Patent Litigation in the U.S. Semiconductor Industry1

Rosemarie Ham Ziedonis

University of Michigan Business School

INTRODUCTION

Firms in many industries utilize and build on the innovations of others, often in the face of short product life cycles. Recognizing this, scholars and industry representatives alike have started to question whether changes in the U.S. patent system over the past two decades are, in effect, hindering rather than promoting this cumulative process of innovation. Record numbers of patents are issuing from the U.S. Patent and Trademark Office (USPTO) in areas ranging from semiconductors and computer software to business methods and human gene sequences, raising concerns about the costs and feasibility of navigating through mazes of overlapping patent rights in these areas (Shapiro, 2001; Heller and Eisenberg, 1998). At the same time, the past two decades have witnessed a noticeable rise in patent litigation in the United States (Merz and Pace, 1994; Moore, 2000) as well as an escalation in the costs associated with enforcing patent rights in court (Ellis, 1999; AIPLA, 1999). Calling for reform, some have started to question whether the direct and indirect costs associated with obtaining and enforcing U.S. patent rights have started to outweigh the benefits provided by this system (Barton, 2000; Pooley, 2000; Mazzoleni and Nelson, 1998).

This chapter aims to shed additional light on the operation of the U.S. patent system by tracing the incidence and nature of patent-related legal disputes over the past three decades in one important cumulative technological setting—semiconductors. Much like software or computer firms, semiconductor firms typically require access to a “thicket” of external intellectual property to advance technol-

1  

This study received financial support from the GE Fund of the Wharton School’s Reginald H. Jones Center for Management Policy, Strategy, and Organization. I gratefully acknowledge exceptional research assistance provided by Leslie Schafer, Yelena Slutskaya, and Owen Smith of Wharton. I also thank Jim Bessen, Wesley Cohen, Judge T. S. Ellis, III, Bronwyn Hall, Robert Merges, Stephen Merrill, Kimberly Moore, David Mowery, Cecil Quillen, Leslie Schafer, Deepak Somaya, Jim Walsh, Arvids Ziedonis, and participants in the October 2001 STEP Board conference for helpful comments and suggestions. All remaining errors and omissions are of course my own.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 180
Patents in the Knowledge-Based Economy Patent Litigation in the U.S. Semiconductor Industry1 Rosemarie Ham Ziedonis University of Michigan Business School INTRODUCTION Firms in many industries utilize and build on the innovations of others, often in the face of short product life cycles. Recognizing this, scholars and industry representatives alike have started to question whether changes in the U.S. patent system over the past two decades are, in effect, hindering rather than promoting this cumulative process of innovation. Record numbers of patents are issuing from the U.S. Patent and Trademark Office (USPTO) in areas ranging from semiconductors and computer software to business methods and human gene sequences, raising concerns about the costs and feasibility of navigating through mazes of overlapping patent rights in these areas (Shapiro, 2001; Heller and Eisenberg, 1998). At the same time, the past two decades have witnessed a noticeable rise in patent litigation in the United States (Merz and Pace, 1994; Moore, 2000) as well as an escalation in the costs associated with enforcing patent rights in court (Ellis, 1999; AIPLA, 1999). Calling for reform, some have started to question whether the direct and indirect costs associated with obtaining and enforcing U.S. patent rights have started to outweigh the benefits provided by this system (Barton, 2000; Pooley, 2000; Mazzoleni and Nelson, 1998). This chapter aims to shed additional light on the operation of the U.S. patent system by tracing the incidence and nature of patent-related legal disputes over the past three decades in one important cumulative technological setting—semiconductors. Much like software or computer firms, semiconductor firms typically require access to a “thicket” of external intellectual property to advance technol- 1   This study received financial support from the GE Fund of the Wharton School’s Reginald H. Jones Center for Management Policy, Strategy, and Organization. I gratefully acknowledge exceptional research assistance provided by Leslie Schafer, Yelena Slutskaya, and Owen Smith of Wharton. I also thank Jim Bessen, Wesley Cohen, Judge T. S. Ellis, III, Bronwyn Hall, Robert Merges, Stephen Merrill, Kimberly Moore, David Mowery, Cecil Quillen, Leslie Schafer, Deepak Somaya, Jim Walsh, Arvids Ziedonis, and participants in the October 2001 STEP Board conference for helpful comments and suggestions. All remaining errors and omissions are of course my own.

OCR for page 180
Patents in the Knowledge-Based Economy ogy or to legally manufacture and sell their products. In contrast to software, business methods, or biomedical inventions, however, innovation in semiconductors was already highly cumulative and subject to patent protection prior to the 1980s “pro-patent” shift in the United States.2 For example, over 20,000 U.S. patents had been issued on inventions pertaining to semiconductor devices and manufacturing processes by 1981 (USPTO, 1995). In contrast, few software or biotechnology-related patents had been awarded before 1980 in part because of the legal uncertainty over patentable subject matter in these emerging areas (see Graham and Mowery, 2003 on software; Merges, 1997 on biotechnology-related inventions). The extent to which changes in the U.S. patent landscape during the 1980s have altered patterns of cooperation and conflict over patented technologies in semiconductors remains unclear. The semiconductor industry is also an important empirical context within which to examine the broader incentives generated by the patent system in cumulative technological settings. In surveys on appropriability conducted in 1983 and 1994, (the “Yale” and “Carnegie Mellon” surveys, respectively), R&D managers in semiconductors consistently report that patents are among the least effective mechanisms for appropriating returns to R&D investments (Levin et al., 1987; Cohen et al., 2000).3 Driven by a rapid pace of technological change and short product life cycles, semiconductor manufacturers tend to rely more heavily on lead time, secrecy, and manufacturing or design capabilities than patents to recoup investments in R&D. However, in a recent study on patenting in semiconductors, Hall and Ziedonis (2001) find that the strengthening of patent protection in the United States in the 1980s had two divergent effects on dedicated U.S. semiconductor firms. On the 2   Throughout this chapter, the term “pro-patent” refers to a series of legal reforms and rulings discussed in the second section that tilted the judicial treatment of patents in the United States more in favor of the patentee (see Merges, 1997 and Jaffe, 2000 for a review of related studies and empirical evidence). It is important to point out, however, that this term does not imply that the patent regime was “strengthened” in the sense of awarding patents more selectively or ensuring that only the rights of “stronger” patents are upheld. In fact, Quillen and Webster (2001) find that the USPTO has screened out a remarkably low percentage of patent applications since the early 1980s (as little as 5-10 percent). Others emphasize that the Federal Circuit’s interpretation of the nonobviousness standard has effectively “lowered the bar” of patentability (see, e.g., Quillen, 1993 and Hunt, 1999) and, in doing so, has generated additional uncertainty in the enforcement process (as discussed by Lunney, 2001). 3   The 1994 Carnegie Mellon Survey on Industrial R&D in the U.S. Manufacturing Sector (Cohen et al., 2000) updated and extended the influential “Yale” survey conducted in 1983 (Levin et al., 1987). Respondents in both surveys were R&D lab managers in a variety of “focus industries.” Both surveys found that R&D managers in only a handful of industries, including pharmaceuticals, chemicals, and (more recently) biotechnology and medical devices, considered patents to be an effective mechanism by which to appropriate the returns to R&D. These results echo the findings of Scherer et al. (1959), Taylor and Silberston (1973), and Mansfield (1986). As discussed below, the Carnegie Mellon survey extends upon the Yale survey by asking questions on why firms seek patent protection.

OCR for page 180
Patents in the Knowledge-Based Economy one hand, the “pro-patent” shift induced capital-intensive firms to “ramp up” their portfolios of patents more aggressively—largely to deter threats of litigation and to improve their bargaining positions in negotiations with external patent owners.4 On the other hand, the strengthening of U.S. patent rights also appeared to facilitate entry into the industry by firms specializing in chip design. Interviews with representatives from design firms suggest that these firms (often relatively small in size) enforce their patent rights quite aggressively in court vis-à-vis direct rivals, primarily to establish proprietary rights in niche product markets.5 If this is true, any apparent increase in patent litigation within this sector may simply reflect the emergence of these specialized firms and their reliance on U.S. courts to bar use of their intellectual assets. Combined, these findings underscore the importance of considering the multifaceted effects of the U.S. patent system even among firms within an industry. With this in mind, this study seeks to address several basic empirical questions. How do the characteristics of semiconductor firms involved in legal patent disputes over the past three decades compare with those of nonlitigating semiconductor firms? To what extent have firms in this sector been involved in more legal disputes over intellectual property during the so-called “pro-patent” era? Is patent litigation in this industry still fairly “uncommon” as sometimes claimed? Finally, do semiconductor design firms and manufacturers differ in their propensity to enforce patents or in the characteristics of their patent-related legal disputes? To address these questions, the study examines the characteristics of patent cases filed in U.S. District Courts and the U.S. International Trade Commission (USITC) from January 1, 1973, through June 30, 2001, that involve 136 dedicated U.S. semiconductor firms as plaintiff, defendant, or owner of a litigated patent. Firms in the sample include the universe of publicly traded U.S. firms during 1973-2000 that either (a) list semiconductors and related devices (SIC3674) as their primary line of business or (b) were identified by industry sources as dedicated U.S. semiconductor firms. In 2000, sample firms collectively generated over $88 billion in revenues, spent $12 billion in R&D, and had been awarded roughly 31,000 U.S. patents. An unfortunate weakness of this approach is the exclusion from the sample of large U.S. “systems” manufacturers (e.g., IBM, 4   Cohen et al. (2000) report similar findings based on responses to the Carnegie Mellon survey. In industries characterized by “cumulative” (or “complex”) innovation, respondents consistently reported that prevention of lawsuits and blocking of patenting by others were among the most important reasons for patenting. 5   This information is based on a series of structured interviews conducted in 1998 with intellectual property managers and executives from seven U.S. semiconductor firms (three specialized design firms and four dedicated manufacturers). Although this small sample of firms is not representative of the industry as a whole, consistent views emerged in roughly 30 informal interviews with outside legal counsel and others involved in managing and evaluating intellectual property within this industry (see Ziedonis, 2000).

OCR for page 180
Patents in the Knowledge-Based Economy AT&T, or Motorola) and non-U.S. firms (e.g., Mitsubishi, Samsung, or Siemens). Although these firms are important patent owners and users of semiconductor technologies, it is not possible to isolate the share of R&D expenditures directed toward semiconductor technologies for these diversified firms. This approach offers several methodological advantages. First, it enables the identification of a fairly large sample of firms whose R&D expenditures are primarily directed toward semiconductor-related innovation, regardless of whether the firms are involved in legal disputes over patents. By focusing on the patent acquisition and enforcement histories at the level of individual firms, it is possible to examine changes in the propensity of firms to enforce their own patents while also observing changes in the propensity of firms of different sizes and types to encounter patent lawsuits initiated by others. The sample also includes a mix of U.S. semiconductor manufacturers and design firms, including 81 “manufacturers” (i.e., firms like Intel, Texas Instruments, and Micron Technologies, which design and manufacture the majority of their products in-house) and 55 “design” firms (i.e., firms like Altera, Xilinx, and SonicBlue, which specialize in chip design but contract out the manufacture of products to third parties).6 Even though most of the design firms in the sample commercialize and sell products of their own, they are typically much smaller in size (in terms of number of employees or sales revenues) than semiconductor manufacturers in the sample and they invest more heavily in R&D. The results of this study may therefore help inform the underlying factors driving patterns of litigation involving small firms—at least in this sector. Finally, comparing patent litigation and patent issuance trends yields “litigation rates” that are somewhat difficult to interpret in the context of the semiconductor industry. Typically, patent litigation rates are calculated by denominating the number of filed patent cases with the number of patents “at risk” for litigation (Lerner, 1995; Lanjouw and Schankerman, 2001, 2003; Somaya, 2003). Yet, as mentioned above, the decision to patent for many semiconductor firms is, in fact, driven by a desire to deter litigation (Hall and Ziedonis, 2001; Cohen et al., 2000). This methodology enables me to offer a different perspective by calculating liti- 6   Most of the design (or “fabless”) firms in the sample commercialize products based on their designs (e.g., Altera and Xilinx in programmable logic devices). Toward the end of the sample period, so-called “chipless” firms entered the industry that specialize in chip design but license out their designs for others to embed into end products (see Arora et al., 2001; Linden and Somaya, 2000). Rambus, a company specializing in interconnection technologies that speed communications between memory chips and microprocessors, is a prominent example of such a company. Because these “chipless” firms are a recent phenomenon and represent less than five percent of the design firms in the sample, the litigation trends and practices discussed in this chapter refer primarily to those involving the former category of design firms (i.e., those that compete directly in semiconductor product markets). Distinguishing between the patent acquisition and enforcement strategies of “traditional” design firms and those of “chipless” firms is an interesting topic for future research.

OCR for page 180
Patents in the Knowledge-Based Economy gation rates in the industry (overall, and for manufacturers and design firms separately) using (a) firm-level patenting activity and (b) firm-level R&D expenditures. Detailed information about all reported patent cases involving one or more of the 136 semiconductor firms in the sample is merged with information about the patents and other parties involved in the disputes. Several empirical patterns emerge, which are summarized as follows: Of the 136 U.S. semiconductor firms in the sample, roughly 56 percent were involved in at least one reported patent case filed in U.S. District Courts and the USITC between January 1, 1973, and June 30, 2001. On average, sample firms involved in patent lawsuits spent more on R&D (in absolute terms and per employee), were larger (in terms of sales or number of employees), and owned more patents than “peer” semiconductor firms not involved in patent litigation during this period. The number of annual cases filed that involve these firms (as a group) increased sharply around the mid-1980s and continued at a higher level throughout the 1990s. This trend is not remarkable when compared to the overall growth in U.S. patent litigation during this period documented elsewhere (Merz and Pace, 1994; Moore, 2000). It is consistent, however, with popular reports that legal disputes over intellectual property have become more common in semiconductors—despite the widespread use of cross-licenses in this industry. Relative to annual R&D spending by these firms, the patent litigation rate in semiconductors rose considerably during 1986-2000 from that in the preceding decade (by as much as 93 percent). In contrast, the number of cases filed per 1,000 patents awarded to these firms (a more common metric used to estimate litigation rates) exhibited a slight decline between the two periods. The apparent decline in litigated patents per patents awarded during the latter period is driven, however, by the dramatic rise in patenting by semiconductor firms since the mid-1980s (as reported in Hall and Ziedonis, 2001). Indeed, updating the trends reported in Hall and Ziedonis (2001) reveals that the “patent portfolio races” of U.S. semiconductor manufacturers continued to accelerate through the end of the 1990s both in absolute terms and relative to firm-level R&D spending. Regardless of how it is measured, the average litigation rate of specialized design firms in the sample is high and is more than twice that of manufacturers in the sample. Manufacturers, on average, are involved in disputes with a more disparate set of parties and tend to enforce patents that are almost 4 years older than the average patent in their portfolios. In contrast, design firms typically enforce their patents against other design firms and litigate over patents that are roughly the same age as the average patent in their portfolios. In addition to these general trends, it was also interesting to observe what appears to be an active “market” for intellectual property that predates the filing

OCR for page 180
Patents in the Knowledge-Based Economy of a patent lawsuit. In at least 30 percent of identified cases, legal title to a litigated patent had been reassigned from the original inventor (or assignee) to one of the litigating parties—typically, to the plaintiff in an infringement suit. Some of these disputes involved plaintiffs that had acquired the intellectual property as part of a broader acquisition of a firm or its physical assets (e.g., SGS Thomson successfully enforced Mostek’s memory chip patents after acquiring the company in the mid-1980s; similarly, Atmel enforced patents awarded to Seeq Technologies after acquiring Seeq’s intellectual and physical assets pertaining to non-volatile memory). In other cases, the plaintiff appeared to be using externally generated patents in reciprocal suits. For example, after failing to reach agreement on the terms of a renewed cross-license agreement, Hyundai sued Texas Instruments in 1991 for infringing five patents—four of which Hyundai had purchased from outside inventors. There also was an apparent rise in infringement suits brought by specialized “patent licensing” companies.7 On one hand, the emergence of specialized patent management and enforcement companies may help “tilt the table” more in favor of independent inventors or patentees from small businesses (see Lerner, 1995; Lanjouw and Lerner, 2001) or may represent the continued development of markets for technology (Arora et al., 2001). On the other hand, others raise concerns that an increased trade in and enforcement of so-called “paper patents” (i.e., “blocking” patents owned by inventors or companies that do not compete in the related product markets) is imposing an implicit tax on innovation (Pooley, 2000). These are interesting issues that warrant future investigation. Before turning to the rest of the chapter, it is important to acknowledge two inherent limitations of this research. First, like any study of litigation events, this study is inherently limited by its examination of the proverbial “iceberg’s tip.”8 7   Some of these firms, for example, General Patent Corporation (or its affiliated IP Holdings LLC), specialize in the management and enforcement of patents on a contingency fee basis. Arora et al. (2001) refer to such companies as “technology intermediaries” and discuss their services in more detail (p. 84-85). 8   Because most companies treat information about licensing negotiations and agreements as highly confidential information, it is unusual to observe directly how the “iceberg’s tip” (in this case, patent-related disputes that involve the filing of a lawsuit) compares with the underlying set of disputes that are settled privately and in the absence of a case being filed with the courts. To my knowledge, the most direct evidence on this point is from a 1994 survey of patent and licensing practices of British and Japanese firms that asked intellectual property managers to estimate the number of complaints of infringement made against the company and to indicate how the complaints were resolved (Pitkethly, 1996). Overall, British managers estimated that 87 percent of the disputes they encounter over alleged patent infringement are settled privately—without a lawsuit being filed. Similarly, Japanese managers estimated that they resolved around 80 percent of infringement complaints either by simple notification or by private negotiations (as reported in Pitkethly, 1996, data table 57; based on responses from 50 British and 120 Japanese managers). More recently and in the context of the United States, Lanjouw and Schankerman (2001) examine this issue indirectly by comparing the characteristics of litigated and nonlitigated patents in the United States. As discussed below, they find that most patents issued

OCR for page 180
Patents in the Knowledge-Based Economy As discussed above, semiconductor firms have long licensed and cross-licensed their intellectual property and private settlement over intellectual property rights is still the rule rather than the exception (Grindley and Teece, 1997). Although future versions of this research will attempt to control more explicitly for this selection bias, this study simply summarizes the characteristics of observable case filings. A second limitation of this study is its primary reliance on data that may underestimate the number of patent cases filed in U.S. District Courts, particularly during the early period of the study (i.e., before 1984). As discussed in the third section of this chapter, several attempts were made to address this shortcoming in the data. Nonetheless, underreporting in the early period may still exist, and the results should be interpreted with this in mind. The remainder of this chapter is organized as follows. The second section summarizes what is often referred to as the “pro-patent” shift in the U.S. legal environment during the 1980s and discusses its effects on the use of patents in the semiconductor industry. The third section presents the data and methodology used to trace patent litigation involving U.S. semiconductor firms during 1973-2001. The descriptive findings are summarized in the fourth section, and concluding remarks follow. THE CHANGING PATENT LANDSCAPE The patent system has long been recognized as an important policy instrument used to promote innovation and technological progress. Two fundamental mechanisms underpin the patent system. First, an inventor discloses to the public a “novel,” “useful,” and “nonobvious” invention. In return, the inventor receives the right to exclude others from using that patented invention for a fixed period of time (now 20 years from the date of patent application in the United States). The rules of the patent game may differ from country to country (e.g., whether rights are assigned to the first inventor or the first to file the patent application), but the underlying principle remains the same. By providing exclusionary rights for some period of time and a more conducive environment in which to recoup R&D investments, the patent system aims to encourage inventors to direct more of their resources toward R&D than would otherwise be the case. At the same time, detailed information about the invention is disclosed to the public when the patent application is published.     in the United States are never involved in litigated disputes; on average, however, litigated patents tend to be more valuable than nonlitigated patents and are more likely to form the basis for a sequence of innovations by the patentee. Lanjouw and Schankerman (2001, 2003) discuss these selection issues in greater detail. See also Siegelman and Donohue (1990) and Siegelman and Waldfogel (1999) on the selection biases inherent in studies of litigation events more generally.

OCR for page 180
Patents in the Knowledge-Based Economy The So-Called Pro-Patent Shift in the United States The creation of the Court of Appeals for the Federal Circuit (CAFC) in 1982 is often credited with ushering in an era that reversed the judicial treatment of patent rights in the United States from the preceding decades in ways that favored patent owners.9 From the trust-busting era of the 1930s through much of the 1970s, patents were largely viewed as anticompetitive weapons used to stifle competition.10 For example, in 1959, Scherer and his co-authors report: During the past two decades a pronounced change has taken place in the policies of governmental bodies towards patents owned by corporations…. The courts have become increasingly critical of patent validity, and cases in which the exercise of patent rights conflicted with antitrust statutes have been prosecuted by denying the exclusiveness of the patent grant. Since 1941, more than 100 judgments have been entered which required corporations to license their patents to all applicants at reasonable royalties or no royalties at all. This trend was brought sharply to the public’s attention in January of 1956 when two of the nation’s foremost leaders in industrial technology, the American Telephone and Telegraph Co. and International Business Machines, Inc., entered into decrees requiring them to license all of their more than 9,000 patents, in most cases without receiving royalties in return.” (Scherer et al., 1959, p. 2-3) Antipatent sentiment continued through much of the 1970s. As Merges (1997) states: “It was difficult to get a patent upheld in many federal circuit courts, and the circuits diverged widely both as to the doctrine and basic attitudes toward patents. As a consequence, industry downplayed the significance of patents” (p. 12). By the early 1980s, the pendulum started to swing away from a restrictive treatment of patents toward a view that patent rights should be construed liberally to stimulate innovation. Driven by general concerns about increased international competition in several key industries—including semiconductors—and a growing belief that stronger intellectual property rights were needed to stimulate innovation, Congress passed a series of laws in the early 1980s aimed at improving the function of the U.S. patent system and at relaxing antitrust constraints on the 9   Although governmental agencies (in the United States, the USPTO) examine applications and decide whether an invention qualifies for patent protection, the courts ultimately determine the strength of patent rights once granted. By deciding whether a patent is valid or whether another party has infringed on the patent owner’s rights, courts play a pivotal role in determining the strength (and hence, the value) of patent rights. 10   As Merges (1997) states: “Unfortunately for the patent system, the identification of patents with big business meant that when big business lost favor, so too would patents…. The exclusive nature of the patent grant, coupled with the actual market power that some patents conferred on their holders, seemed closely related to many of the monopolists’ oppressive practices” (p. 11).

OCR for page 180
Patents in the Knowledge-Based Economy collaborative R&D activities of firms.11 Unique to semiconductors, the 1984 Semiconductor Chip Protection Act (SCPA) also conferred protection against theft of the “mask works,” the overall layout of the chip designs (see Samuelson and Schotchmer, 2001 for a recent review of this form of protection). No other event signaled the shift toward stronger legal protection for patents in the United States than the 1982 creation of the CAFC, a centralized appellate court with jurisdiction over all patent infringement appeals (Jaffe, 2000).12 Although the driving force behind the legal reform was a need to unify U.S. patent doctrine, the Federal Circuit put in place a number of procedural and substantive rules that collectively favored patent owners. For example, the new court endorsed the broad, exclusionary rights of patent owners through its interpretation of patent scope, increased evidentiary standards to make it more difficult to invalidate the rights of patent owners, was more willing to halt allegedly infringing actions early in the dispute process by granting preliminary injunctions, and was more willing to sustain large damage awards and thereby penalize infringing parties more severely.13 The plaintiff success rates in patent infringement suits also increased substantially during this period (Lerner, 1995). Although the CAFC was created in 1982 and issued a flurry of written opinions during 1983 (Adelman 1987; Nies 1993), the impact of the CAFC on the favorable legal treatment of patent rights in U.S. courts was not widely publicized until the mid-1980s.14 The “surprising new power of patents” was perhaps most clearly revealed by Polaroid’s success in a longstanding lawsuit against Kodak for infringing certain instant photography patents awarded to Polaroid. In a 1985 11   For example, the 1984 National Cooperative Research Act reduced the antitrust penalties for collaboration among firms in “pre-commercial” research, which paved the way for the subsequent formation of SEMATECH, a large ongoing research consortium in the semiconductor industry. See also Kortum and Lerner (1998) on the legislative initiatives aimed at improving the operation of the USPTO: “More patent legislation was enacted between 1980 and 1982 than had passed in the previous two decades” (p. 7). 12   Until 1982, patent appeals were primarily heard in the court of appeals of the district in which the case was tried, which led to “forum shopping” among firms (Kortum and Lerner, 1998). Adelman (1987), however, argues that the 1982 establishment of the new court represented an “outgrowth of the dissatisfaction with the functioning of both the Supreme Court and the federal appellate courts” and a “realization by Congress that a uniform and more reliable patent system was necessary for sustained economic growth and to rise to the challenge of Japanese and German industrial competition” (p. 983). 13   Merges (1997) and Lanjouw and Lerner (2001) provide additional information on these and other effects of the court. The main point, however, is that many of the rules and decisions of the CAFC during this period favored the rights of patent owners. 14   A series of articles surfaced in the popular press during 1985-1986 that proclaimed the “new” legal environment for patent owners. See, for example, “A Change in the Legal Climate,” Forbes, Oct. 7, 1985, p. 41; “A Weapon at Last [pro-patent decisions],” Forbes, Mar. 10, 1986, p. 46; and “The Surprising New Power of Patents,” Fortune, June 23, 1986, p. 57.

OCR for page 180
Patents in the Knowledge-Based Economy ruling, Kodak was required to pay Polaroid almost $1 billion in damages and interest, was barred from manufacturing and selling instant cameras, and was forced to close its instant camera production line (Warshofsky, 1994). As reported by Hall and Ziedonis (2001), representatives from the semiconductor industry emphasized the importance of this case, along with Texas Instruments’ successful patent infringement cases against Japanese and Korean firms during 1985-1986, in demonstrating the “new power of patents.” Not only did TI and other large patent owners such as IBM and Motorola increase the price (i.e., royalty rates) charged for “rights to use” their patents, but the increased value associated with patents may have induced entry into the patent licensing business as the licensure of patents became more profitable under the new regime (as discussed below). The Evolving Role of Patents in Semiconductors Not surprisingly, the use and importance of U.S. patents in semiconductors was affected by this changing patent landscape, albeit in some unanticipated ways. By the early 1980s, a broad range of semiconductor technologies, including methods for manufacturing semiconductors and integrated circuit design, had diffused widely across the industry (Levin, 1982). The “technological giants” in semiconductors, including AT&T and IBM, were effectively curtailed from aggressively enforcing their patent rights against rival firms (either merchant manufacturers or other users of semiconductor technologies) from the 1950s to the late 1970s by the antitrust constraints discussed above. As a result of its 1956 consent decree with U.S. antitrust authorities, for example, AT&T had licensed its semiconductor inventions widely to other firms in return for access to subsequent inventions by licensees. AT&T’s active role in licensing and disseminating semiconductor technologies is credited with stimulating the early growth of the U.S. merchant semiconductor industry (Tilton, 1971; Grindley and Teece, 1997; Mowery and Rosenberg, 1998). Nonetheless, Tilton (1971, p. 76) concludes: Certainly the great probability that other firms were going to use the new technology with or without licenses is another reason for the liberal licensing policy. Secrecy is difficult to maintain in the semiconductor field because of the great mobility of scientists and engineers and their desire to publish. Moreover, semiconductor firms, particularly the new, small ones, have demonstrated over and over again their disposition to infringe on patents. The prospect of lengthy and costly litigation in which its patents might be overturned could not have been very attractive to AT&T. Similarly, Von Hippel (1988) emphasizes that the semiconductor field was a very fast-moving one that, even by the early 1980s, contained many unexpired patents with closely related subject matter and claims. He writes: Since patents challenged in court are unlikely to be held valid, the result of high likelihood of infringement accompanying use of one’s own patented—or unpat-

OCR for page 180
Patents in the Knowledge-Based Economy ented—technology is not paralysis of the field. Rather, firms in most instances simply ignore the possibility that their activities might be infringing the patents of others. The result is what Taylor and Silberston’s interviewees in the electronics components field termed ‘a jungle’ and what one of my interviewees termed a ‘Mexican standoff’…. The usual result is cross-licensing, with a modest fee possibly being paid by one side or the other. (p. 52-53)15 Two other factors contributed to the infrequent patent litigation and widespread cross-licensing that have historically characterized the semiconductor industry. First, as reported in the introduction to this chapter, semiconductor firms tend to rely on mechanisms other than patents to recoup their R&D investments, including being first to market and safeguarding the “know-how” (often through secrecy) required to manufacture commercially viable chips (Levin et al., 1987; Cohen et al., 2000). Indeed, there is little evidence that the strengthening of U.S. patent rights boosted aggregate R&D spending by firms in this sector (Hall and Ziedonis, 2001; Bessen and Maskin, 2000). Second, to reduce the risk of disruptions in supply, large customers of chips (e.g., IBM and the U.S. government) typically required dedicated manufacturers to transfer to a competing supplier the know-how and patent rights required to manufacture a compatible product (Shepard, 1987). These second source agreements further promoted cross-licensing in the industry but declined in use over the decade of the 1980s as the industry matured and built up capacity (Grindley and Teece, 1997). Although cross-licensing continues to be an important mechanism by which firms trade access to one another’s patents, the terms of these agreements appear to have changed (not surprisingly) as the rights of patent owners have grown stronger. Firms with large patent portfolios, such as Texas Instruments, IBM, AT&T, and Motorola, adopted a more aggressive licensing and litigation strategy to profit directly from their patent portfolios—both by seeking licenses from a larger number of firms and by increasing royalty rates on use of their inventions. For example, in the early 1980s, Texas Instruments launched a more aggressive patent licensing program—initially against Japanese and Korean competitors in the market for memory chips. During 1986-1993, TI earned almost $2 billion from licensing rights to its semiconductor patents (Grindley and Teece, 1997). Similarly, IBM increased its royalty rates around 1987-1988 from 1 percent of sales revenues for products using IBM patents up to a range of 1 to 5 percent (Shinal, 1988). By 2000, IBM earned over $1.5 billion in income from licensing its intellectual property portfolio, up from $646 million in 1995.16 According to industry representatives as well as accounts in the general business press (e.g., Warshofsky, 1994; Rivette and Kline, 2000), the increased value associated with 15   Similarly, writing in 1987, Levin et al. report: “In the semiconductor industry … the cumulative nature of technology makes it difficult to participate legally without access to the patents of numerous firms. In consequence, there is widespread cross-licensing” (p. 798, fn 29). 16   As reported in IBM annual reports, available at www.ibm.com.

OCR for page 180
Patents in the Knowledge-Based Economy TABLE 4B Profile of Cases Involving Sample Design Firms, by Type of Case     Infringement Suits Declaratory Judgment Suits Panel B. Cases Involving Design Firms Total (Includes 3rd party) As Plaintiff As Defendant As Plaintiff As Defendant Overview Number of casesa 90 43 48 5 4 Number of unique parties involved 92 57 47 10 4 Average number of parties/case (median) 2.41 (2) 2.30 (2) 2.27 (2) 3 (2) 2.75 (3) Number of patents involved 134 60 80 19 6 Average number of patents/case (median) 2.19 (1) 1.91 (1) 2.12 (1) 4.8 (4) 3.25 (3.5) By characteristics of opposing party:b Percent cases “within sample” Percent with other US semiconductor manufacturers 27.78% 9.30% 39.58% 20.00% 0.00% Percent with US Design firms 18.89% 37.21% 33.33% 0.00% 0.00% Percent cases with foreign firms 11.1% 13.95% 6.25% 0.00% 0.00% Percent cases with non-semiconductor firmsc 43.3% 34.88% 25.00% 60.00% 100.00% Percent cases with independent inventors 2.2% 0.00% 0.00% 20.00% 0.00% Percent cases with univs or govt labs 0.0% 0.00% 0.00% 0.00% 0.00% By characteristics of litigated patents: Percent in electronics-related classes (G01-G21; H- ) 98.51% 100.00% 98.75% 100.00% 100.00% Percent that pertain to new or improved semiconductor devices (Invention Type=1)d 79.9% 93.33% 82.50% 57.89% 100.00% Percent that pertain to new or improved manufacturing processes (Invention Type=3)d 6.77% 0.00% 8.75% 5.26% 0.00% aOverall, manufacturers were involved in 209 cases and design firms were involved in 90 cases during the sample period. This exceeds the number of cases in the sample (287) because of 12 cases that involve both manufacturers and design firms in the sample. bPercentages exceed 100% since multiple parties (of different types) can be involved in a case. cDefined as firms for which SIC3674 is not listed as a primary or secondary class among its lines of business. dSee Appendix A for information about invention types and how they were coded.

OCR for page 180
Patents in the Knowledge-Based Economy of infringement (defendants in infringement suits), or are engaged in declaratory judgment suits. Because of the limited number of declaratory judgment suits in the sample, the discussion below focuses on the trends reported in the overall and patent infringement columns. In general, we see from Tables 4A and 4B that cases involving manufacturers tend to include a more disparate set of parties and inventions than is true of disputes involving design firms. Almost 25 percent of cases involving manufacturers are against foreign firms, and almost 10 percent of the cases include an independent inventor as an opposing party. In contrast, less than 12 percent of cases involving design firms are against non-U.S. firms, and only 2 percent are in opposition with an independent inventor. Similarly, opposing parties in disputes involving design firms are more heavily concentrated among other design firms or U.S. semiconductor manufacturers. Design firm disputes also involve a more focused set of technologies more targeted toward product-related inventions (as revealed in the lower rows of Tables 4A and 4B). Many of these trends are, of course, not surprising given the broader range of technological and commercial activities in which manufacturers are involved. Table 4B also reveals, however, that design firms (on average) tend to enforce their patent rights most frequently against other design firm rivals. Interestingly, they most commonly defend themselves in litigation initiated by domestic semiconductor manufacturers, followed by disputes initiated by other design firms. To the extent that design firms are using the courts to protect market share (as suggested in interviews discussed in the second section of this chapter), we should expect them to enforce their patents relatively early in the patent’s lifetime to block competition in related markets. Consistent with this view, I find that design firms enforce patents that are roughly the same age as the average patent in their portfolios. In contrast, manufacturers in the sample enforce patents that are, on average, almost 4 years older than the average patent in their portfolios.35 Somaya (2003) argues that the “strategic stakes” are higher for patents enforced early in their lifetimes. In a study comparing case filings and settlements in computers and research medicines, Somaya finds that the patentee’s strategic stakes renders settlement less likely in both sectors. My descriptive results are consistent with this finding, albeit in a different setting. CONCLUDING REMARKS This chapter examines the enforcement of U.S. patents in semiconductors— an industry characterized by a rapid, cumulative process of innovation. Starting with a sample of 136 dedicated U.S. semiconductor firms, the study compares the 35   More specifically, patents enforced by design firms were (on average) issued within 1 year of the average patent in the litigating design firm’s portfolio. These results do not appear to be driven by outliers.

OCR for page 180
Patents in the Knowledge-Based Economy characteristics of litigating and nonlitigating firms and explores the incidence and nature of litigation events involving these firms from 1973 through June 30, 2001. Despite active cross-licensing in this industry, the results suggest that litigation events over patented technologies have become more frequent during the period associated with stronger U.S. patent rights. Previous research suggests that the aggressive patenting by manufacturing firms in this sector is driven by a desire to deter such litigation and to negotiate more favorable access to external technologies (Hall and Ziedonis, 2001; Cohen et al., 2000). Indeed, this study finds that the “patent portfolio races” of U.S. semiconductor manufacturers identified by Hall and Ziedonis continued apace throughout the decade of the 1990s, dwarfing overall patent litigation trends in this sector. Although the number of patent cases involving these firms has declined slightly relative to their patenting activity since the mid-1980s, I nonetheless find that it has increased relative to the R&D investments of these firms during the same period. Assuming that the direct and indirect costs associated with litigation have also increased over time (Barton, 2000; AIPLA, 1999), these trends suggest that semiconductor firms have been directing a larger share of their innovation-related resources toward defending, enforcing, and challenging patents in court since the mid-1980s than was true in the preceding period. The descriptive findings yield somewhat mixed results regarding the litigation behavior of small firms—a matter explored in detail in recent studies by Lanjouw and Lerner (2001) and Lanjouw and Schankerman (2003). On one hand, I find that semiconductor firms involved in litigation over the sample period are larger, invest more in R&D, and own more patents than nonlitigating firms in the industry. This result holds both across the sample and within each group of design and manufacturing firms. Yet the high average litigation propensity of design firms in the sample is quite striking. These firms, which typically employ less than 500 employees, enforce an average of 4 out of every 100 patents they own—a litigation rate that is not only high relative to semiconductor manufacturers but closely resembles that of dedicated biotechnology firms in the early 1990s (as reported by Lerner, 1995). As “technology specialists” lacking complementary manufacturing assets of their own, semiconductor design firms appear to rely quite heavily on U.S. courts to protect their intellectual assets—primarily against other design firm rivals. At least within this sector, these results call into question whether the propensity of small firms to enforce patents stems from a desire to aggressively defend technological niches (“high stakes”) or from a lack of large portfolios with which to trade (as Lanjouw and Schankerman’s 2003 study suggests). The next step in this research is to test between these competing explanations econometrically, which will also enable us to investigate a broader range of factors shaping cooperation and conflict over intellectual property rights in this sector. Although suggestive, these results highlight several important (but unresolved) questions about the role of patents, and patent portfolios, in the complex

OCR for page 180
Patents in the Knowledge-Based Economy process of innovation. What is the effect of litigation—and threats of litigation— on the R&D and patenting behavior of firms? Do firms in cumulative technological settings “avoid the shadows” of better-capitalized rivals (or those with larger patent portfolios), as Lerner (1995) finds in the biotechnology sector? Alternatively, if “mutual blocking” conveys value to these firms in terms of more favorable cross-licensing deals or implicit design freedom, firms may seek to “race into” the patent thickets by amassing larger portfolios of their own with which to trade. Addressing these questions would enrich our understanding of the underlying incentives generated by the patent system. Finally, to what extent is the emergence of patent thickets deterring entry or “tilting the tables” more in favor of large firms (or, somewhat separately, firms with large patent portfolios)? This question extends well beyond the scope of this study but is important from a policy perspective and interesting to consider within the context of semiconductors.36 As mentioned in the second section of this chapter, the early success of the U.S. semiconductor industry is often attributed to the liberal licensing terms offered by firms such as AT&T and IBM for rights to use their portfolios of patents in the 1950s through 1970s—in part because of constraints imposed by antitrust authorities (Tilton, 1971; Levin, 1982; Grindley and Teece, 1997). Yet, as demonstrated by the emergence of design firms within this sample, widespread entry continued to characterize this industry during the period associated with stronger U.S. patent rights and a less restrictive antitrust regime. Several recent studies provide partial insights into this apparent paradox. For example, by modeling the decision of patent owners to invest in monitoring potential infringers, Crampes and Langinier (2002) show that strengthening the rights of patent owners may deter entry if an incumbent already had incentives to negotiate licenses with the entrant (to settle rather than sue) but now imposed a higher license fee. Alongside this conventional finding, however, they also find that strengthening the rights of patent owners may (1) expand the “settlement area” under which it is becomes profitable for the patent owner to license rather than sue (and thereby induce entry) and (2) induce entry by firms with highly differentiated products (where the bargaining surplus is greatest). Recent empirical findings by Gans et al. (2002) similarly suggest that, in the shadows of stronger patent protection, the relationships between incumbents and start-up firms may become more cooperative than competitive in nature, whereas others emphasize the importance of patent rights in sustaining entry by specialized firms (e.g., Arora et al., 2001). In summary, these studies highlight the importance of 36   To empirically estimate the effects, if any, of incumbent portfolios on observed patterns of entry, one would need some way of establishing the counterfactual: Absent such portfolios (or given portfolios of smaller size), what pattern of entry would we expect to observe? Studies along these lines would also need to take into account the demand-side factors and technological opportunities that also influence the entry decision (as discussed in Cohen, 1995).

OCR for page 180
Patents in the Knowledge-Based Economy understanding the economic incentives generated by the patent system but also underscore the multifaceted effects that may arise even within one technological sector. REFERENCES Adelman, M. J. (1987). “The New World of Patents Created by the Court of Appeals for the Federal Circuit.” Journal of Law Reform 20(4): 979-1007. American Intellectual Property Law Association (AIPLA). (1999). Report of Economic Survey 1999. Arlington, VA: AIPLA. Arora, A., A. Fosfuri, and A. Gambardella. (2001). Markets for Technology: The Economics of Innovation and Corporate Strategy. Cambridge, MA: MIT Press. Barton, J. H. (2000). “Reforming the Patent System.” Science 287: 1933-1934. Bessen, J., and E. Maskin. (2000). “Sequential Innovation, Patents, and Imitation,” Working Paper No. 00-01, Department of Economics, Massachusetts Institute of Technology. Cohen, W. M. (1995). “Empirical Studies of Innovative Activity,” in P. Stoneman, ed., Handbook of Economics of Innovation and Technological Change. Oxford, UK: Blackwell Publishers Inc. Cohen, W. M., R. R. Nelson, and J. Walsh. (2000). “Protecting Their Intellectual Assets: Appropriability Conditions and Why U.S. Manufacturing Firms Patent (or Not),” NBER Working Paper 7552. Crampes, C., and C. Langinier. (2002). “Litigation and Settlement in Patent Infringement Cases.” RAND Journal of Economics 33(2): 258-274. Ellis, Judge T. S., III. (1999). “Distortion of Patent Economics by Litigation Costs.” In K.M. Hill, T. Takenaka, and K. Takeuchi, eds., Streamlining International Intellectual Property: Enforcement and Prosecution, University Technology Transfer, and Incentives for Inventors. Seattle, Washington: University of Washington School of Law Center for Advanced Study and Research on Intellectual Property, publication series No. 5. Farn, M. (1996). “A Quasi-Statistical Analysis of the Acquisition and Enforcement of Patent Rights in the Computer and Semiconductor Industries for the Period 1986-1995,” Stanford Law School directed research, Fall 1996. Gans, J. S., D. H. Hsu, and S. Stern. (2002). “When Does Start-Up Innovation Spur the Gale of Creative Destruction?” RAND Journal of Economics 33(4): 571-586. Graham, S., and D. C. Mowery. (2003). “Intellectual Property Protection in the U.S. Software Industry.” In W. Cohen and S. Merrill, eds., Patents in the Knowledge-Based Economy.Washington, D.C.: The National Academies Press. Grindley, P. C., and D. J. Teece. (1997). “Managing Intellectual Capital: Licensing and Cross-Licensing in Semiconductors and Electronics.” California Management Review 39(2): 1-34. Hall, B. H., and R. H. Ziedonis. (2001). “The Patent Paradox Revisited: An Empirical Study of Patenting in the US Semiconductor Industry, 1979-95.” RAND Journal of Economics 32(1): 101-128. Heller, M. A., and R. S. Eisenberg. (1998). “Can Patents Deter Innovation? The Anticommons in Biomedical Research.” Science 280: 698-701. Hunt, R. M. (1999). “Nonobviousness and the Incentive to Innovate: An Economic Analysis of Intellectual Property Reform,” working paper no. 99-3, Economic Research Division, Federal Reserve Bank of Philadelphia. Integrated Capital Monitor. (1999). “Company Analysis: Rambus, Inc.” 1(1): 6-7. Integrated Circuit Engineering Corporation (ICE). (1995). Cost Effective IC Manufacturing, 1995. Scottsdale, AZ: Integrated Circuit Engineering Corporation. Integrated Circuit Engineering Corporation (ICE). (1976-1998). Status: A Report on the Integrated Circuit Industry. Scottsdale, AZ: Integrated Circuit Engineering Corporation.

OCR for page 180
Patents in the Knowledge-Based Economy Jaffe, A. (2000). “The U.S. Patent System in Transition: Policy Innovation and the Innovation Process.” Research Policy 29: 531-557. Kortum, S., and J. Lerner. (1998). “Stronger Protection or Technological Revolution: What Is Behind the Recent Surge in Patenting?” Carnegie-Rochester Conference Series on Public Policy 48: 247-304. Lanjouw, J. O., and M. Schankerman. (2001). “Characteristics of Patent Litigation: A Window on Competition.” RAND Journal of Economics 32(1): 129-151. Lanjouw, J. O., and M. Schankerman. (2003). “Enforcement of Patent Rights in the United States.” In W. Cohen and S. Merrill, eds., Patents in the Knowledge-Based Economy. Washington, D.C.: National Academies Press. Lanjouw, J., and J. Lerner. (2001). “Tilting the Table? The Use of Preliminary Injunctions.” Journal of Law and Economics 44(2): 573-603. Lerner, J. (1995). “Patenting in the Shadow of Competitors.” Journal of Law and Economics 38: 563-595. Levin, R. C. (1982). “The Semiconductor Industry.” In Richard R. Nelson, ed., Government and Technical Progress: A Cross-Industry Analysis. Oxford, UK: Pergamon Press. Levin, R. C., A. K. Klevorick, R. R. Nelson, and S. G. Winter. (1987). “Appropriating the Returns from Industrial Research and Development.” Brookings Papers on Economic Activity 3: 783-820. Linden, G., and D. Somaya. (2000). “System-on-a-Chip Integration in the Semiconductor Industry: Industry Structure and Firm Strategies,” Walter A. Haas School of Business working paper, University of California, Berkeley. Lunney, G. S. (2001). “E-Obviousness.” Michigan Telecommunications and Technology Law Review 7(363): 363-422. Macher, J., D. C. Mowery, and D. Hodges. (1998). “Reversal of Fortune? The Recovery of the U.S. Semiconductor Industry.” California Management Review 41(1): 107-136. Mansfield, E. (1986). “Patents and Innovation: An Empirical Study.” Management Science 32(2): 173-181. Mazzoleni, R., and R. R. Nelson. (1998). “Economic Theories About the Benefits and Costs of Patents.” Journal of Economic Issues 32(4): 1031-1052. Merges, R. P. (1997). Patent Law and Policy: Cases and Materials, second edition. Charlottesville, VA: The Mitchie Company. Merz, J. F., and N. M. Pace. (1994). “Trends in Patent Litigation: The Apparent Influence of Strengthened Patents Attributable to the Court of Appeals for the Federal Circuit.” Journal of the Patent and Trademark Office Society 76: 579-590. Moore, K. A. (2000). “Judges, Juries, and Patent Cases—An Empirical Peek Inside the Black Box.” Michigan Law Review 99(281): 365-409. Mowery, D. C., and N. Rosenberg. (1998). Paths of Innovation: Technological Change in 20th-Century America. Cambridge, UK: Cambridge University Press. Mutti, J. (1993). “Intellectual Property Protection in the United States under Section 337.” The World Economy 16: 339-357. Mutti, J., and B. Yeung. (1996). “Section 337 and the Protection of Intellectual Property in the United States: The Complaints and the Impact.” The Review of Economics and Statistics 78: 510-520. Nies, H. W. (1993). “Ten Years of Patent Law Development Under the U.S. Court of Appeals for the Federal Circuit.” IIC 24(6): 797-803. Pitkethly, R. (1996). The Use of Intellectual Property in High Technology Japanese and Western Companies. DPhil Thesis, Oxford University. Pooley, J. (2000). “The Trouble with Patents.”California Lawyer, October. Quillen, C. D. (1993). “Proposal for the Simplification and Reform of the United States Patent System.” American Intellectual Property Law Association Quarterly Journal 21(3): 189-212.

OCR for page 180
Patents in the Knowledge-Based Economy Quillen, C. D., and O. H. Webster. (2001). “Continuing Patent Applications and the Performance of the U.S. Patent Office.” Federal Circuit Bar Journal 11(1): 1-21. Rivette, K. G., and D. Kline. (2000). Rembrandts in the Attic: Unlocking the Hidden Value of Patents. Boston, MA: Harvard University Press. Samuelson, P., and S. Schotchmer. (2001). “The Law and Economics of Reverse Engineering,” Boalt Hall School of Law Working Paper, University of California, Berkeley. Scherer, F. M., S. E. Herzstein, Jr., A. W. Dreyfoos, W. G. Whitney, O. J. Bachmann, C. P. Pesek, C. J. Scott, T. G. Kelly, and J. J. Galvin. (1959). Patents and the Corporation: A Report on Industrial Technology Under Changing Public Policy, second edition. Boston, MA: Harvard University, Graduate School of Business Administration. Shapiro, C. (2001). “Navigating the Patent Thickets: Cross-Licenses, Patent Pools, and Standard-Setting.” In A. B. Jaffe, J. Lerner, and S. Stern, eds., Innovation Policy and the Economy, volume 1. Cambridge, Mass: National Bureau of Economic Research. Shepard, A. (1987). “Licensing to Enhance Demand for New Technologies.” RAND Journal of Economics 18(3): 360-368. Shinal, J. (1988). “IBM Again Tops Technology Rivals for Most Patents.” The San Diego Union-Tribune, 20 January. Siegelman, P., and J. J. Donohue III. (1990). “Studying the Iceberg from Its Tip: A Comparison of Published and Unpublished Employment Discrimination Cases.” Law and Society Review24 (5): 1133-1170 Siegelman, P., and J. Waldfogel. (1999). “Toward a Taxonomy of Disputes: New Evidence Through the Prism of the Priest/Klein Model.” Journal of Legal Studies 28: 101-130. Somaya, D. (2003). “Strategic Determinants of Decisions Not to Settle Patent Litigation.” Strategic Management Journal 24(1): 17-38. Taylor, C. T., and Z. A. Silberston. (1973). The Economic Impact of the Patent System: A Study of the British Experience. Cambridge, UK: Cambridge University Press. Tilton, J. E. (1971). International Diffusion of Technology: The Case of Semiconductors, Washington, D.C.: Brookings Institution. U.S. Patent and Trademark Office (USPTO). (1995). Technology Profile Report: Semiconductor Devices and Manufacture, 1/1969-12/1994. Washington, DC: U.S. Department of Commerce. Varchaver, N. (2001). “The Patent King.” Fortune, May 14, pp. 203-216. Von Hippel, E. (1988). The Sources of Innovation. Oxford, UK: Oxford University Press. Warshofsky, F. (1994). The Patent Wars: The Battle to Own the World’s Technology. New York, NY: John Wiley & Sons, Inc. Ziedonis, R. H. (2000). Firm Strategy and Patent Protection in the Semiconductor Industry, unpublished doctoral dissertation, Walter A. Haas School of Business, University of California, Berkeley.

OCR for page 180
Patents in the Knowledge-Based Economy APPENDIX A CLASSIFYING LITIGATED PATENTS AS PROCESS OR PRODUCT-RELATED TECHNOLOGIES In surveys on appropriability (Levin et al., 1987; and Cohen et al., 2001), R&D managers consistently report that patents for new or improved products are generally more effective at preventing duplication or securing royalty income than those for new or improved innovations used in the manufacturing process. The general argument is that process innovations are less subject to public scrutiny and, therefore, are more easily kept secret than new or improved features of a final product. Similarly, infringement of process innovations may be more difficult to detect if they are embedded in difficult-to-observe manufacturing operations. At the same time, some of the most publicized patent infringement suits in the semiconductor industry over the past two decades involved manufacturing-related inventions. For example, Texas Instruments earned millions in royalties on its patents covering methods to transport wafers after successfully suing Korean and Japanese firms in the mid-1980s. More notorious are the successful lawsuits by Jerome Lemelson, an independent inventor who owned patents covering scanning technologies used to monitor and control production systems. By the end of 2000, the Lemelson Foundation had collected over $1.5 billion in licensing fees from users of more modern bar code scanners, including manufacturers in the automobile and semiconductor industries (Varchaver, 2001). Has the composition of patent cases filed in this industry tilted more toward these “royalty-seeking” disputes in which the value of the exclusionary right increases as the technology becomes more widely adopted throughout the industry? I hope to investigate this question econometrically in future versions of this research by examining the subset of cases that involve manufacturers and comparing the age and type of technologies involved in those disputes before and after the mid-1980s. Although we have made progress in this direction, we need to resolve two issues before understanding how much credence we should lend to our results. First, as discussed in the text, I need to confirm whether I have a representative sample of cases in the early (1973-1984) period. Secondly, I need to verify that our coding of inventions as product- or process related is reasonable and accurate. I would also need to normalize any findings based on the litigated patents with patents that are not involved in filed disputes. Otherwise, I could simply be picking up the fact that the stock of semiconductor-related patents “at risk” for litigation may have grown older or more toward process innovation as the technology has matured. With these caveats and longer-term objectives in mind, this appendix summarizes the approach I used to classify “product”- and “process”-related inventions within the litigation sample.

OCR for page 180
Patents in the Knowledge-Based Economy TABLE A.1 Litigated Patents by Invention Type Code Innovation Type Definition and Notes Examples #Litigated Patents by Type (%total) 1 Product: Semiconductor Devices Definition: inventions that pertain to new or improved semiconductor devices Note: Must include claims that cover the device but may also cover methods claims US3138742: Miniaturized Electronic Circuits (TI Kilby patent,issued 1964); US4609986: Programmable Logic Device Using EPROM Technology (Altera Corp, issued 1986) 277 (55%) 2 Process: Manufacturing Definition: inventions that pertain to new or improved manufacturing processes Note: includes inventions related to materials and equipment used in semiconductor manufacture or more general manufacture. Processes US4256534: Device Fabrication by Plasma Etching (Bell Labs, issued 1981); US3735350: Code Scanning System (Jerome Lemelson, issued 1973) 118 (23%) 3 Product: Broader Electronics or Computer-Related Definition: inventions that pertain to new or improved products that use or embed semiconductor devices but that are not stand-alone semiconductor products Note: dominated by inventions related to data processing and computing US4074351: Variable Function Programmed Calculator (Boone calculator patent,TI,issued 1978) 103 (20%) 4 Other/Unclear Definition: inventions that pertain to new or improved products or processes that appear unrelated to semiconductors US4553515: Cylinder Head for Spark Ignition Internal Combustion Engines (BL Technology Limited; issued 1985) 6 (1%)

OCR for page 180
Patents in the Knowledge-Based Economy Although conceptually straightforward, determining whether the claims of a patent cover product- or process-related improvements is far from simple. As a first approach, I identified patents pertaining to semiconductor devices (“products”) or their manufacture (“processes”) with the U.S. patent classification scheme. The results, however, were unsatisfactory. Of 504 litigated patents involving semiconductor firms in the sample, less than 30 percent fell into the three-digit U.S. patents classes that cover device inventions (classes 326, 327, 365, and 257) and less than 10 percent of the litigated patents fell into the three-digit class for “semiconductor device manufacturing: process” (438). These results were not entirely surprising for the process-related inventions given the wide range of materials, methods, and equipment used in semiconductor manufacturing. To overcome this problem, I manually classified each litigated patent into one of four mutually exclusive categories. The main objective was to classify a patent as “product” if it pertained primarily to new or improved semiconductors or related devices (e.g., memory chips, logic devices, or analog-digital converters) and “process” if it pertained to new or improved manufacturing processes but did not claim rights to specific devices per se. Categories 1 and 2 include semiconductor-related “product” and “process” inventions, respectively. In reviewing the patents, however, it was clear that many of the patents pertained to products that embed semiconductors (e.g., computer systems, hand-held calculators) instead of stand-alone semiconductor devices. Category 3 includes these “downstream” product-related inventions. Finally, Category 4 contains patents that were difficult to classify into one of the above categories or seemed to pertain to innovations unrelated to semiconductors. Table A.1 summarizes each “invention type” category and the number of litigated patents assigned to each category. Of the 504 litigated patents, I identified 277 that made specific claims to new or improved semiconductor devices (Category 1), 188 that pertained solely to methods, equipment, or materials used in the manufacturing process (Category 2), and 103 that pertained to innovations in downstream products (Category 3). A small number of patents (6) seemed unrelated to semiconductors or did not fall clearly within one of these categories. This suggests that 75 percent of the litigated patents in the sample pertain, broadly, to product innovation and 25 percent focus solely on processes used in manufacturing. In follow-on studies, I hope to use these results to inform whether concerns about “hold up” by manufacturers are illuminated by a change in the age or types of patents enforced by or against these firms in the period associated with stronger patent rights.

OCR for page 180
Patents in the Knowledge-Based Economy This page in the original is blank.