. To assess the novelty of the application, he searches previous patents issued in the original and related subclasses and several on-line databases. To be entitled to utility patent protection (the most common form of U.S. grant), an innovation must satisfy three criteria. Under 35 U.S.C. 101-103 and 112, it must be:

• a process, machine, manufacture, or composition of matter.
• new, useful, and nonobvious.
• disclosed in sufficient detail that a skilled person could build and operate it.

The scope of patentable subject matter has traditionally not included fundamental scientific discoveries. A frequently invoked rationale for this omission is that many scientists care little for monetary rewards, and would consequently have pursued the discoveries in any case. To grant patent awards for purely scientific discoveries would consequently be socially wasteful. [See, for instance, Katz v. Horni Signal Manufacturing Corp., 52 F.Supp. 453, 59 U.S.P.Q. (BNA) 196 (D.N.Y. 1943), rev'd, 145 F.2d 961, 63 U.S.P.Q. (BNA) 190 (2d Cir. 1944), cert. denied, 324 U.S. 882, 65 U.S.P.Q. 588 (U.S. 1945).]

If the application appears to conform to the other standards for patentability, the patent examiner will then determine whether the claimed innovation conflicts with any in-process applications or recent patent awards. Unlike most nations, the United States grants patents to the party that is "first to invent" a new product or process rather than the one who is "first to file" for an award. If the application appears to supersede another application or a recent award, the examiner will declare the patent application to be in interference. Often applicants will provoke interferences. The standard approach is for the firm to "copy claims": i.e., to include in its application the same wording as in another firm's claim. Even though USPTO holds pending applications confidential, firms often provoke interferences with patents that are still pending. They are able to do so because of the timing of the patent application process. To receive European patent protection, firms must file an application at the European Patent Office (EPO) within one year of the U.S. application. Eighteen months after the original application, the EPO publishes the key information about the pending patent.

Disputes are turned over to the USPTO's Board of Patent Appeals and Interferences. The firm whose patent has been interfered with (the senior party) is notified, and each party reviews the other's application. The Board will hold a hearing to determine which inventor first made the discovery. It will decide whether the junior party's patent should be allowed in its entirety or whether some claims should be disallowed. If the senior party's patent has not yet been issued, its claims may be scaled back or even rejected entirely. If the senior party's patent has been issued, the Board cannot take away the award or retroactively reduce its scope. A finding casting doubt on the senior party's patent, however, can be used by the junior party if and when it challenges the patent in Federal court.

Thus, prior to the award of the junior party's patent, the two parties are often aware of each other's patent position. The lengthy interference process serves to bring firms together early in the dispute process and facilitates the negotiation of a cross-licensing agreement to settle the controversy. In recent years, 80% of the interferences have been settled prior to a final hearing by the Board [Calvert and Sofocleous, 1992].

At the time of award, the patent examiner assigns the patent to one or more U.S. patent subclasses. The examiner has a strong incentive to classify these patents carefully, because he uses these classifications in his searches of the prior state-of-the-art. To insure the accuracy of the classification and to maintain consistency across examining groups, an official known as a "post classifier" reviews the classification of all issuing patents.

After the patent is issued, the primary forum for formally resolving disputes is the Federal courts. The Federal courts have exclusive jurisdiction over disputes involving the infringement of patents, as well as over appeals of USPTO decisions. Other disputes--e.g., a disagreement between a firm and an employee over a royalty--are routinely referred back to the state courts. If a firm believes that a patent is being infringed, it may sue the infringer for damages and/or injunctive relief (a judgment ordering the defendant to cease infringing the patent). Conversely, the alleged infringing firm may preemptively sue the other firm for declaratory relief (a judgment that the plaintiff is not infringing any patent held by the defendant).

In either event, the initial litigation must be undertaken in a district court. Prior to 1982, appeals were heard in the court of appeals of the district in which the case was tried. These circuit courts varied considerably in their interpretation of patent law, and the resolution of these differences through appeals to the U.S. Supreme Court was a lengthy and uncertain process. Consequently, the Court of Appeals for the Federal Circuit (CAFC) was established as the appellate court for all patent-related Federal cases. CAFC decisions may still be appealed to the U.S. Supreme Court, but the latter seldom agrees to hear such appeals.

At any point in the litigation process, the adversarial parties may settle their dispute. This agreement may or may not be accompanied by compensation for retroactive relief and/or a patent license or cross-license agreement. If the settlement is reached before the filing of a suit or a decision within USPTO, the existence of the dispute is unlikely to become public knowledge. The settlement of interferences or post-award disputes are rarely announced, and certainly not in any systematic manner.

Practitioner accounts suggest that the impact of patent litigation has grown with the strengthening of patent rights. This shift towards a more "pro-patent" policy has been effected partially through legislation—e.g., the Computer Software Protection of Act of 1980 and the Semiconductor Chip Protection Act of 1984—but even more so through the decisions of the CAFC. When the CAFC was created in 1982, its stated purpose was to be a streamlined venue for treating patent cases in a systemized manner. But as Merges [1992] notes,

While the CAFC was ostensibly formed strictly to unify patent doctrine, it was no doubt hoped by some (and expected by others) that the new court would make subtle alterations in the doctrinal fabric, with an eye to enhancing the patent system. To judge by results, that is exactly what happened.

This claim is supported through a comparison of CAFC's rulings with previous appellate decisions in patent infringement cases. Between 1953 and 1978, circuit courts affirmed 62% of district court decisions holding patents to be valid and infringed, and reversed 12% of the decisions holding patents to be invalid or not infringed [Koenig, 1980]. In the years 1982-90, the CAFC affirmed 90% of district court decisions holding patents to be valid and infringed, and reversed 28% of the judgments of invalidity or noninfringement [Harmon, 1991].

This shift seems to have been associated with shifts in both patenting behavior [see Kortum and Lerner, 1998, for an extended discussion] and litigation. This note concerntrates on the rise in corporate patent litigation. In an analysis of intellectual property litigation involving a sample of 530 firms based in Middlesex County, Massachusetts, Lerner [1994] finds that these firms engaged in 78 distinct patent suits between January 1990 and June 1994 in Federal District for Massachusetts and Middlesex County Superior Court. During the same period, these firms were awarded 2,533 patents. Because firms generally attempt to litigate cases in the district encompassing their headquarters, this litigation probably represents about one-half the patent suits involving these firms. (Because there was considerable variation in the interpretation of patent laws across the federal circuits before the creation of the CAFC in 1982, firms frequently attempted to litigate cases in districts that they believed were predisposed to their arguments. The creation of the centralized appellate court has led to a considerable reduction in the differences in intellectual property law across the circuits. As a result, "forum shopping"—the filing of cases in districts perceived to be favorable—declined sharply. Firms today generally attempt to litigate these cases in the same district as their headquarters. This allows them to employ the same outside lawyers that they usually utilize, and to make greater use of internal corporate counsel.) The analysis suggests that approximately 6 patent suits are filed for each 100 corporate patent awards. Particularly striking, practitioner accounts suggest, has been the growth of litigation--and threats of litigation--between large and small firms. Several well-capitalized firms, including Texas Instruments, Digital Equipment, and Intel, have established groups that approach rivals to demand royalties on old patent awards. In addition to litigation in the courts, in recent years nearly 4,000 quasi-judicial administrative procedures have been conducted annually within the USPTO [U.S. Department of Commerce, 1991].

These suits lead to significant expenditures by firms. Based on historical costs, the patent litigation within USPTO and the Federal Courts begun in 1991 will lead to total legal expenditures (in 1991 dollars) of about $1 billion, a substantial amount relative to the $3.7 billion spent by U.S. firms on basic research in 1991. This estimate of litigation costs is based on documents by the USPTO and other sources. USPTO's estimate of the number of patent suits is likely to be low, because (i) many cases that involve patents are classified as contract, miscellaneous tort, or other cases, and (ii) many suits classified as patent cases are nonetheless not reported by the federal clerks to USPTO.

Litigation also leads to substantial indirect costs. The discovery process is likely to require the alleged infringer to produce extensive documentation, as well as time-consuming depositions from researchers and general managers. The firm may be portrayed unfavorably in the trade press, and be disparaged by rival salesmen. In addition, the CAFC has repeatedly emphasized a patentee's "affirmative duty to exercise due care to determine whether or not he is infringing" other firms' patents [Underwater Devices v. Morrison-Knudsen, 717 F.2d 1380, 1389; 219 U.S.P.Q. (BNA) 569 (Fed. Cir. 1983)]. If the infringement is held to be willful, the firm risks being assessed trebled damages, opponent's legal fees, and court costs. Its officers and directors may also be held individually liable.

Event studies can provide one indication of the total costs of litigation. Bhagat, Brickley, and Coles [1994] examine the market reaction to the filing of 20 patent infringement lawsuits between 1981 and 1983 where (i) the filing was reported in the Wall Street Journal, (ii) there is only one plaintiff and defendant, and (iii) both the plaintiff's and defendant's stock returns are included in the Center for Research in Security Prices' (CRSP) Daily Returns File. In the two-day window ending on the day the story appears in the Journal, the combined market-adjusted value of the firms fell by an average of -3.1% (significant at the one percent confidence level).

These developments have sparked concern that the pattern of costly litigation--or payments to forestall litigation--are leading to reductions or distortions in innovative investments, particularly for small firms. [Several examples are discussed in Rutter, 1993 and Chu, 1992. These concerns also appear to have been an important motivation for the Department of Justice's emphasis on "innovation markets" in their proposed new intellectual property guidelines.] This anecdotal evidence is supported by more formal examinations. Lanjouw [1994] uses European patent renewal data and a model of patenting behavior to estimate how litigation affects the pace of innovation. Her simulations suggest that a doubling of legal costs will lead about a 30% reduction in the value of the average patent. Claims regarding differences with firm size are corroborated by a 1990 survey of 376 firms [Koen, 1992]. This survey found that the time and expense of intellectual property litigation was a major factor in the decision to pursue an innovation for 55% of the enterprises with under 500 employees, but was a major concern for only 33% of larger businesses. In general, small firms believed that their patents were infringed more frequently, but were considerably less likely to litigate these infringements. Lerner [1995] showed that concerns about patenting near active and experienced litigators also appeared to be shaping small bitoechnology's decisions about where to pursue research activities.