Do Not Delete
11/24/2013 11:28 AM
ARTICLE
PROLIFERATING PATENTS AND PATENT
LAW’S “COST DISEASE”
John M. Golden∗
ABSTRACT
Whether or not the patent system is truly in “crisis,” it is
certainly under considerable strain. This Article uses a variety of
statistics to analyze the U.S. patent system’s development
through time. These include not only already readily available
statistics such as the numbers of U.S. patents issued per year,
but also yearly statistics for the number of U.S. patent examiners
and for the number of full-time-equivalent U.S. research-anddevelopment employees—statistics that seem to have previously
appeared only in relatively isolated and scattered fashion.
Through analysis of various selections, combinations, and
variants of these statistics, including logarithms and ratios, the
Article seeks to develop a better sense of what concerns with the
U.S. patent system such figures most persuasively suggest. A key
point is that, although recent percentage rates of growth
characteristic of the U.S. patent system might be nowhere near
the ballistic levels of the nineteenth century, the rate at which
patents issue is still typically accelerating. This acceleration
predictably creates tension as public and private bureaucracies,
as well as individuals, struggle to keep up. The Article
characterizes resulting strains as symptoms of a variant of
∗
Professor in Law, The University of Texas at Austin. Thanks to Maria Amon
and Jane O’Connell for research assistance. For comments, I thank John Duffy, Mark
Lemley, Jonathan Masur, Laura Pedraza-Fariña, Arti Rai, Greg Reilly, Polk Wagner,
Christopher Yoo, an anonymous reviewer, the editors of the Houston Law Review, and
participants at a symposium hosted by the University of Houston’s Institute for
Intellectual Property & Information Law, the 2013 Intellectual Property Scholars
Conference, the 2013 Conference on Empirical Legal Studies, and a colloquium at the
George Washington University Law School.
455
Do Not Delete
456
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
William Baumol’s more general “cost disease” for services. The
Article discusses potential responses, including greater
worksharing with foreign patent offices, increased privatization
of examination functions, more effective U.S. Patent and
Trademark Office rulemaking, or partial automation of
examination.
TABLE OF CONTENTS
I.
INTRODUCTION...................................................................... 456
II. PATENT PROLIFERATION AS HISTORICAL NORM, NOT
RECENT PATHOLOGY ............................................................. 462
A. The Historical Trend of Accelerating Patenting ........... 463
B. Patenting by Non-U.S. Residents and the Uptick in
Patenting’s Percentage Growth ..................................... 470
C. An Overwhelming “Patent Tax”? Patent Issuance Per
Dollar of GDP................................................................ 472
III. PROLIFERATING PATENTS’ RELENTLESS PRESSURE FOR
ADMINISTRATIVE REFORM .................................................... 476
A. Patent Law’s Labor Problem ......................................... 476
B. Patent Law’s History of Administrative Change and
Growth........................................................................... 483
IV. INSTITUTIONAL REFORMS TO MEET THE CHALLENGE:
BEYOND MERELY SCALING UP THE PATENT
BUREAUCRACY ...................................................................... 489
A. Pooling Resources with Foreign Patent Offices ............ 490
B. Privatization Options .................................................... 492
C. Rulemaking or Other Approaches to “Partially
Automated” Examination.............................................. 497
V.
CONCLUSION ...................................................................... 500
I.
INTRODUCTION
Since at least 1999, the exact words “The patent system is in
crisis” have appeared so often in academic literature that they
might be considered a meme.1 Whatever the cultural status of the
1. See, e.g., DAN L. BURK & MARK A. LEMLEY, THE PATENT CRISIS AND HOW THE
COURTS CAN SOLVE IT 3 (2009) (“The patent system is in crisis.”); Colleen V. Chien,
Reforming Software Patents, 50 HOUS. L. REV. 325, 327 (2012) (“The patent system is in
crisis.”); Paul Edward Geller, An International Patent Utopia?, 85 J. PAT. & TRADEMARK
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
457
“patent crisis” clause, however, its descriptive utility seems open
to question. If there is a crisis in patent law, it appears to be of a
long-lasting and distinctively vibrant sort. The number of patents
issued each year continues to grow quite steadily: the number of
utility patents issued in 2012 was 253,155, nearly 100,000 more
patents than issued in 1999.2 The number of utility patent
applications has grown even more dramatically during the same
time period: 270,187 were filed in 1999; more than double that
number, 542,815, were filed in 2012.3 Moreover, “patent system”
growth has been international. The number of patent
applications filed worldwide has recently grown by more than 7%
annually,4 with the People’s Republic of China alone
“account[ing] for 72% of the world’s patent-filing growth between
2009 and 2011.”5 In short, despite a prolonged patent “crisis” at
the start of the twenty-first century, the century seems destined
to be patent-rich. Is the constant cry of “crisis” no more than an
eye-catching headline?
“Crisis”—with its connotation of a turning point—might not
be the right word. Part II of this Article argues that the
relentless proliferation of patents in the past few decades is
really nothing terribly new. Recent increases in rates of
patenting are actually reasonably modest by the standards of the
nineteenth century, and much of the recent overall increase is
attributable to growth in U.S. patents issued to foreign
individuals or entities—i.e., to greater internationalization of
patenting and, arguably, innovation. Any sense that patents
have become an unsustainable burden on the economy seems at
least somewhat leavened when one considers that the number of
U.S. patents issued annually per real dollar of U.S. gross
OFF. SOC’Y 582, 582 (2003) (“The patent system is in crisis.”); Robert P. Merges, As Many
as Six Impossible Patents Before Breakfast: Property Rights for Business Concepts and
Patent System Reform, 14 BERKELEY TECH. L.J. 577, 591 (1999) (“The concerns about
quality, especially in light of the data on overall volume, point to one conclusion: the
patent system is in crisis.”); cf. JAMES BESSEN & MICHAEL J. MEURER, PATENT FAILURE:
HOW JUDGES, BUREAUCRATS, AND LAWYERS PUT INNOVATORS AT RISK 5 (2008) (“Overall,
the performance of the patent system has rapidly deteriorated in recent years.”).
2. U.S. PATENT & TRADEMARK OFFICE, U.S. PATENT ACTIVITY: CALENDAR YEARS
1790 TO THE PRESENT 1 (2013), available at http://www.uspto.gov/web/offices/ac/ido/oeip/
taf/h_counts.pdf [hereinafter U.S. PATENT ACTIVITY].
3. Id.
4. Patent Applications, ECONOMIST, Dec. 15, 2012, at 89 (noting that growth in
patent-application filings “was over 7% for the second year running”).
5. Id. (“China’s patent office received more applications than any other country’s in
2011 . . . .”). See generally How Innovative Is China?, ECONOMIST, Jan. 5, 2013, at 52
(noting that a surge in Chinese patenting might be “because government incentives have
prompted people to file lots of iffy patent applications, which the local patent office has a
tendency to approve”).
Do Not Delete
458
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
domestic product (GDP) remains at a level roughly comparable to
or substantially less than the levels prevailing from 1929 through
1970.6
In short, although not definitive or necessarily even strongly
indicative about how satisfied or dissatisfied we should feel with
the current state of patent law, these and other statistics suggest
that contentions that there is a “crisis” and assertions about the
nature of any such “crisis” need to do more than reference recent
multiplicative growth in patent-issuance or application rates as
probative of a “crisis” or its causes.7 In light of the patent
system’s long history of patent proliferation, it in fact seems very
difficult to know what the proper quantitative level of patenting
should be at any point in time. Even after “normalizing”
patenting levels by dividing patents issued in a given year by
that year’s real GDP or by another figure such as a measure of
the number of people in the United States engaged in research
and development, it seems hard to find any obviously optimal
baseline or natural level of equilibrium with which one should
compare.8 As a result, the question of what, if anything, modern
patenting rates and their growth with time signals about the
relative health of the U.S. patent system seems substantially
open to debate.
Nonetheless, whether or not “crisis” is the right word to
describe the current state of the U.S. patent system, there does
seem to be substantial strain on that system. Indeed, such strain
seems a predictable symptom of the system’s very success at
proliferation. Although the percentage rate of growth in
patenting might be nowhere close to the ballistic levels of the
nineteenth century, the rate at which patents issue is still
9
accelerating. This acceleration creates constant tension as public
and private bureaucracies, as well as individuals, struggle to
6. Compare U.S. PATENT ACTIVITY, supra note 2, at 1–2, with Current-Dollar and
“Real” GDP, BUREAU OF ECON. ANALYSIS, U.S. DEP’T OF COMMERCE, http://bea.gov//
national/index.htm#gdp (last visited Nov. 19, 2013).
7. Cf., e.g., BURK & LEMLEY, supra note 1, at 22 (“Even if the world is more
innovative than it used to be, we doubt it is four times more innovative than . . . in the
1980s, or . . . nearly twelve times as innovative as the 1870s . . . . The more logical
explanation is that it is simply easier to get a patent today than it used to be . . . .”);
Patrick Doody, The Patent System Is Not Broken, INTELL. PROP. & TECH. L.J., Dec. 2006,
at 12 (“The critics bemoan the number of patent applications filed and issued per year and
remarkably rely on that fact as an indicator that the patent system is broken.”). See
generally Ronald J. Mann, Do Patents Facilitate Financing in the Software Industry?, 83
TEX. L. REV. 961, 1003 n.220 (2005) (expressing an intention “to rebut the idea that the
proliferation of software patents has stifled R&D spending in any measurable way”).
8. See infra Part II.C.
9. U.S. PATENT ACTIVITY, supra note 2, at 1–5.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
459
keep up. This Article, composed for a symposium oriented toward
intersecting concerns of intellectual property and administrative
law, ultimately focuses on the tensions generated for the
USPTO’s public bureaucracy, but the fact that patent law also
generates strains on the information-processing, decisionmaking, and financial capacities of private firms, individuals,
and courts should not be forgotten.
As Parts II and III indicate, the number of patents per real
dollar of GDP might not be at a historically high level,10 but the
numbers of patents per member of the U.S. civilian workforce
and per full-time-equivalent U.S. research-and-development
11
employee apparently are. Further, for the last three decades,
percentage growth in the number of U.S. patent examiners has
proceeded at a pace more characteristic of the mid-nineteenth
12
century than most of the twentieth. With the U.S. Patent and
Trademark Office (USPTO) striving to increase the number of
patent examiners and administrative patent judges to
unprecedented levels—and opening regional offices apparently in
part to facilitate this13—the patent system’s capacity to strain
relatively scarce human capital seems clear. Indeed, such strains
have, like the patent system’s accelerating growth, been an
apparently continual feature of its existence for more than two
centuries.14 These strains have forced administrative change in
15
the past, and we should expect that they will continue to do so
10.
11.
12.
13.
Infra Part II, Figure 3.
Infra Part III, Figures 4, 5.
See infra text accompanying note 96.
See U.S. PATENT & TRADEMARK OFFICE, PERFORMANCE AND ACCOUNTABILITY
REPORT: FISCAL YEAR 2012, at 3, 38, available at http://www.uspto.gov/about/stratplan/ar/
USPTOFY2012PAR.pdf [hereinafter 2012 USPTO PERFORMANCE REPORT].
14. See infra Part III.B (discussing patent law’s history of administrative change
and growth). See also Officers and Employees—Training—Private Persons, 42 Comp.
Gen. 673 (1963) (referencing the “shortage of well-trained patent examiners outside of
government” in the 1960s); LEE A. HOLLAAR, LEGAL PROTECTION OF DIGITAL
INFORMATION 244–46 (2002) (noting that, under the Patent Act of 1790, three Cabinet
members charged with reviewing patent applications struggled with the time demands of
the task); Warren K. Mabey, Jr., Deconstructing the Patent Application Backlog . . . A
Story of Prolonged Pendency, PCT Pandemonium & Patent Pending Pirates, 92 J. PAT. &
TRADEMARK OFF. SOC’Y 208, 231 (2010) (“The rise of the technology driven global
economy, expansion of patentable subject matter, and increase in patent value has not
been met with a comparable increase in the office’s ability to dispose of
applications . . . and the backlog continues to increase. . . . There are also simply not
enough examiners to review the flood of applications.”).
15. See infra Part III.B (discussing patent law’s history of administrative change
and growth); see also U.S. PATENT & TRADEMARK OFFICE, 2010–2015 STRATEGIC PLAN 8,
12, 14 (2013), available at http://www.uspto.gov/about/stratplan/USPTO_2010-2015_
Strategic_Plan.pdf (outlining the USPTO’s plan to combat the patent backlog); John D.
Grier, Comment, Chasing Its Own Tail? An Analysis of the USPTO’s Efforts to Reduce the
Patent Backlog, 31 HOUS. J. INT’L LAW 617, 628 (2009) (“[T]he USPTO responded to the
Do Not Delete
460
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
in the future. The continuing story of the patent system’s
administrative and substantive evolution is thus in significant
part a story of the response to the patent system’s variant of
William Baumol’s “cost disease” for services, in accordance with
which more obstinately labor-intensive endeavors struggle to
keep up with productivity gains in the rest of the economy, and
thus effectively operate as a comparative drag on overall
growth.16
How can the present-day administrative apparatus of patent
law adapt so that it can continue to bear the weight of an everincreasing number of applications? In this Article, I will focus for
the most part on the upfront processing of applications by the
USPTO. Back-end review of patents, as by the courts and
International Trade Commission, is a related but distinct
problem, complicated in terms of reform options by the United
States’ constitutional provision for civil jury trial rights through
17
the Seventh Amendment.
There seem to be at least three main ways in which the
USPTO can respond to the ever-rising tide of applications:
(1) greater international work-sharing arrangements that might
provide a relatively immediate, albeit perhaps only temporarily
satisfactory, response to the internationalization of patent
applications flowing into the USPTO; (2) greater privatization of
functions of assessing patents’ or patent applications’ validity
and proper scope; and (3) greater use of rulemaking or other
means of “partially automating” processes of individual
examination.18 The first approach, international work-sharing,
appears to have received significant recent attention from the
patent backlog by implementing procedure changes and proposing rule changes to the
patent examination process . . . .”).
16. See WILLIAM J. BAUMOL, THE COST DISEASE: WHY COMPUTERS GET CHEAPER
AND HEALTH CARE DOESN’T 22 (2012) (noting the “endurance of productivity stagnancy” in
industries such as health care, education, and law characterized by a “handicraft—or inperson—attribute of their supply processes”); William J. Baumol, The Two-Sided Cost
Disease and Its Frightening Consequences, in THE HANDBOOK OF INNOVATION AND
SERVICES 84, 85 (Faïz Gallouj & Faridah Djellal eds., 2010) (observing that items
suffering from persistently rising costs “generally have a handicraft element in their
production process, whose labor content is therefore difficult to cut down”).
17. See The Supreme Court, 1995 Term—Leading Cases, 110 HARV. L. REV. 135, 266
(1996) (discussing the Supreme Court’s decision that the construction of a patent is within
the province of the court, and that the Seventh Amendment does not guarantee that a
jury may determine the meaning of a disputed term of art); see also Colleen V. Chien,
Patently Protectionist? An Empirical Analysis of Patent Cases at the International Trade
Commission, 50 WM. & MARY L. REV. 63, 108 (2008) (“Empowering the ITC, which lacks a
jury, to award damages might appear to run afoul of the Seventh Amendment. . . . The
Supreme Court, however, has held that the Seventh Amendment is not applicable to
administrative proceedings.”).
18. See infra Part IV (discussing these proposed reforms in detail).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
461
USPTO and foreign patent offices,19 although progress has likely
been slower than many would desire—presumably at least partly
as a result of political hurdles and idiosyncrasies of individual
20
national or regional patent systems. The second approach has
received glimmers of attention but is likely still underexploited.
The third approach likewise seems substantially underexplored.
A possible barrier to the third approach is the desire for
individualized consideration that is often an obstacle to
implementation of a system of “mass” or “bureaucratic” justice.21
But here is another point at which the reader is meant to be
impressed by emphasis in Parts II and III on the current and
likely future volumes of patents and patent applications—in
particular, on the apparently likely continuation of the historical
trend of accelerating growth in patenting. Mechanisms for
administering the patent system, including private mechanisms
for the clearance and licensing of patent rights, threaten to
become relatively ineffectual under the crushing weight of everaccumulating patents and patent applications. Unless we wish to
dramatically limit access to the U.S. patent system by very
substantially ratcheting up gatekeeping devices such as USPTO
fees, procedural burdens, or substantive requirements of
patentability, we need to develop more efficient ways of
processing patents, applications, and information relating to
them.
A key step in moving toward more efficient processing
within the patent system might involve more open
acknowledgement of the reality that the patent system has, at its
core, always been a system designed to provide only a relatively
19. See David Kappos, Under Sec’y of Commerce & Dir. of USPTO, Remarks at the
Global & European Initiatives for Work Sharing Conference in Budapest: Collaborative
Innovation—A Global Call for Enhancing Work Sharing (May 18, 2011), available at
http://www.uspto.gov/news/speeches/2011/kappos_budapest.jsp.
20. David Kappos, Under Sec’y of Commerce for IP & Dir. of USPTO, Remarks at IP
Week 2011 in Brussels: Harmonizing Patent Systems—A Common Mission at an
Economic Turning Point (Dec. 9, 2011), available at http://www.uspto.gov/news/speeches/
2011/kappos_ipweek2011.jsp; see also Anneliese M. Seifert, Will the United States Take
the Plunge into Global Patent Law Harmonization? A Discussion of the United States’
Past, Present, and Future Harmonization Efforts, 6 MARQ. INTELL. PROP. L. REV. 173,
178–79, 197, 203 (2002) (discussing how U.S. concern with the interests of small-entity
inventors has impeded harmonization of patent law across countries).
21. See, e.g., David Rosenberg, Class Actions for Mass Torts: Doing Individual
Justice by Collective Means, 62 IND. L.J. 561, 561 (1987) (describing bureaucratic justice
as “a mode of decision-making associated with administrative agencies, which lacks the
common law’s traditional commitment to party control and focus on the discrete merits of
each claim”); Charles E. Reuther, Class Actions and the Quest for a Fair Resolution in
Mass Tort Litigation, N.J. LAW. MAG., Aug. 2011, at 25 (discussing concerns with
individual justice in mass tort litigation).
Do Not Delete
462
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
crude form of justice—one responsive to individual initiative but
also bound to rough legal proxies for merit, such as
“nonobviousness”22 and bound also by exogenous circumstances
that can make an apparently meritorious innovation a relative
market failure and can make a seemingly trivial innovation a
great monetary success.23 Recognizing that the patent system in
effect implements a form of quite crude, unevenly distributed
“mass justice” might well be crucial to improving and
streamlining its operations to best serve the needs of a patentrich twenty-first century. If patents continue to issue at
accelerating rates, “muddling through”24 on matters of the patent
system’s administrative organization might not be enough.
Patents’ proliferation generates a fundamental numerical
challenge for patent-system administration that might require
one or more paradigm shifts in order to produce a reasonably
sustainable answer. A “mass justice” model for patent
administration, using rulemaking and imprecisely controlled
private mechanisms, might be a substantial part of the best
answer we can have.
II. PATENT PROLIFERATION AS HISTORICAL NORM, NOT RECENT
PATHOLOGY
The United States currently issues three basic types of
patents—utility patents for inventions of seemingly almost any
25
functional, technological kind; plant patents for new, asexually
26
reproduced plant varieties; and design patents for ornamental
22. See 35 U.S.C. § 103(a) (2006) (“A patent may not be obtained . . . if the
differences between the subject matter sought to be patented and the prior art are such
that the subject matter as a whole would have been obvious at the time the invention was
made . . . .”); How to Get a Patent, U.S. PATENT & TRADEMARK OFFICE, http://
www.uspto.gov/web/patents/howtopat.htm (last updated Mar. 26, 2008) (explaining that
inventors must take initiative and file an application to receive a U.S. patent).
23. Cf. A. Michael Noll, Anatomy of a Failure: Picturephone Revisited, 16
TELECOMM. POL’Y 307, 307, 309 (1992) (describing the market failure of the highly
anticipated Picturephone); Michelle Fox, Surprising Ideas that Made Millions, CNBC
(Mar. 15, 2012), http://www.cnbc.com/id/46747902/page/4 (noting the financial success of
“Billy Bob Teeth”).
24.
John M. Golden, Biotechnology, Technology Policy, and Patentability:
Natural Products and Invention in the American System, 50 EMORY L.J. 101, 191
(2001) (concluding that, “given present uncertainties, purposeful and careful
muddling through [with respect to ‘patent law’s problems’] may be the best way to
begin the new century”).
25.
See 35 U.S.C. § 101 (2012) (“Whoever invents or discovers any new and
useful process, machine, manufacture, or composition of matter, or any new and
useful improvement thereof, may obtain a patent thereof, subject to the conditions
and requirements of this title.”).
26.
Id. § 161 (“Whoever invents or discovers and asexually reproduces any
distinct and new variety of plant . . . other than a tuber propagated plant or a plant
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
463
aspects of “article[s] of manufacture.”27 As discussed below, the
general trend has been for the rate at which all three types of
patents issue—i.e., the number of patents of each type issued
28
each year—to grow with time. Indeed, at least for the more than
century-old categories of utility and design patents, the general
long-term trend has been for the number of patents issued each
29
year to increase at an accelerating pace —albeit a pace that has
tended not to keep up with the growth of the United States’ GDP
in inflation-adjusted real dollars.30 These long-term trends
suggest that some of the recent talk of a “patent crisis” might be
overblown. In the longer historical view, the oft-noted shift in
U.S. patenting from relative stasis or decline prior to the mid31
1980s to accelerating growth after the mid-1980s could be
considered more of a “return to normalcy” than a descent into
intellectual property extravagance. As Part III will discuss,
various concerns associated with the return to accelerating
growth are real and are particularly important for those of us
interested in the proper structuring of patent law’s
administrative institutions.
A. The Historical Trend of Accelerating Patenting
The United States has been issuing utility patents since
1790, when three such patents issued.32 The number of patents
issued each year has grown prodigiously with time.33 Such growth
was particularly explosive in the nineteenth century, a fact that
can be appreciated by consideration of some relatively
straightforward numbers. In the period from 1800 through 1809,
34
911 utility patents issued. Only 50 years later, in the 1850s,
found in an uncultivated state, may obtain a patent therefor, subject to the
conditions and requirements of this title.”).
27. Id. § 171 (“Whoever invents any new, original and ornamental design for an
article of manufacture may obtain a patent therefor, subject to the conditions and
requirements of this title.”).
28. See infra text accompanying notes 37–57.
29. See U.S. PATENT ACTIVITY, supra note 2, at 1–3.
30. Compare id. at 1–3, with Current-Dollar and “Real” GDP, supra note 6.
31. See, e.g., Adam B. Jaffe, The U.S. Patent System in Transition: Policy
Innovation and the Innovation Process, 29 RES. POL’Y 531, 536 (2000) (observing that
“[p]atenting by U.S. inventors in the U.S. had been constant or declining for much of the
1970s and early 1980s” but that, “beginning in 1984, and accelerating in 1988, patent
applications by U.S. residents began to increase”).
32. U.S. PATENT ACTIVITY, supra note 2, at 5; see also ROBERT PATRICK MERGES &
JOHN FITZGERALD DUFFY, PATENT LAW AND POLICY: CASES AND MATERIALS 7–8 & n.22
(2013) (reporting that the first U.S. patent, which was “for a process for making potash,”
issued on July 31, 1790).
33. U.S. PATENT ACTIVITY, supra note 2, at 1–5.
34. Id. at 4–5.
Do Not Delete
464
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
more than 20 times that number of patents, 19,661, issued.35
Another 50 years on, from 1900 through 1909, the number of
utility patents issued—304,726—was more than 15 times greater
36
still.
In the twentieth and early twenty-first centuries, the
absolute number of utility patents issued in a year continued
generally to grow, ultimately overcoming notable but temporary
downturns from about 1933 through 1947 and from about 1972
37
through 1983. Despite the generally continuing growth trend,
however, the overall percentage increases in patenting tended to
be much smaller than in the 1800s.38 In the first half of the
39
twentieth century, a time marked by two world wars and the
massive economic dislocation of the Great Depression,40 a norm of
single-digit multiplicative growth per half century replaced the
nineteenth-century norm of double-digit multiplicative growth
over such a span. Thus, from 1950 through 1959, a total of
425,988 utility patents issued, about 1.4 times the figure for the
period from 1900 through 1909.41 By the decade from 2000
through 2009, U.S. utility patents had returned to a somewhat
higher percentage-growth rate, albeit nothing like the
percentage-growth rates characteristic of the 1800s. The total
number of U.S. utility patents issued from 2000 through 2009
was 1,624,154—nearly 4 times the total a half century earlier.42
One might conjecture that some of the resumed growth in
the latter half of the twentieth century was little more than
“paper growth” that resulted, for example, from more frequent
use of continuation applications or other applications that might
be viewed as directed toward obtaining patents that are to a
significant degree redundant of patents obtained from parent
applications.43 Such redundancy has apparently been a possibility
35. Id. at 3–4.
36. Id. at 3.
37. Id. at 1–3.
38. Id. at 1–5.
39. See R.R. PALMER & JOEL COLTON, A HISTORY OF THE MODERN WORLD 665 (5th
ed. 1978) (“The First World War lasted over four years, from 1914 to the end of
1918 . . . .”); id. at 819 (describing the “Second World War of the twentieth century” as the
greatest conflict in human history).
40. Id. at 762 (describing consequences of “the oceanic wreckage of the great
depression”); see also HISTORY OF THE WORLD: THE LAST FIVE HUNDRED YEARS 633
(Esmond Wright ed., 1984) (observing that, during the Great Depression, “industrial
production in Germany and the United States fell by almost fifty percent in three years”).
41. U.S. PATENT ACTIVITY, supra note 2, at 2–3.
42. Id. at 1–2.
43. JANICE M. MUELLER, PATENT LAW 76 (4th ed. 2013) (“An obvious-type double
patenting rejection can be overcome by the filing of a terminal disclaimer [according to]
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
465
throughout since at least the mid-nineteenth century,44 but a recent
study has indicated that the number of continuation and
continuation-in-part applications filed with the USPTO has recently
45
risen more steeply than the total number of applications. Another
study has indicated that, during the 5 years from 1996 through
2000, 15.5% of issued patents “were issued based on disclosures
46
that were in common with other patents to the same applicant.” If
there has been a growth in such redundancy over time, raw patent
issuance numbers might overstate to at least some degree the
significance of patent proliferation. On the other hand, for the
patent-administration “long game” that is the fundamental concern
of this paper, even if we make the unfavorable assumptions (a) that
there is a present redundancy of about one sixth in issued patents,
(b) that this redundancy is something wholly new within the last
few decades, and (c) that this redundancy is 100% relevant for
purposes of determining whether there has been pertinent “double
counting,” the effect seems little more than a rounding error. Where
the concern is multifold increases in patenting over decades, even a
possible need to correct certain numbers by approximately 16%
apparently does little to change the basic qualitative picture.
What has driven the basic trend of ever accelerating growth?
Growth in the numbers of utility patents issued each year has in
part reflected growth in the array of technologies potentially
covered by such patents, a growth that has been partly a product
of the proliferation of technologies,47 but also partly a product of a
which the applicant formally agrees that if he is awarded a patent on the application
claims [at issue], that patent will be deemed to expire on the same date as the applicant’s
first patent (on which the double patenting rejection was based.”); Robert A. Clarke, U.S.
Continuity Law and Its Impact on the Comparative Patenting Rates of the U.S., Japan
and the European Patent Office, 85 J. PAT. & TRADEMARK OFF. SOC’Y 335, 338 (2003) (“By
filing multiple continuing applications it is possible for an applicant to obtain a number of
patents based on the same initial invention, subject to the possible need to file terminal
disclaimers. . . .”).
44. Stephen T. Schreiner & Patrick A. Doody, Patent Continuation Applications:
How the PTO’s Proposed New Rules Undermine an Important Part of the U.S. Patent
System with Hundreds of Years of History, 88 J. PAT. & TRADEMARK OFF. SOC’Y 556, 559
(2006) (noting that a statutory provision for continuation applications enacted in 1952
“simply codified long-standing practice that has been in place for well over a century”).
45. Compare Christopher A. Cotropia, Cecil D. Quillen, Jr. & Ogden H. Webster,
Patent Applications and the Performance of the U.S. Patent and Trademark Office, 23
FED. CIR. B.J. 179, 181 (2013) (“The number of applications filed increased by 179% from
1996 to 2012 . . . .”), with id. at 183 tbl.2 (reporting that the number of continuation and
continuation-in-part filings grew by nearly a factor of four from a total of 18,203 in 1996
to a total of 72,079 in 2012).
46. Clarke, supra note 43, at 338.
47. Cf. W. BRIAN ARTHUR, THE NATURE OF TECHNOLOGY 175–76 (2009) (observing
that “every technology requires supporting technolog[y]” and the “vast majority of niches
for technology are created not from human needs, but from the needs of technologies
themselves”); KEVIN KELLY, WHAT TECHNOLOGY WANTS 284 (2010) (“The number of
Do Not Delete
466
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
view of patentable subject matter that, at least until the last
several years, had seemed to become more expansive with time.48
Expansion of the U.S. patent system has also occurred through
the addition of the two other types of patents mentioned above:
design patents, added by statute in 1842 as a separate option for
ornamental aspects of functional objects,49 and plant patents,
50
added in the 1930s as an option for asexually reproduced plants.
Since 1842, growth in the number of design patents has
51
followed an upward trajectory that has been roughly similar to
that for utility patents: rapid upward growth in the nineteenth
century and still accelerating growth but a lower percentage growth
rate in the twentieth. The number of design patents issued in the
1850s was 925; the number issued in the period from 1900 through
1909 was 8,369; the number issued in the 1950s was 30,287; and
the number issued in the period from 2000 through 2009 was
188,663.52 The multiplicative factors by which design patent
issuance increased from half century to half century were thus 9.0,
3.6, and 6.2, respectively.
Because U.S. plant patents began issuing only in the 1930s, we
only have a couple half-century figures for them. For example, 992
plant patents issued during the 1950s, and 9,436 issued during the
53
period from 2000 through 2009. Thus, the average number of
species of technology invented every year is increasing at an increasing rate.”); John M.
Golden, Innovation Dynamics, Patents, and Dynamic-Elasticity Tests for the Promotion of
Progress, 24 HARV. J.L. & TECH. 47, 61–62 (2010) (discussing positive-feedback effects
that can lead to both technological advances and technologies themselves proliferating
with time).
48. See Edward Van Gieson & Paul Stellman, Killing Good Patents to Wipe Out Bad
Patents: Bilski, the Evolution of Patentable Subject Matter, and the Inability to Save
Valuable Patents Using the Reissue Statute, 27 SANTA CLARA COMPUTER & HIGH TECH.
L.J. 403, 409 (2011) (“The effect of Bilski on future patents is that it imposes additional
requirements for obtaining valid, enforceable software and business method patents.”); see
also John R. Thomas, The Patenting of the Liberal Professions, 40 B.C. L. REV. 1139, 1139
(1999) (“Times have changed . . . . The sheer range of recently issued patents suggests
that few restraints bound the sorts of subject matter that may be appropriated via the
patent system.”).
49. See 35 U.S.C. § 171 (2012) (“Whoever invents any new, original and ornamental
design for an article of manufacture may obtain a patent therefor, subject to the
conditions and requirements of this title.”); see also Jason J. Du Mont, A Non-Obvious
Design: Reexamining the Origins of the Design Patent Standard, 45 GONZ. L. REV. 531,
542 (2009) (“[O]n August 29, 1842, Congress enacted the first statute authorizing patent
protection for designs.”).
50. Emily Marden & R. Nelson Godfrey, Intellectual Property and Sharing Regimes
in Agricultural Genomics: Finding the Right Balance for Innovation, 17 DRAKE J. AGRIC.
L. 369, 387 (2012) (“The United States initially established limited patent rights for
asexually reproduced varieties of plants covered under the Plant Patent Act of 1930.”).
51. See U.S. PATENT ACTIVITY, supra note 2, at 1–4.
52. Id.
53. Id.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
467
plant patents issued each year increased by nearly tenfold over a
period approximately the latter half of the twentieth century.
Interestingly, however, in the first decade or so of the twentyfirst century, growth in plant-patent issuance appears potentially
to have topped off or even gone into decline. Since the number of
plant patents issued in a year surpassed 1,000 in 2002, during
which 1,133 of such patents issued, the number of plant patents
issued in a given year has been largely flat or decreasing: the
current all-time maximum occurred in 2008, when 1,240 issued,
and the numbers issued in 2010 through 2012—981, 823, and 860,
respectively—have all been below 1,000.54 In contrast, utility-patent
issuance reached an all-time high of 253,155 in 2012.55 In that same
year, all-time highs were reached for the numbers of both utilitypatent applications and design-patent applications filed—542,815
and 32,799, respectively—whereas the number of plant-patent
applications filed in that year was 1,149, a number below the peak
numbers of more than 1,200 plant patents filed in each of 2004,
2005, and 2008.56 Such figures suggest that plant patents are
currently on course to become an increasingly unimportant part of
the U.S. patent system, perhaps at least a partial result of the
U.S. Supreme Court’s confirmation in 2001 that, generally
speaking, standard utility patents are a viable alternative for
achieving patent protection for plants.57 Thus, trends with
respect to plant patents might properly be viewed as a no-morethan-incidental aspect of the overall story of relatively relentless
growth in patent-issuance rates.
With respect to this dominant story of relentless growth, one
might wonder about the representativeness of the various “halfcentury numbers” recited above. The use of aggregate figures for
decades was meant to guard against such concerns, but graphical
illustrations of the number of U.S. patents issued per year might be
the easiest way to confirm both (1) the generally accelerating nature
of growth over time and (2) the general slowdown of percentage
growth in the twentieth century. To this end, Figure 1 graphs as a
function of time the total number of U.S. patents issued—including
any utility, design, and plant patents issued—in each year from
58
1790 through 2012. The generally accelerating nature of patent
issuance, which I have previously discussed and sought at least
54. Id. at 1.
55. Id. at 1–5.
56. Id.
57. See J.E.M. Ag Supply, Inc. v. Pioneer Hi-Bred Int’l, Inc., 534 U.S. 124, 127
(2001) (“We hold that utility patents may be issued for plants.”).
58. See infra Figure 1.
Do Not Delete
468
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
partially to explain,59 is indicated by the long-term tendency of the
resulting curve to slope upward at an ever sharper tilt as one moves
to the right—i.e., as time elapses.
Figure 1: Numbers of U.S. Patents Issued Per Year
from 1790 through 2012
6QVCN75
2CVGPVU+UUWGF
2GT;GCT
752CVGPVU
+UUWGFVQ75
4GUKFGPVU2GT
;GCT
Figure 2 sheds further light on the nature of this acceleration
by graphing as a function of time the natural logarithm of the total
number of U.S. patents issued in a given year.60 The slope of the total
patents graphed in Figure 2 basically reflects the percentage change
61
in the numbers of patents issued annually. Consequently, the
generally declining nature of this slope indicates that, even though
the number of patents issued in a given year is still tending to
increase (a fact indicated by the upward-curving nature of the graph
in Figure 1), the percentage increase in the number of patents issued
from year to year has tended to be substantially lower since 1900
than it was before 1900. In short, although growth in the absolute
number of patents issued each year is still accelerating, the
percentage growth in the rate of patent issuance has generally
slowed.62 Further, recent rates of percentage growth—i.e.,
59. See Golden, supra note 47, at 90–96 (analyzing the time-dependent behavior of the
total number of U.S. patents issued since 1790).
60. See infra Figure 2.
61. See GEORGE B. THOMAS, JR. & ROSS L. FINNEY, CALCULUS AND ANALYTIC GEOMETRY
395 (6th ed. 1984) (noting that the derivative of the natural logarithm of u with respect to x is
(1/u)(du/dx)).
62. Consider, for example, that the total number of patents issued to U.S. residents was
(a) 112,621 for the period from 1898 through 1902; (b) 202,256 for the period from 1923
through 1927; and (c) 475,921 for the period from 1998 through 2002. U.S. PATENT ACTIVITY,
supra note 2, at 1–5. On its face, the last figure—nearly half a million patents issued—might
seem very impressive as it is more than double the corresponding 1923–1927 figure. But the
Do Not Delete
11/24/2013 11:28 AM
2013]
PROLIFERATING PATENTS
469
percentage growth rates in patenting since the mid-1980s—do show
an uptick relative to the immediately preceding decade but do not
appear terribly anomalous for the post-1900 “long century,” at least
outside periods of historically great slowdown, conflict, or
dislocation such as the 1930s, 1940s, 1970s, and early 1980s.63
Figure 2: Natural Logarithms of the Numbers of U.S.
Patents Issued Per Year
0CVWTCN
.QICTKVJOQH
6QVCN0WODGTQH
752CVGPVU
+UUWGF2GT;GCT
0CVWTCN
.QICTKVJOQH75
2CVGPVU+UUWGFVQ
754GUKFGPVU
2GT;GCT
1923–1927 figure is itself nearly double the 1898–1902 figure. More specifically, the 1923–1927
figure is about 1.8 times the 1898–1902 figure even though only 25 years separated the
start points for the five-year spans of 1898–1902 and 1923–1927. If the rate of patenting
by U.S. residents had increased at the same percentage rate in the 75 years after 1923–
1927 as it increased in the 25 years between 1898–1902 and 1923–1927, the number of
patents issued to U.S. residents from 1998 through 2002 would have been approximately
equal to the 1923–1927 total multiplied by the cube of 1.8—or about 1.2 million. Thus, the
actual rate of patenting by U.S. residents from 1998 through 2002 was less than half the
rate of patenting that would have occurred if the average percentage growth rate between
1898–1902 and 1923–1927 had been maintained.
63. See supra note 9 and accompanying text. Despite seemingly apparent questions
about the historical representativeness of a period running from 1930 to 1982, a period
whose endpoints roughly coincide with (1) the beginning of the twentieth century’s first
major slowdown in patenting and (2) the end of the twentieth century’s second major
slowdown in patenting, Adam Jaffe and Josh Lerner have used this period as a foil for
recent growth in patenting, which they characterize as a “Patent Explosion.” ADAM B.
JAFFE & JOSH LERNER, INNOVATION AND ITS DISCONTENTS: HOW OUR BROKEN PATENT
SYSTEM IS ENDANGERING INNOVATION AND PROGRESS, AND WHAT TO DO ABOUT IT 11
(2004) (comparing the percentage growth rate from 1983 through 2002 to that “from 1930
until 1982”).
Do Not Delete
470
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
B. Patenting by Non-U.S. Residents and the Uptick in
Patenting’s Percentage Growth
Despite the relative historical modesty of patenting’s recent
percentage growth, some commentary has suggested that this
development signals that the patent system has spun out of
control, in substantial part because of failures of the USPTO.64
Reference to patenting’s long-term trends casts some doubt on
this inference. As is commonly the case when one tries to draw
inferences from numbers or numerical trends, a key question is
what baseline we should use as a reference point for current
65
growth rates. Thus, as suggested above, if one views the
patenting rates of the 1970s as more of an anomaly than a proper
norm, the apparent uptick in patenting rates starting in the mid1980s looks more like a “restoration” than a revolution, a shift
bringing overall patenting up toward where we might have
expected it to be if the trends of the pre-Great Depression
twentieth century—or even those of the more recent post-World
War II 1950s and 1960s—had continued.66
Moreover, much of the uptick in percentage growth of U.S.
patenting over the last half century or so seems likely to be more
a product of greater internationalization of patenting than of any
pathology peculiar to the U.S. patent system in the past few
decades. If we remove from our tabulation U.S. patents issued to
nonresidents, much of the recent growth in patenting disappears.
Consider the following numbers for the total numbers of patents
issued to U.S. residents or non-U.S. residents during our
established half-century periods: 911 from 1800 through 1809;
20,586 from 1850 to 1859; 313,095 from 1900 to 1909; 457,267
67
from 1950 to 1959; and 1,822,253 from 2000 to 2009. These
figures correspond to multiplicative factors for half-century
growth of 22.6, 15.2, 1.5, and 4.0, respectively. Consider now the
corresponding total numbers of patents issued to U.S. residents
only: 911 from 1800 to 1809; 20,276 from 1850 to 1859; 277,884
from 1900 to 1909; 400,844 from 1950 to 1959; and 948,198 from
2000 to 2009.68 These figures correspond to multiplicative factors
for successive half-century growth of 22.3, 13.7, 1.4, and 2.4,
respectively. Although the more than doubling of U.S. patenting
by U.S. residents since the 1950s is impressive, the fact that this
64. See id. at 11–13 (“Much of the problem stems from the organization of the
patent office itself.”).
65. See supra note 31 and accompanying text.
66. See supra note 63 and accompanying text.
67. U.S. PATENT ACTIVITY, supra note 2, at 1–5.
68. Id.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
471
multiplicative factor is greater than that associated with the first
half of the twentieth century might have something to do with
the occurrence of two world wars and a “Great Depression”
during the period from 1914 to 1945.69 Further, this
multiplication of domestic patenting activity by 2.4 pales in
comparison to the fourfold multiplication of U.S. patenting
overall. This fourfold multiplication is necessarily the product of
a more than fourfold multiplication of U.S. patenting by foreign
residents. Indeed, foreign residents experienced a more than
fifteenfold multiplication in the number of U.S. patents issued to
them from the 1950s to the first decade of the 2000s. Foreign
residents obtained only 56,423 U.S. patents in the 1950s but
874,055 U.S. patents from 2000 to 2009.70 Foreign patenting has
grown so rapidly that it now accounts for the majority of newly
issued U.S. patents: in each year from 2008 to 2012, the majority
71
of U.S. patents issued to foreign residents.
Because of the importance of foreign patenting to the
modern story of U.S. patent law, Figures 1 and 2 graph as a
function of time not only the total number of U.S. patents issued
per year or the logarithm of that number, but also the total
number of U.S. patents issued to U.S. residents per year or the
72
logarithm of that number. In each figure, the gap between these
69. See supra notes 39–40 and accompanying text.
70. U.S. PATENT ACTIVITY, supra note 2, at 1–2. Growth in the number of patents
issued to foreign residents has been notably uneven, in part because of changes in U.S.
law. Although the U.S. Patent Act of 1790 did not discriminate against non-citizens, B.
ZORINA KHAN, THE DEMOCRATIZATION OF INVENTION: PATENTS AND COPYRIGHTS IN
AMERICAN ECONOMIC DEVELOPMENT, 1790–1920, at 57 (2005), the Patent Act of 1793 did
not allow foreign citizens to obtain U.S. patents, and the Acts of 1800 and 1832 only
opened the door for foreign citizens who had lived in the United States for at least a year.
Jennifer A. Crane, Riding the Tiger: A Comparison of Intellectual Property Rights in the
United States and the People’s Republic of China, 7 CHI.-KENT J. INTELL. PROP. 95, 103–
04 (2008) (noting that the U.S. amended its patent laws “in 1800 to allow foreign citizens
who resided in the United States for two years to petition for a patent” and in 1832 “to
enable foreig[n] citizens who lived in the United States for twelve months and who swore
an oath of intention for U.S. citizenship to petition”). “In 1836, the stipulations on
citizenship or residency were removed, but were replaced with discriminatory patent
fees,” which were ultimately repealed in 1861. KHAN, supra, at 57. Unsurprisingly,
therefore, the USPTO does not report numbers for patents issued to foreign residents
until 1836. U.S. PATENT ACTIVITY, supra note 2, at 4. For the decade of the 1850s, when
the discriminatory fees for foreign residents were still in place, the total number of U.S.
patents reported as issued to foreign residents is 310. Id. at 3–4. For the period from 1900
to 1909, after the repeal of those discriminatory fees, the total number—35,211—is more
than 110 times higher. Id. at 3. Foreign patenting fell substantially during the period
marked by the Great Depression and World War II. The number of U.S. patents issued to
foreign residents fell from a then historical peak of 7,376 in 1932 to a mid-century low of
1,617 in 1947, a number lower than that of any year since 1888. Id. at 2–3.
71. U.S. PATENT ACTIVITY, supra note 2, at 1.
72. See supra Figures 1, 2.
Do Not Delete
472
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
two graphs reflects the increasingly substantial contribution of
foreign patenting to the U.S. patent picture. If one removes
patents issued to foreign residents from the story, the growth in
patenting over the last few decades looks far less explosive,
indeed arguably quite tame. If one suspects that the effective
internationalization of U.S. patenting is likely to be more of a
reflection of globalization trends and economic development
elsewhere than of the USPTO’s peculiar dysfunctionality,73 the
notion that the recent “explosion” in U.S. patenting is
predominantly a product of system breakdown becomes subject to
additional doubt.
C. An Overwhelming “Patent Tax”? Patent Issuance Per Dollar
of GDP
But even if the system has not yet broken down, could it be
on the verge of collapse? Does the historically relentless tendency
toward increased patenting—albeit at generally reduced
percentage rates over time—mean that we are now suffering
74
under an overwhelming “patent tax”?
As a first cut at addressing the second question, we might
consider how the number of patents issued each year has
compared to the United States’ GDP as a function of time. This
effectively corresponds to considering the average “tax” that an
individual patent imposes on the U.S. economy to have stayed
relatively constant in real dollars and, effectively, to have been
paid at the time of patent issuance. In addition to making
objections that one should use cumulative numbers with lags
from issuance to represent an effective “patent tax,” one might
object that individual patents are likely to be worth more today
and thus impose a higher per-patent “tax,” even in real dollars,
in part because larger real GDPs correspond with larger
effective markets and also larger markets for more expensive
75
goods and services. As I only wish to make the patents-toGDP comparison primarily for the rough qualitative purpose of
73. See Mabey, supra note 14, at 221 (“With the shift towards international
innovation and market globalization, the demand for patent protection both domestically
and abroad has skyrocketed.”).
74. See John M. Golden, Patent Privateers: Private Enforcement’s Historical
Survivors, 26 HARV. J.L. & TECH. 545, 589 (2013) (“[A]s Macauley noted long ago with
respect to copyright, patents effectively act as a private-taxation regime in which
patentees use litigation or its threat to effect transfers of wealth.”).
75. See BAUMOL, supra note 16, at 5 (“This unprecedented productivity
growth . . . has all but eliminated famine in wealthy countries, created technology
unimaginable in earlier eras, given us ever rising standards of living, and greatly reduced
the rising costs of health care, education, and other important services.”).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
473
stimulating thought about the nature of current concerns with
the patent system, I do not believe I need to try to devise the
best patents-to-GDP comparison—an exercise that would
likely require answering difficult empirical questions.76
Figure 3 presents the results of graphing the number of patents
issued per year divided by U.S. GDP in billions of chained real
77
dollars from 1929 through 2012.
Figure 3: Ratio of U.S. Patents Issued to Real GDP
6QVCN75
2CVGPVU+UUWGF
2GT;GCT&KXKFGF
D[4GCN)&2
6QVCN75
2CVGPVU+UUWGFVQ
754GUKFGPVU
2GT;GCT&KXKFGF
D[4GCN)&2
76. If one considers damage awards in patent-infringement cases, work by Michael
Mazzeo, Jonathan Hillel, and Samantha Zyontz suggests that mean patent damage
awards per year might have increased by about a factor of three between the mid-1990s
and the end of the last decade, but that median patent damage awards—which one might
think are at least somewhat more indicative of the value of the great mass of patents
never litigated at all—might have decreased over the same time period. Michael J.
Mazzeo, Jonathan Hillel & Samantha Zyontz, Predicting the “Unpredictable”: An
Empirical Analysis of U.S. Patent Infringement Awards 13 fig.3 (2012) (unpublished
manuscript), available at http://ssrn.com/abstract=2164787. Compare Jean O. Lanjouw &
Mark Schankerman, Characteristics of Patent Litigation: A Window on Competition, 32
RAND J. ECON. 129, 146–147 & tbl.6 (2001) (showing typical litigation probabilities of
only about a couple percent for U.S. owned corporate patents), with Mark A. Lemley,
Rational Ignorance at the Patent Office, 95 NW. U. L. REV. 1495, 1501 (2001) (describing
grounds for “estimat[ing] that at most only about two percent of all patents are ever
litigated”). Which result or what combination of these results is most representative of
overall trends in the size of the “patent tax” are questions beyond this Article’s scope.
77. See infra Figure 3. The figures used represent annual U.S. GDP from 1929
through 2012 in billions of chained 2009 dollars—i.e., inflation-adjusted GDP values from
1929 through 2012—as indicated by the U.S. Department of Commerce’s Bureau of
Economic Analysis. Current-Dollar and “Real” GDP, supra note 6 (providing various GDP
figures). I only use figures from 1929 or later because the relevant table from the Bureau
does not go back further than 1929. See id.
Do Not Delete
474
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
Figure 3 at least suggests that, if one wishes to contend
that patents “overtax” social GDP, one might need to contend
that individual patents are now on average more burdensome
in real dollars than they were, say, fifty years ago or,
alternatively, that such overtaxation has long been the case. In
any event, the sense of relative tameness to the recent uptick
in patenting, already suggested by Parts II.A and II.B, seems
even more pronounced in Figure 3. This figure shows a rather
dramatic fall in the number of patents per real GDP dollar
from the 1930s through the 1970s. In the last forty years,
there has been a relatively slow but small rise in the number
of U.S. patents issued annually to U.S. residents per real GDP
dollar, with the result being a return to values for U.S.
resident patents per real GDP dollar that were typical of the
the mid-1970s. The growth in foreign patenting has generated
an approximate doubling of the overall number of U.S. patents
issued annually per real GDP dollar, but even this more
dramatic recent rise has only returned the number of U.S.
patents issued annually per real GDP dollar to a size typical of
the early 1950s, rather than some unprecedented historical
height. As I have previously noted, for several decades now,
U.S. patenting per real GDP dollar has overall been
remarkably flat compared to raw patenting rates.78
Indeed, the notion that, in light of modern economies’ size
and investments in innovation, modern rates of patenting are
not surprisingly out of control finds further support in Figure
79
4 below. Figure 4 shows the ratio of U.S. patents issued in
any given year to the number of U.S. full-time equivalent
employees engaged in research and development during that
year according to National Science Foundation figures
reported annually in the U.S. Census Bureau’s Statistical
Abstract of the United States.80 As with the patents-per-GDP
figures, recent patents-per-R&D-employee figures are actually
81
relatively low, historically speaking. Although there has been
a trend toward increases in the yearly ratios in the past couple
82
decades, these increases do not appear facially overwhelming.
78. See John M. Golden, “Patent Trolls” and Patent Remedies, 85 TEX. L. REV. 2111,
2111 n.3 (2007) (comparing “average number[s] of issued patents per billions of real gross
domestic product”).
79. See infra Figure 4.
80. Annual census information is available at Statistical Abstracts, U.S. CENSUS
BUREAU, U.S. DEP’T OF COMMERCE, http://www.census.gov/prod/www/statistical_abstract.
html.
81. See infra Figure 4.
82. See infra Figure 4.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
475
Figure 4: Ratio of U.S. Patents to 1,000 U.S. R&D
Employees
In short, if one only looked at Figures 3 and 4, perhaps adding
a glance at Figure 2’s illustration of generally slowed growth in the
logarithm of number of patents issued annually, one might wonder
why there is any fuss at all about supposedly excessive patenting.
Indeed, one might wonder why our productivity in generating
patents has apparently slowed and what might be done to increase
current patent application and issuance rates. As Part III will
argue, this does not mean there are no problems with the patent
system—or more specifically that there is no truth to contentions
that the USPTO is currently issuing too many patents. The basic
point is that because of the need to consider how patent issuance
rates should be normalized to account for other developments—i.e.,
to consider to what baselines patent issuance rates should be
compared—one should not feel confident that citation of raw patent
issuance numbers proves much of anything. As a logarithmic graph
for which slope indicates percentage changes over immediately
prior years, Figure 2 implicitly makes prior years’ issuance rates
the baseline for normalization. Figures 3 and 4 more explicitly
make values for real GDP and numbers of R&D employees
baselines for normalization. I do not claim to have used the limited
space available in this short piece to have provided a decisive
argument for any of these potential normalization schemes. But I
hope their illustrative use is at least suggestive of a lack of easy
answers about whether the last few decades have witnessed an
unprecedented and obviously dysfunctional “patent explosion.”83
83.
JAFFE & LERNER, supra note 63, at 11.
Do Not Delete
476
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
III. PROLIFERATING PATENTS’ RELENTLESS PRESSURE FOR
ADMINISTRATIVE REFORM
A. Patent Law’s Labor Problem
Part II suggests that the present-day “patent crisis” might
not be as sharp or fearsome as sometimes thought. Nonetheless,
the historically relentless acceleration of patenting that Part II
highlights also points to a very real problem. The continual
acceleration of patenting rates strains both public and private
actors’ capacities to perform all the various activities necessary
for a patent system to function effectively. Many of these tasks—
such as drafting patent applications, processing and evaluating
such applications, making and adjudicating challenges to patent
validity or enforceability, fulfilling requests for patent
“clearance” (i.e., confirmation that relevant patent claims are
invalid or not infringed), negotiating patent licenses, and
identifying and taking legal action against potential patent
infringement—all tend to require human labor for which the
“Silicon Revolution” has not yet generated a workable
technological equivalent.84 In short, the patent system, like legal
services more generally, suffers from a form of the Baumol “cost
disease” under which “handicraft” sectors struggle to keep up
with the productivity gains in endeavors more subject to
standardization or automation—or, at least, benefiting more
greatly from standardization or automation.85 Indeed, the patent
system’s cost disease is arguably worse than the normal case, in
which the focus is commonly on how the cost disease attaches to
set amounts of work. For the USPTO, increased productivity in
other economic sectors not only tends to have an inflationary
effect on USPTO costs but also, through its association with
increased innovation, helps increase the actual amount of work
that the USPTO needs to do.
Relatedly, as the numbers of patent applications, patents
and resultant clearance questions, licensing negotiations, or
lawsuits increase, the system’s demands on a relatively scarce
supply of people with appropriate scientific, technological, or
legal backgrounds increase. Thus, not only can the patent system
act as a relative economic drag by struggling to keep up with
84. Cf. BAUMOL, supra note 16, at 22–24 (discussing how many tasks demand
“substantial personal attention” and thus do not share as fully as others in productivity
gains from automation).
85. Id. (listing “legal services” as one of “the stagnant services” and contrasting
these services to services such as telecommunications that have experienced much greater
productivity growth (emphasis omitted)).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
477
productivity gains in the innovation that it is meant to support, it
can also impose a sort of “diversion of labor” cost on the economy,
pulling skilled labor away from economic sectors with greater
opportunities for growth in productivity.86 Figure 5, which plots
the number of U.S. patents issued per year divided by the size of
that year’s U.S. civilian workforce,87 gives some sense of the
88
“patent labor tax” that likely comes from having to devote more
and more human labor to generating, studying, licensing, and
litigating the ever-growing number of patents that issue.
Figure 5: Ratio of U.S. Patents Issued to Civilian
Workforce89
6QVCN0WODGTQH
752CVGPVU
+UUWGF2GT;GCT
&KXKFGFD[5K\GQH
75%KXKNKCP
9QTMHQTEGKP
6JQWUCPFU
But Figure 5 likely understates the likely rate of growth in
the “patent labor tax.” By simply using raw patent numbers,
Figure 5 does not reflect apparently increasing complexity in
patents and patent applications themselves. Even aside from the
arguably greater complexity or difficulty of some of the
underlying technology appearing in modern utility patent
86. Cf. U.S. PATENT & TRADEMARK OFFICE, supra note 15, at 12 (outlining the
USPTO’s strategy to hire more patent examiners by targeting science and engineering
graduates as well as “experienced IP professionals” such as patent attorneys, patent
agents, former examiners, and skilled technologists).
87. See infra Figure 5.
88. See infra Part III.A (describing the “patent labor tax”).
89. The figures for the number of people in the U.S. civilian workforce for each year
from 1947 through 2012 come from the U.S. Department of Labor’s Bureau of Labor
Statistics and are available at http://www.bls.gov/cps/cpsaat01.htm.
Do Not Delete
478
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
applications,90 those applications tend now to have more pages
and more individual claims than they did in times past.91
Consequently, one might expect that a representative individual
application today makes more demands on human labor than a
representative individual application in the past. Unless we are
improving the speed with which we can process complexity or
difficulty, the “patent labor tax” can therefore be expected to be
increasing not only as a result of increases in sheer numbers of
patents and patent applications but also as a result of increases
in the complexity or difficulty of processing individual
applications themselves.
Along these lines, one might conjecture that growth in the
number of U.S. patent examiners is likely to reflect both these
contributions to the “patent labor tax.” Other things being
equal, the United States can be expected both to need more
examiners to process more patents and to need more
examiners to process a constant number of patents marked by
increased complexity. Figure 6 shows that, since the
92
establishment of the “modern” examination system in 1836,
the number of U.S. patent examiners, as indicated in patent
93
office reports, has indeed grown prodigiously and, like the
90.
Cf. Benjamin F. Jones, As Science Evolves, How Can Science Policy?, 11
INNOVATION POL’ Y & ECON. 103, 106–07 (Josh Lerner & Scott Stern eds., 2011)
(describing how “the increasing narrowness of expertise and the shift toward team
production in science” has generated “evaluation challenges” that seem likely to
produce “increasing difficulties in allocating intellectual property rights”).
91.
Mabey, supra note 14, at 231 (“The length of applications has grown both
constantly and considerably over the years. The number of claims per application has
also skyrocketed.” (footnote omitted)); see also John R. Allison & Mark A. Lemley,
The Growing Complexity of the United States Patent System, 82 B.U. L. REV. 77, 79
(2002) (“By almost any measure—subject matter, time in prosecution, number of
prior art references cited, number of claims, number of continuation applications
filed, number of inventors—the patents issued in the late 1990s are more complex
than those issued in the 1970s.”).
92.
See infra notes 116–17 and accompanying text (noting the reinstitution of
pre-issuance examination in 1836); see also MERGES & DUFFY, supra note 32, at 8
(describing the U.S. patent system as “not grow[ing] to its full stature until the 1836
revision, when a formal system of examination, with professional examiners, was
substituted for the pro forma registration system of the 1793 Act”).
93. Numbers for U.S. patent examiners were generally obtained from annual
reports of the USPTO or its relevant institutional predecessor, the U.S. Patent Office.
Additional numbers for middle years of the nineteenth century—i.e., for years from
1836 to 1875—were obtained from Andrew P. Morriss and Craig Allen Nard. Andrew
P Morriss & Craig Allen Nard, Institutional Choice & Interest Groups in the
Development of American Patent Law: 1790–1865, 19 SUP. CT. ECON. REV. 143, 149
n.16 (2011) (discussing numbers for patent examiners in 1836, 1837, 1839, 1848,
1856, and 1875). Some of Morriss and Nard’s numbers might be overestimates in that
at least their 1836 number appears to reflect the number of examiners (namely, one)
that were then statutorily authorized, a number that presumably represented a
ceiling but not necessarily a floor for the number of examiners actually employed. See
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
479
number of issued patents, has tended to grow at an
accelerating rate.94 Less predictably and thus more
remarkably, Figure 7, which graphs the logarithm of the
number of U.S. patent examiners over time, shows that, since
1982, the year in which the U.S. Court of Appeals for the
Federal Circuit was created,95 the rate of percentage growth in
the number of U.S. patent examiners—i.e., the slope of the
logarithmic curve—has tended to reach levels more
characteristic of the nineteenth century than the twentieth.96
This aspect of the growth in numbers of patent examiners
contrasts with the corresponding aspect of the growth in
numbers of issued patents: as Figure 2 shows, percentage
growth in numbers of issued patents has undergone something
of a revival in the wake of the Federal Circuit’s creation, but
percentage growth rates for numbers of issued patents have
tended not to reach nineteenth-century levels.97 The
comparison between Figures 2 and 7 suggests that, instead of
focusing so much on how much patenting has increased over
the last few decades, commentators might more properly focus
on the comparatively more runaway growth in the patent
bureaucracy.
id. (“The 1836 Patent Act authorized the Commissioner of Patents to appoint an
‘Examiner of Patents,’ a machinist, and two draftsmen.”). For purposes here, the
possibility that some of Morriss and Nard’s numbers are mere ceilings seems to have little
importance: in any case, the number of examiners shown for the period up through 1875
will be quite close to zero on the vertical scale of Figure 6. A more serious problem could
be that the manner of counting “examiners” or nature of the title or job description of an
“examiner” has likely not been entirely consistent through time. Even recently, the
USPTO has shifted at least terminologically speaking between reporting numbers of
“patent professionals” and numbers of “patent examiners.” Compare 2012 USPTO
PERFORMANCE REPORT, supra note 13, at 203 tab.29 (reporting numbers of “Patent
Examiners” in terms of “[n]umber of people on-board” at the end of fiscal years 2008
through 2012), with ASSISTANT SECRETARY OF COMMERCE AND COMMISSIONER OF
PATENTS AND TRADEMARKS, ANNUAL REPORT: FISCAL YEAR ‘93, at 50 tbl.3 (reporting
numbers of “Patent Professional[s]” corresponding to “[t]otal paid staff years, excluding
overtime hours worked,” at the end of fiscal years 1987 through 1993). For purposes here,
I have treated both sets of numbers as indicating the number of patent examiners
representative for that year. In any event, whatever precision or lack of precise
comparability through time that results seems unlikely to alter very substantially the
general qualitative picture suggested by Figures 6 and 7, including the upward turn in
examiner numbers in the early 1980s.
94. See infra Figure 6; see also supra Figure 1.
95. See infra note 120.
96. See infra Figure 7.
97. See supra Figure 2.
Do Not Delete
480
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
Figure 6: Number of U.S. Patent Examiners from 1836
to 2012
Figure 7: Natural Logarithm of Number of U.S. Patent
Examiners from 1836 to 2012
Figures 8 and 9 further serve to drive home this point
regarding the patent bureaucracy’s prodigious growth. By
plotting the ratio between the number of U.S. patent examiners
and the number of U.S. patents issued in years since 1836,
Figure 8 shows how growth in the number of U.S. patent
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
481
examiners has historically tended, generally speaking, to outstrip
growth in the number of issued patents.98 By plotting the ratio
between the number of U.S. patent examiners and the number of
thousands of full-time equivalent research-and-development
employees in the United States, Figure 9 shows that, at least
since the late 1980s, the number of examiners has typically
grown more rapidly than overall engagement of personnel in
research and development.99
Figure 8: Number of U.S. Patent Examiners Per New
U.S. Patent
98. See infra Figure 8. One might infer that this graph suggests that the USPTO
has suffered not only from “stagnant” productivity but, in fact, from declining
productivity. But such an inference might well be unfair. As discussed in the text, the
complexity of applications—at least in the case of applications for utility patents—
appears to have increased with time. See supra text accompanying notes 90–91; infra text
accompanying note 129. Further, patentability requirements, such as nonobviousness,
have emerged and become more refined, thereby likely adding to the demands placed on
an examiner in the typical individual case. See John F. Duffy, Inventing Invention: A Case
Study of Legal Innovation, 86 TEX. L. REV. 1, 19 (2007) (“The relatively recent
development of the nonobviousness doctrine is explained in part by the overall youth of
the entire field of patent law.”); Gregory Mandel, The Non-Obvious Problem: How the
Indeterminate Nonobviousness Standard Produces Excessive Patent Grants, 42 U.C. DAVIS
L. REV. 57, 81–83 (2008). Such developments might help explain how, during a decade in
which the USPTO “more than doubled” its number of examiners, relevant times for
processing patent applications increased by forty percent. See James D. Ridgway,
Equitable Power in the Time of Budget Austerity: The Problem of Judicial Remedies for
Unconstitutional Delays in Claims Processing by Federal Agencies, 64 ADMIN. L. REV. 57,
66 (2012) (discussing the USPTO’s travails in the context of other agencies’ claimsprocessing challenges).
99. See infra Figure 9.
Do Not Delete
482
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
Figure 9: Ratio of U.S. Patent Examiners to 1,000 U.S.
R&D Employees
Of course, the rise in the number of USPTO examiners is
only one part of a larger growth in the patent system’s public and
private personnel. Patents require not only patent examiners but
patent drafters, and patents asserted in or out of court commonly
require lawyers to be involved in the processes of licensing or
litigation. An article by John Barton in 2000 indicated that
membership in the American Bar Association’s Section on
Patents, Trademarks, and Copyright had been “growing faster
100
than the amount of research.” In a like vein the rate at which
the USPTO issued new patents multiplied approximately by 3.5
101
from 1980 to 2010,
the rate at which the USPTO issued
registration numbers to new patent attorneys or agents
multiplied by about 4.3.102 As such figures highlight, the patent
system’s personnel demands extend far beyond the confines of
USPTO offices, and this has significant implications for the
100. John H. Barton, Reforming the Patent System, 287 SCIENCE 1933, 1933 & n.1
(2000) (concluding from the ABA membership numbers that “[t]he number of intellectual
property lawyers in the United States is growing faster than the amount of research”).
101. The USPTO issued an average of 64,369 total patents per year during the five
years from 1978 through 1982, whereas the USPTO issued an average of 228,419 total
patents per year during the five years from 2008 through 2012. See U.S. PATENT
ACTIVITY, supra note 2, at 1.
102. According to figures from the USPTO Office of Enrollment and Discipline on file
with the Author, the USPTO issued an average of 425 new registration numbers per year
during the five years from 1978 through 1982, whereas it issued an average of 1,838
registration numbers per year during the five years from 2008 through 2012.
Interestingly, however, the real take-off point for issuance of greater numbers of
registration numbers per year appears to have occurred no earlier than the late 1980s and
thus to have lagged the earlier take-off points for patenting and numbers of patent
examiners, which, as Figures 2 and 6 indicate, appear to have occurred in the early 1980s.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
483
system’s overall costs and the sustainability of its current growth
trajectory.
If we imagine that the most crucial scarce resource for which
proliferating patents are competing is not capital but instead
labor—whether the labor of patent prosecutors, patent
examiners, attorneys advising on whether a planned course of
action is likely to infringe another’s patent rights, attorneys and
businesspeople trying to agree on licensing terms, or litigators,
judges, and juries looking to resolve patent disputes in the
courts—then I think we quickly come to confront what seems to
me one of the more undeniable dilemmas for the modern patent
system: whether we can continue to reconfigure the system’s
institutions and processes to sustain something like the
established trajectory of continually accelerating growth, a
trajectory that reflects the likely continually accelerating growth
of innovation itself. If we cannot do this, we will likely need to
contain the patent system’s growth in potentially draconian
ways—for example, by significantly weakening or limiting patent
rights’ desirability or accessibility through severely truncated
definitions of patentable subject matter, a significantly higher
threshold for an invention to be considered patentably
nonobvious, systematically narrower claim constructions, heavily
diluted remedies for patent infringement, much higher fees for
patenting or maintaining patents, or delays in processing
applications or infringement claims that are so long that they
much more substantially deter use of the system altogether.
Some readers might champion one or another or perhaps even all
of the latter set of potential reforms. But as this Article is for a
conference focused on the institutions for administering our
intellectual property laws, I will focus here on the former
alternative—adapting patent-law administration to sustain the
increasing weight of applications that it has been receiving and,
other things being equal, should continue to expect.
B. Patent Law’s History of Administrative Change and Growth
Patent law’s existing administrative challenges are far from
being entirely new. Reform of patent law’s institutions has
occurred repeatedly throughout history,103 and a significant
number of these reforms have been designed to respond to the
ever more crushing weight that continually accelerating
patenting places on patent-law institutions. This was true from
the very start. The U.S. Patent Act of 1790 adopted an ambitious,
103.
See infra notes 98–108 and accompanying text.
Do Not Delete
484
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
high-level approach to the granting of patents: patents would
issue on approval by a majority of a “patent board” composed of
the Secretary of State, the Secretary of War, and the Attorney
General.104 Predictably (at least in historical hindsight), these
officials found themselves short on time and resources to review
dozens of incoming applications.105
Congress’s solution, implemented through the U.S. Patent
Act of 1793, was to eliminate substantive examination of patents
before issuance—i.e., to turn the U.S. patent system into a
registration system.106 Generally speaking, the Secretary of State
was empowered—albeit not explicitly required107—to issue a
patent once a petitioner had fulfilled a set of basic ministerial
requirements.108 Policing the validity and scope of patents was
substantially left to post-issuance action by the courts.109
The 1793 Act’s resolution of the patent system’s bureaucratic
problems—achieved by largely eliminating the need for
substantial pre-issuance bureaucracy—worked for a time.110 But
with the number of patents issued per year increasing more than
tenfold over the next few decades,111 strains inevitably showed.
Once several hundreds of patents were issuing each year, even a
residual ministerial requirement requiring the mere signing of
each patent by the President, the Secretary of State, and the
104. EDWARD C. WALTERSCHEID, TO PROMOTE THE PROGRESS OF USEFUL ARTS:
AMERICAN PATENT LAW AND ADMINISTRATION, 1798–1836, at 168 (1998) (“Under the Act
of 1790 the power to issue patents was completely vested in three high officials . . . .”).
105.
[I]t was the dawning recognition by the members of the patent board, and
particularly by Jefferson, that they simply had insufficient time to properly
carry out the tasks assigned to them under the Act, that more than anything
else soon produced an understanding in the Congress that the Act of 1790 had to
be amended or in some manner changed to avoid having high government
officials responsible for the issuance of patents.
Id. at 195; see also Crane, supra note 70, at 103 (“In three years (1790–1793), 57 U.S.
patents were granted, 114 petitions were still pending, and dozens of petitions were
denied.”).
106. WALTERSCHEID, supra note 104, at 223 (“The most fundamental change [made
by the U.S. Patent Act of 1793] was of course from examination to registration by
elimination of the requirement that the patent board determine that an invention was
‘sufficiently useful and important’ for a patent to issue.”).
107. Id. at 224 (“[T]he Act merely made it lawful for the Secretary of State to issue a
patent [under specified circumstances]; it did not require him to do so.”).
108. Id. at 223 (describing patent issuance under the 1793 Act as “a pro forma
process dependent only on the completion of the required ministerial acts by the
petitioner for patent”).
109. Morriss & Nard, supra note 93, at 151 (“The 1793 Act shifted the ex ante
gatekeeper role performed by an examination to an ex post proceeding in the courts . . . .”).
110. Id.
111. See U.S. PATENT ACTIVITY, supra note 2, at 4–5 (listing numbers of utility
patents issued for individual years from 1790 on).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
485
Attorney General became a burdensome source of delay.112 With
more patents in private parties’ hands, patent litigation became
more common—and a more common source of complaint.113 An
1836 Senate committee report lamented that “[a] considerable
portion of all the patents granted [we]re worthless and void,” that
lawsuits were “daily increasing in an alarming degree,” and that
“frauds”—invalid patents obtained under false pretenses—had
“become extensive and serious.”114 The patent bar’s incipient
development and refinement of patent claims as a means to
115
clarify patent scope and validity apparently did not suffice to
relieve a perception of undue pressure on both courts and private
parties who either feared or experienced unjust accusations of
infringement.
The revised administrative solution, instituted through the
Patent Act of 1836 and largely continued to this day, was to
reinstate a form of pre-issuance examination while still retaining
significant roles for the courts both in policing patentability post116
issuance and in acting as primary interpreters of the Patent
Act itself.117 A revival of substantive examination of patents prior
to issuance was made feasible by placing responsibility for such
examination in the hands of “a specialized bureaucracy” designed
for the task, rather than making this task the responsibility of
Cabinet officers having other, largely disconnected duties.118
Comparatively speaking, the basic institutional structure of the
patent system has been only tweaked since then, prominent
tweaks including, of course, the successive creation of semispecialized appeals courts by Congress, the Court of Customs and
Patent Appeals,119 and its successor, the Court of Appeals for the
112. See WALTERSCHEID, supra note 104, at 422 (noting that, in 1836, the
Superintendent of Patents “readily acknowledged that patents were frequently delayed
for months due to the inability to obtain the signatures of the three top government
officials required, i.e., the President, the Secretary of State, and the Attorney General”).
113. 1836 Senate Committee Report, 18 J. PAT. OFF. SOC’Y 853, 857 (1936).
114. Id.
115. Morriss & Nard, supra note 93, at 153–54.
116. See Patent Act of 1836, ch. 357, sec. 7, 5 Stat. 117, 119–20 (discussing preissuance examination); John F. Duffy, The FCC and the Patent System: Progressive Ideals,
Jacksonian Realism, and the Technology of Regulation, 71 U. COLO. L. REV. 1071, 1138–
39 (2000) (“[I]n 1836, when Congress conferred a limited administrative power to the
Patent Office, it consciously designed institutional constraints—including an
administrative appeals process and judicial review . . . .”).
117. Cf. Duffy, supra note 116, at 1134 (“[U]nlike agencies in the twentieth century,
the Patent Office was given no power to issue substantive regulations—a limitation that
continues to have significant legal implications.”).
118. Morriss & Nard, supra note 93, at 159.
119. William J. Graham, The United States Court of Customs and Patent Appeals, Its
History, Functions and Jurisdiction, 14 J. PAT. OFF. SOC’Y. 932, 937, 939–40 (1932)
Do Not Delete
486
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
Federal Circuit.120 Some scholars have suggested that the recent
America Invents Act has shifted the balance of power between
the USPTO and the courts in ways that might amount to more
121
than another mere “tweak,” but the extent to which this is true
122
remains to be seen.
In any event, since 1836, the principal way that the U.S.
patent system has responded to the proliferation of patent rights
and would-be patent rights is to multiply the number of hands on
deck—most particularly, to multiply the number of examiners at
a percentage rate that overall has exceeded that for patents
123
themselves. Andrew Morriss and Craig Nard report that, after
the 1836 Act, the number of examiners—or at least, the number
of statutorily authorized examiners124—grew nearly onehundredfold in four decades: from one in 1836 to two in 1837,
four in 1839, eight in 1848, twenty-four in 1856, and ninety-six in
1875.125 As Figures 6 and 7 have suggested, generally speaking,
the numbers have kept growing since. Economic historian Zorina
(noting the establishment of a court of customs appeals in 1910 and the conferring of
jurisdiction over appeals from the U.S. Patent Office in 1929).
120. Paul D. Carrington & Paulina Orchard, The Federal Circuit: A Model for
Reform?, 78 GEO. WASH. L. REV. 575, 576 (2010) (“The Federal Circuit was established
through the Federal Courts Improvement Act of 1982 as a partial response to fifteen
years of agitation over the structure of the federal judiciary.” (footnote omitted)); Jeffrey
A. Lefstin, The Constitution of Patent Law: The Court of Customs and Patent Appeals and
the Shape of the Federal Circuit’s Jurisprudence, 43 LOY. L.A. L. REV. 843, 847 (2010)
(“The Federal Circuit was created from the merger of the Court of Customs and Patent
Appeals (CCPA) and the appellate division of the Court of Claims.”).
121. See, e.g., Leahy-Smith America Invents Act, Pub L. No. 112-29, 125 Stat. 284
(2011) (codified as amended in scattered sections of 35 U.S.C.); Arti K. Rai, Patent
Validity Across the Executive Branch: Ex Ante Foundations for Policy Development, 61
DUKE L.J. 1237, 1239, 1243 (2012) (“[T]o the extent [USPTO guidelines on topics such as
patentability] were to be implemented in a postgrant review proceeding that resembled a
formal adjudication, the strong form of deference enunciated by the Court in
Chevron . . . and its progeny would be applicable.”); Sarah Tran, Policy Tailors and the
Patent Office, 46 U.C. DAVIS L. REV. 487, 493 (2012) (“The freshly enacted Leahy-Smith
America Invents Act . . . gives the USPTO opportunities to play a much larger role in
policymaking than ever before.”); Melissa F. Wasserman, The Changing Guard of Patent
Law: Chevron Deference for the PTO, 54 WM. & MARY L. REV. 1959, 1965 (2013) (“This
Article concludes that the AIA rejects over two hundred years of court dominance in
patent policy by anointing the PTO as the chief expositor of substantive patent law
standards.”).
122. See John M. Golden, The USPTO’s Soft Power: Who Needs Chevron Deference?,
SMU L. REV. (forthcoming) (“Because the legislative history of the AIA prominently
featured the trouncing of a proposal to give the USPTO general rulemaking authority, I
am skeptical that the AIA has worked such a sea change.”).
123. See supra Figure 8 (showing that, generally speaking, the ratio between the
number of patent examiners and the number of newly issued patents has increased with
time).
124. See supra note 93.
125. See Morriss & Nard, supra note 93, at 149 n.16.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
487
Khan reports that in 1892 the U.S. Patent Office had “over six
hundred employees, including some two hundred specialized
technical examiners.”126 At the end of fiscal year 2012, which
was 120 years later, the USPTO had 7,935 patent examiners
and 11,531 employees overall.127 Thus, during a 120-year span
from 1892 to 2012 in which the number of applications filed
and patents issued both multiplied by less than a factor of
fifteen,128 the number of examiners multiplied by a factor of
approximately forty. Perhaps in part because of the increasing
129
complexity of patent applications or the technologies they
130
describe —but in any case despite the productivity-enhancing
benefits of the late twentieth century’s information-technology
revolution—individual U.S. patent examiners now apparently
process less than half as many applications per year as their
late nineteenth-century forebears.131
The “more hands on deck” approach seems destined to
continue indefinitely. The USPTO’s opening of four new
regional offices in or about Dallas, Denver, Detroit, and Silicon
Valley—all to complement the USPTO’s main office in
Alexandria, Virginia—seems partly designed to serve this
end.132
But
this
blunt,
expansion-of-the-patent-workforce
approach seems likely to be ultimately problematic. Polk
Wagner has noted that even the massive scaling up of the
USPTO workforce in recent decades appears to have had
relatively little success in addressing fundamental concerns
133
about patent quality. Moreover, quality aside, such scaling
up seems likely to become progressively more difficult and
perhaps ultimately undesirable if the goal is to at least keep
126.
KHAN, supra note 70, at 53.
127.
2012 USPTO P ERFORMANCE REPORT, supra note 13, at 10.
128.
See U.S. P ATENT ACTIVITY , supra note 2, at 1, 3 (listing patent statistics for
years 2012 and 1892).
129.
See supra note 91 and accompanying text.
130.
See supra note 90 and accompanying text.
131.
See supra Figure 8.
132.
Press Release, U.S. Patent & Trademark Office, U.S. Commerce
Department to Open Four Regional U.S. Patent Offices That Will Speed Up the
Patent Process and Help American Businesses Innovate, Grow, and Create Jobs (July
2, 2012), available at http://www.uspto.gov/news/pr/2012/12-40.jsp (contending that
additional regional offices “will help the USPTO attract talented IP experts
throughout the country who will work closely with entrepreneurs to process patent
applications, reduce the backlog of unexamined patents, and speed up the overall
process”).
133. R. Polk Wagner, Understanding Patent-Quality Mechanisms, 157 U. PENN. L. REV.
2135, 2158–59 (2009) (observing that the “hiring and training of thousands of new examiners”
have generated “only modest results, even at the most basic level”).
Do Not Delete
488
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
pace with a continually increasing number of incoming patent
applications.134 Mark Lemley has suggested that relatively high
turnover of USPTO examiners “means that the [US]PTO needs to
hire more than 1,000 examiners a year just to keep even with
attrition.”135 Unless the USPTO can significantly reduce attrition,
this replacement number will grow as the size of the examining
corps rises, presumably causing the USPTO to have to run
harder just to stay in place.136 Further, as the patent workforce
rises relative to the size of the overall workforce, the search for
additional qualified hands is only likely to become more
difficult. Silicon Valley firms have become notoriously
competitive in their pursuit of engineering talent—and
assertive in their pursuit of immigration reform to help expand
the practically available talent pool.137 If the USPTO continues
to
need
to
expand
proportionately—or
even
superproportionately—with the number of patents or patent
applications, and if the number of patents or applications
continues to grow at an accelerating rate, the workforce
demands for the USPTO can be expected to become more and
more difficult to fill. Even if the USPTO can somehow continue
to fill ever greater demand for examiner labor by opening more
regional offices or substantially increasing examiner salaries
and patent-associated fees to pay for those salaries, one might
wonder about the wisdom of an increasingly significant
diversion of labor into administration of the U.S. patent
system or increased fees that make the patent system less
accessible to undercapitalized innovators.
134. Id. at 2163 (“[M]erely keeping pace with increased filings will create an enormous
(and possibly unsustainable) drag on resources without clear evidence that such
expenditures will improve pendency, much less patent quality.”).
135. Mark A. Lemley, Fixing the Patent Office, 13 INNOVATION POL’Y & ECON. 83, 89
(2013); see also Mabey, supra note 14, at 256 (“[T]he office has one of the highest attrition
rates in the government.”).
136. Cf. Mabey, supra note 14, at 256–57 (describing the USPTO’s efforts to hire
1,200 examiners a year between 2007 and 2012 as a consequence of having “one of the
highest attrition rates in the government”); Grier, supra note 15, at 645 (observing that
patent-examiner attrition “is so severe that in some years the USPTO loses more
examiners than it hires”).
137. See generally Jennifer Martinez, Silicon Valley Readies All-Out Push for
Immigration Reform, HILL, Oct. 22, 2013, at 8 (“Silicon Valley groups and top executives
like Facebook CEO Mark Zuckerberg are planning a flurry of events and media
campaigns aimed at pressuring the House to vote on immigration bills before the end of
the year.”); Sarah McBride, Twitter Pays Engineer $10 Million as Silicon Valley Tussles
for Talent, REUTERS (Oct. 13, 2013), http://www.reuters.com/article/2013/10/13/net-ussiliconvalley-engineers-twitter-idUSBRE99C03R20131013 (“Welcome to Silicon Valley,
where a shortage of top engineering talent amid an explosion of venture capital-backed
start-ups is inflating paychecks. . . . Stories abound about the lengths to which employers
will go to attract engineering talent.”).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
489
Moreover, the USPTO’s workforce demands are not the only
“tax” on labor supply that the patent system generates.
Increasing numbers of patents and patent applications mean,
generally speaking, increased work for patent attorneys, patent
agents, and paralegals involved with the process of drafting them
and seeking their issuance through the USPTO. If the average
number of lawsuits per patent remains relatively stable or
increases, more patents will mean more litigation and thus more
work for litigators, judges, and juries. There is also the less
publicly visible transactional and counseling work of patent law:
for example, the work on patent licenses and assignments, the
work on assessing whether patent licenses are needed in the first
place, the strategic assessment of whether one’s patent
portfolio—or perhaps another’s patent portfolio that is ripe for
acquisition—suffices to protect core business interests.
In short, in private as well as public sectors, our evergrowing patent system can be expected to strain the capacity of
our society to supply in an efficient way the labor required to
administer and implement the system as currently configured.
We thus have a classic problem of “bureaucratic rationality.”138
How can we process such a large and increasing number of
claims in a way that cost-effectively advances social goals? The
suggestion above is that the “more hands on deck” response will
ultimately prove inadequate, even if it is not inadequate already.
The patent system needs to put greater emphasis on exploring
options other than sheer force-of-numbers responses. Some
preliminary thoughts on potential directions to pursue are the
subject of Part IV.
IV. INSTITUTIONAL REFORMS TO MEET THE CHALLENGE: BEYOND
MERELY SCALING UP THE PATENT BUREAUCRACY
What non-brute-force innovations can the United States use
to relieve the human-labor bottlenecks that the patent system’s
own flourishing creates? This Part discusses three potential
approaches: greater work-sharing with foreign patent offices,
greater leveraging of private initiative and knowledge, and more
generalized USPTO fact-finding. Aspects of at least the first two
approaches would likely face significant opposition from the U.S.
patent examiners’ labor union, the Patent Office Professional
Association, a powerful force in patent-law politics that has a
138. See JERRY L. MASHAW, BUREAUCRATIC JUSTICE: MANAGING SOCIAL SECURITY
DISABILITY CLAIMS 26 (1983) (“From the perspective of bureaucratic rationality,
administrative justice is accurate decisionmaking carried on through processes
appropriately rationalized to take account of costs.”).
Do Not Delete
490
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
history of opposing the outsourcing of examiners’ work.139 But
this short piece will not propose how to resolve such concerns of
political economy. Instead, it will simply discuss some of the
major options for addressing the current and future patent
system’s entirely predictable, essentially inexorable workload
needs. In any case, as these needs place greater and greater
pressure on the system, one might expect that various seemingly
insurmountable political barriers will ultimately give way.
A. Pooling Resources with Foreign Patent Offices
Given the recently dominant contribution of foreign
applications to U.S. patenting growth,140 perhaps the most
obvious response is for the U.S. to seek more substantial help
from abroad. Many foreign applicants for U.S. patents
simultaneously apply for foreign patents on the same subject
matter, and many domestic U.S. applicants seek patent
protection in other countries.141 Greater worksharing with foreign
patent offices, which commonly are doing largely duplicative
142
research into questions relating to patentability, is perhaps the
most straightforward response to the foreign-filing “explosion.”
Patent offices for twenty-four countries and the European
Union have sought to achieve such worksharing through a
number of “Patent Prosecution Highway” (PPH) programs, under
139. See Michael Abramowicz & John F. Duffy, Ending the Patenting Monopoly, 157
U. PA. L. REV. 1541, 1573 & n.110 (2009) (noting that, in response to a USPTO proposal to
“privatiz[e] prior art searches,” “political opposition has been voiced most fiercely by the
examiners’ union”); David P. Irimies, Why the USPTO Should Adopt a Deferred Patent
Examination System, 20 DEPAUL J. ART, TECH. & INTELL. PROP. L. 355, 386 (2010)
(contending that the “main policy rationale” for a provision in a patent reform bill that
would have “foreclose[d] the possibility of work sharing substantive examination of patent
applications and prior art searches amongst national patent offices” is “that labor unions
are a major stakeholder in Democratic Party politics”); S. Jay Plager, Foreword: The
Evolving Future of Law and Technology, 8 J. TECH. L. & POL’Y xiii, xv (2003) (observing
that a proposal to “contract[] out prior art search responsibilities” was “not surprisingly a
move opposed by the examiner corps union”); cf. Chris J. Katopis, Perfect Happiness?:
Game Theory as a Tool for Enhancing Patent Quality, 10 YALE J.L. & TECH. 360, 379
(2008) (“POPA has demonstrated ability to influence both Republican and Democratic
members of Congress. . . . POPA’s two primary goals are essentially to obtain higher pay
and more examining time per application for its personnel.”). See generally Arti K. Rai,
Growing Pains in the Administrative State: The Patent Office’s Troubled Quest for
Managerial Control, 157 U. PA. L. REV. 2051, 2065 (2009) (noting that USPTO reforms
“may be difficult to implement in a work environment where the patent examiners’ union
tends to view management-proposed changes as attempts to squeeze examiners further
with respect to the time that they have to examine applications”).
140. See supra text accompanying notes 67–71.
141. See Mabey, supra note 14, at 222 (“[A]pplications claiming identical inventions
are routinely filed in numerous global patent offices.”).
142. See id. (“The basic requirements of patentability are remarkably similar across
the world’s industrialized nations.”).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
491
which the results of at least partial examination of an application
in one office can help expedite review in another.143 Based on
various reports, the PPH programs appear to be delivering faster
and cheaper prosecution results for applications in the
program.144
Nonetheless, despite robust growth in the last few years, the
percentage of PPH filings is still miniscule: a record number of
approximately 5,000 PPH filings at the USPTO in the USPTO’s
145
fiscal year 2012 apparently represented only about one percent
of the more than 500,000 applications the USPTO received that
year.146 Moreover, whatever savings of time and effort that
worksharing through the PPH achieves, these savings are
generally less than 100%. The PPH provides significantly less
than a full pass through the USPTO’s examination process.147 In
short, the PPH is currently only saving the USPTO and patent
applicants a fraction of their time, effort, and expense on less
than 1% of currently pending applications.
Further substantive harmonization of the patent laws of
different countries might enable greater worksharing. By moving
the United States’ novelty rules toward the first-to-file rules
more typical of other countries, the United States’ America
Invents Act148 might represent a step toward greater capacity for
worksharing. But the miniscule percentage of patents running
through the PPH suggests we are at least a decade away from
worksharing at a level that would take a truly significant bite out
of the workload facing the USPTO and other patent offices.
Meanwhile, patenting and efforts to patent continue to
143. See Toshinao Yamazaki, Patent Prosecution Highways (PPHs): Their First Five
Years and Recent Developments Seen from Japan, 34 WORLD PAT. INFO. 279, 279 (2012)
(providing an overview of PPH programs).
144. See id. at 280–82 (claiming PPH benefits in terms of speed of “patent
acquisition,” increased allowance rates, and reduced costs); see also 2012 USPTO
PERFORMANCE REPORT, supra note 13, at 20 (“[PPH] [u]sers benefit not only by fast
portfolio-building, but also by enjoying the collective savings of millions of dollars in the
process.”).
145. See 2012 USPTO PERFORMANCE REPORT, supra note 13, at 21 fig.8 (showing
that about 5,000 additional PPH filings occurred between the end of September 2011 and
the end of September 2012).
146. Id. at 175 tbl.1 (listing preliminary patent application numbers for fiscal year
2012).
147. See Patent Prosecution Highway (PPH)—Fast Track Examination of
Applications, U.S. PATENT & TRADEMARK OFFICE, http://www.uspto.gov/patents/init_
events/pph/index.jsp (last visited Nov. 19, 2013) (describing the PPH as enabling “fasttrack examination” of patents, rather than issuance without further examination).
148. MUELLER, supra note 43, at 242 (observing that the AIA’s implicit “firstinventor-to-file standard” is “characterized by Congress as more in keeping with
international standards” and is not absolute but instead “subject to three important
exceptions”).
Do Not Delete
492
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
accelerate. By the time substantial worksharing is able to take,
say, a twenty-percent bite out of the workload faced by the
USPTO, the size of the remaining eighty percent of the workload
might already be overwhelming.
The likely inadequacy of international worksharing to
relieve the USPTO’s long-term bottleneck problems reflects the
fact that such worksharing is, to some degree, just another
variant of the traditional approach of bringing more
governmental hands on deck. The helpful twist is that
international worksharing enables the USPTO to seek help from
hands not on the USPTO’s deck, but instead already performing
essentially redundant work in patent offices abroad. But this
twist is of limited long-term value. Even perfect worksharing
cannot alter the fact that, as the number of patent applications
continues to increase, there will ultimately be a need for an
additional governmental hand somewhere. The personnel
shortfalls and cost concerns that plague USPTO operations thus
merely become globalized. Unless we look forward to a day when
every hundredth person works for a patent office, we will
presumably need to seek further help elsewhere.
B. Privatization Options
Of course, another way of obtaining the benefit of additional
hands is to look beyond the government itself. Although William
Baumol has declared that “privatization is no cure for the cost
disease”149 (in which the patent system shares), he has recognized
this as one potential response and seems primarily to have
reacted negatively because of concerns about “threat[s] to liberty”
and “politically irresistible” backlash to privatization in the form
150
of price controls —concerns that seem likely to have less
resonance with respect to privatization of certain aspects of the
U.S. patent system’s operation. In any event, there are at least
149. BAUMOL, supra note 16, at 64. Baumol defines “the cost disease” as “a
widespread pattern of increasing costs” caused by “the unprecedented and spectacular
productivity growth that the world’s industrialized nations have achieved since the
Industrial Revolution.” Id. at 3, 5; see supra Part III.A (discussing patent law’s labor
problem as a form of Baumol’s “cost disease”).
150. BAUMOL, supra note 16, at 64–65 (noting the “threat to liberty posed by reliance
on private armies” and contending that, “if rising costs are caused by unavoidably slow
productivity growth in personal and handicraft services, cost controls can only lead to
deterioration in the quality of those services or, worse, to their partial or total
disappearance”); see also William J. Baumol, Health Care, Education and the Cost
Disease: A Looming Crisis for Public Choice, 77 PUB. CHOICE 17, 27 (1993) (arguing that,
with costs inevitably rising even after privatization, “[t]he prospect is that public pressure
will be the enemy of the public interest as in so many cases where price control has been
invoked as the magic formula for cost reduction”).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
493
three ways in which “privatization” of USPTO work might be
increased. First, one can try to extract more information or work
in “patent filtering” from patent applicants themselves. Second,
one can look for similar voluntary contributions from interested
third parties. Third, one can more actively contract out aspects of
the government’s work to private entities looking to make a
direct profit from that work.
In line with patent law’s historical development of
specification and claiming requirements,151 the most obvious
“privatization” alternative is to consider what more might be
extracted from applicants themselves.152 Whether because
applicants have developed the claimed inventions themselves or
because the applicants have chosen to acquire prospective rights
to those inventions, applicants might be presumed to start with
some degree of knowledge that would be helpful to examiners. In
some circumstances, applicants might even be better able to acquire
additional relevant knowledge. Already existing disclosure
requirements, including a duty of candor153 whose breach can lead to
a later charge of inequitable conduct,154 attempt to leverage such
private knowledge and capacity to some degree. Moreover, a
particularly promising avenue for reform might involve both
empowering and requiring the USPTO to demand or impose greater
clarification of patent scope during a patent’s initial examination.155
The result might be significantly greater applicant investment in
the clarity of patent claims and specifications, and the resulting
increased clarity might help reduce costs of “patent processing” both
156
by the USPTO and by private parties and courts later on.
151. John M. Golden, Construing Patent Claims According to Their “Interpretive
Community”: A Call for an Attorney-Plus-Artisan Perspective, 21 HARV. J.L. & TECH. 321, 349–
51 (2008) (discussing the development of written description and claiming requirements).
152. Cf. Abramowicz & Duffy, supra note 139, at 1574 (viewing as a form of privatization
a “set of new initiatives . . . impos[ing] more burdens on the patent applicant to develop the
information necessary to evaluate the application”).
153. 37 C.F.R. § 1.56(a) (2012) (“Each individual associated with the filing and
prosecution of a patent application has a duty of candor and good faith in dealing with the
Office, which includes a duty to disclose to the Office all information known to that individual
to be material to patentability as defined in this section.”).
154. MUELLER, supra note 43, at 551 (“[A] severe penalty is set for violations of the duty
[of candor to the USPTO] that come to light in subsequent litigation of a patent: if the defense
of inequitable conduct is proved, the entire patent (i.e., all claims, regardless of their validity) is
rendered unenforceable . . . .”).
155. See Wagner, supra note 133, at 2166 (“One potentially fruitful area of
experimentation would require the USPTO to conduct much more thorough claim-construction
analyses . . . defining key terms according to public reference works.”).
156. See Michael J. Meurer, Patent Examination Priorities, 51 WM. & MARY L. REV. 675,
702 (2009) (noting how defects in “notice” of the scope of patent rights can generate costs
or problems for patent examiners, courts, lawyers, and innovators).
Do Not Delete
494
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
Moreover, as the U.S. Supreme Court’s decision in Microsoft
Corp. v. i4i Limited Partnership confirmed by reaffirming the
present existence of a strong, across-the-board presumption of
the validity of issued patent claims,157 the U.S. patent system can
do more to place burdens of production or proof on applicants for
patent rights or the patentees who hold any rights that result.
On the other hand, substantially increasing burdens on
applicants—such as by requiring proof of a plausibly adequate
prior-art search and further requiring a relatively impartial
report on that search’s results—would likely substantially
increase the effective cost of applying for patents. Such an
effective cost increase could further discourage patenting by
already cost-conscious innovators, potentially discouraging
patenting by start-ups158 who might be among patent law’s most
159
socially appropriate clientele.
A second “privatization” approach is to leverage third
parties’ knowledge or incentives to acquire knowledge about
potential or actual patent rights. With some aid from Congress,
the USPTO has taken a number of incremental steps in this
direction over the last few decades. A variety of reexamination or
post-grant review proceedings enable private parties to challenge
patent rights in administrative proceedings that tend to be much
cheaper than litigation in Article III courts and, because of lower
thresholds for administrative standing as compared to Article III
standing, more accessible to parties not immediately threatened
160
with an infringement suit. The America Invents Act has also
facilitated private parties’ submission of relevant information to
the USPTO even prior to a patent’s grant.161 Moreover, a Peer-to-
157. Microsoft Corp. v. i4i Ltd. P’ship, 131 S. Ct. 2238, 2244 (2011) (rejecting a
contention that, for challenges to the validity of issued patent claims, “a preponderance
standard [of proof, rather than a clear-and-convincing standard,] must apply at least
when an invalidity defense rests on evidence that was never considered by the [US]PTO
in the examination process”).
158. See Stuart J.H. Graham et al., High Technology Entrepreneurs and the Patent
System: Results of the 2008 Berkeley Patent Survey, 24 BERKELEY TECH. L.J. 1255, 1310
(2009) (“We find that, among technology startups, the cost of getting a patent is the
most common reason cited for not patenting a major technology.”).
159.
See Golden, supra note 24, at 169–70 (describing how patents can help
support the business models of small biotechnology companies).
160.
Golden, supra note 74, at 613–14 (discussing the emergence of various forms
of reexamination or post-grant review by the USPTO since 1980).
161.
35 U.S.C. § 122(e) (2012) (permitting pre-issuance submissions by third
parties regarding any information that may be relevant to the USPTO); see also
Dylan M. Aste, Note, To Disclose or Not to Disclose: Why the United States Properly
Adopted the European Model for Third-Party Participation During Patent
Prosecution, 3 CASE W. RES. J.L. T ECH. & INTERNET, no. 1, 2012, at 1, 3 (discussing a
new provision of U.S. statutory law “that allows third-party submission of prior art
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
495
Patent Project has shown how private parties might productively
contribute to examination of at least a portion of a select group of
applications, in this case applications selected based on the
willingness of patent applicants to subject them to a more open
pre-grant review process.162
I have previously expressed hope that, through such private
challenges or inputs, private parties can help to check at least
163
But because
some of the limitations of USPTO processes.
private incentives to bring such challenges, absent substantial
publicly funded “bounties” provided in great numbers, seem
likely to suffice only in a relatively small set of cases, even much
greater provision of opportunities for private challenges and
inputs appears unlikely to solve the general problem of dealing
with the great mass of applications that are flowing into the
USPTO at an accelerating rate.
Naturally, therefore, we might turn to the next set of
“privatization” possibilities, a set characterized by much more
ambitious privatization through such measures as the
widespread provision of monetary “bounties” to challenge patent
164
rights, as proposed in different ways by John Thomas and
165
Joseph Miller;
or the partial or general delegation of
examination itself to regulated private firms, a proposal made by
166
Michael Abramowicz and John Duffy.
Such proposals are
certainly intriguing. By offering hope that competition and
[to the USPTO] along with a description of its relevance before the USPTO issues a
patent”).
162. Aste, supra note 161, at 18 & n.107 (describing the USPTO’s Peer-to-Patent
Project, which opened the examination of a couple hundred applications “to online public
participation,” which “allows anyone and everyone to participate as a patent reviewer
and/or a patent facilitator”); see also Giuseppina D’Agostino, Challenges to the Patent
System 23 (Dec. 19, 2011) (unpublished manuscript), available at http://papers.ssrn.com/
sol3/papers.cfm?abstract_id=1991929 (describing the USPTO’s Peer-to-Patent project as
successful but noting concern that “there is a clear challenge in scaling-up the program”
in light of apparently limited interest in participation).
163. Golden, supra note 74, at 615–17.
164. John R. Thomas, Collusion and Collective Action in the Patent System: A
Proposal for Patent Bounties, 2001 U. ILL. L. REV. 305, 307 (“I propose that the Patent
Office award cash prizes to informants who timely disclose patent-defeating references.”).
165. Joseph Scott Miller, Building a Better Bounty: Litigation-Stage Rewards for
Defeating Patents, 19 BERKELEY TECH. L.J. 667, 677 (2004) (proposing monetary rewards
for patent challenges that operate “at the litigation stage, in an amount that varies as a
function of the patentee’s net profits from practicing the technology set forth in the
asserted patent claims”).
166. Abramowicz & Duffy, supra note 139, at 1576 (suggesting a system in which
examination of patent applications is conducted substantially by private firms regulated
by the USPTO); cf. Wagner, supra note 133, at 2160 (reporting that Japan has
“outsourc[ed] the majority of searches [for prior art relevant to patent applications] to
private organizations”).
Do Not Delete
496
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
decentralization will help spark significantly greater efficiency,
Abramowicz and Duffy in particular suggest reason to think that
systematic privatization might substantially improve society’s
capacity to shoulder the ever-growing patent workload.167
But how can such efficiencies be achieved? Abramowicz and
Duffy offer their proposal as an alternative to lumbering
government-based approaches that commonly end up embracing
either inefficient decision-making through “multimember
panel[s]” or unsatisfactory decision-making through “specific
168
rules” that purport to tightly constrain examiner discretion.
Thus, they appear to contemplate that private firms will review
169
patents through a process largely of single-examiner review.
But if decision-making is still largely the function of
individual examiners conducting highly individualized studies of
individual patent applications, one must wonder how great the
efficiency gains can be. By how much might private firms,
working through the mechanism of individualized examination,
cut the required examination workforce? The USPTO currently
receives over 500,000 patent applications a year and employs
170
Commentators
about 7,000 examiners to review them.
commonly view as manifestly insufficient for thorough review the
estimated twenty hours or so that an examiner has, on average,
171
to work on an individual application.
Let’s assume,
nonetheless, that examiners employed by private firms are so
efficient, perhaps because of lower turnover and better personal
and technological support, that they can complete a satisfactory
examination of individual patents in an average of twenty hours.
Let’s also assume that the private firms squeeze an average of
2,000 examination hours out of each examiner each year—a quite
optimistic figure, I think, given that law firms often fall short of
achieving such an average for the total annual billable hours of
law-firm associates presumably having much higher salaries
167. See Abramowicz & Duffy, supra note 139 at 1555 (noting that “the most
significant advantages of private, rather than governmental, provision arise from the
process of competition” and that “[t]he inability of the [US]PTO to adhere to basic
commercial standards and adopt new technology is representative of the organizational
challenges that government-owned enterprises face”).
168. Id. at 1559–60.
169. See id. at 1576 (discussing private examination as a possible solution to achieve
demonopolization).
170. 2012 USPTO PERFORMANCE REPORT, supra note 13, at 10, 176.
171. See, e.g., Lemley, supra note 76, at 1496 n.3 (“Examiners have astonishingly
little time to spend on each application—on average, a total of eighteen hours . . . .”);
Meurer, supra note 156, at 677 (observing that an estimated average of eighteen hours
per patent for examination “compares unfavorably to what I expect of my students when
they write research memos”).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
497
than patent examiners.172 How many examiners would the
private firms need to employ to process 500,000 applications per
year? Multiply 500,000 by 20 hours and then divide by 2,000
hours per examiner, and one ends up with an answer of 5,000
examiners, a number roughly like the total number of examiners
currently employed by the USPTO.173
In short, at least at the level of processing volume, as
opposed to producing quality results, privatization of
individualized examination seems unlikely to hold much magic
that can make the patent workforce problem go away.
Ultimately, if we wish to continue to increase patent-processing
capacity in line with accelerating growth in patenting, we seem
destined to need either a far vaster patent workforce or
qualitatively different approaches to examination of individual
patents. Part IV.C considers some possibilities.
C. Rulemaking or Other Approaches to “Partially Automated”
Examination
How can the USPTO or even private-sector patentcertification entities adjudicate the patentability of more and
more claimed inventions other than by adding hands? A variety
of incremental efficiency-enhancing steps can undoubtedly be
taken: as with other government agencies, the USPTO’s
information technology has a reputation for being somewhat
embarrassingly backward.174 But as pointed out above, basic
mathematics suggests that, as long as we remain true to the
current individualized examination model, straightforward
efficiency-increasing reforms such as mere updating of
172. See Number of Associate Hours Worked Increases at Largest Firms, NALP
BULL., Feb. 2012, tbl.2, available at www.nalp.org/billable_hours_feb2012 (showing
reported average billable hours of less than 1,900 for all categories of firms in 2012).
173. Cf. Doug Lichtman & Mark A. Lemley, Rethinking Patent Law’s Presumption of
Validity, 60 STAN. L. REV. 45, 53 (2007) (noting that to do significantly more than the
USPTO does in examining applications, such as hiring outside experts to comment on
individual applications, “would be enormously costly”).
174. See Diane Bartz, Ex-IBM Exec Tries to Reboot U.S. Patent Office, REUTERS (Jan.
27, 2010), http://www.reuters.com/article/2010/01/26/us-usa-budget-patents-idUSTRE60P
66220100126 (“An antiquated computer system which crashes and idles thousands of
workers is ‘frankly embarrassing,’ says U.S. Patent and Trademark Office Director David
Kappos, a former International Business Machines executive who estimates it would take
several hundred million dollars to upgrade.”); Jake Ward, President Obama Calls USPTO
Case Management System “Embarrassing”, ANTICIPATE THIS!TM PATENT & TRADEMARK
LAW BLOG (Jan. 15, 2010), http://anticipatethis.wordpress.com/2010/01/15/presidentobama-calls-uspto-case-management-system-embarrassing/ (quoting prepared remarks by
President Obama describing the USPTO as “receiv[ing] more than 80 percent of patent
applications electronically, then manually print[ing] them out, scan[ning] them, and
enter[ing] them into an outdated case management system”).
Do Not Delete
498
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
information technology will likely leave intact the bulk of the
patent-processing challenge.
My sense is that, in a patent system that already employs
crude proxies for social value such as novelty, nonobviousness,
and a mostly feeble sense of utility,175 we need to be more willing
to abandon illusions that the patent system can hope regularly to
attain truly individualized, precise justice. In the spirit of Jerry
Mashaw’s “bureaucratic rationality,” what we can more
realistically hope to attain are better overall results with respect
to the mass of patent cases, where the concept of “better”
includes consideration of patents’ public and private costs,
including administrative costs, as well as their public and private
benefits.176 To this end, rulemaking that effectively truncates the
need for individualized fact-finding and decision-making on at
least some cross-sections of issues seems a promising way
forward. Here again, I follow the more general path of Baumol,
who has noted that the so-called “stagnant services,”177 those
marked by relatively low productivity growth, tend to be ones
that have resisted standardization and automation.178 An obvious
response is to seek ways to overcome or circumvent this
resistance, whose inherency Baumol himself has suggested can
sometimes be more illusion than reality.179 More prevalent and
effective USPTO rulemaking is in effect a way of trying to escape
some of the basic causes of stagnancy by introducing more
standardization and automation into USPTO review of individual
patent applications.
The USPTO has already shown a capacity to develop rules
that effectively guide and simplify decision-making in significant
classes of cases, most notably through guidelines on utility and
written description in the context of biochemical innovation.180
175. See John M. Golden, Patentable Subject Matter and Institutional Choice, 89 TEX.
L. REV. 1041, 1065 (2011) (describing various tests for patentability as “no more than
crude proxies for the question of whether any individual patent grant will further overall
social goals”).
176. See generally MASHAW, supra note 138, at 26 (“From the perspective of
bureaucratic rationality, administrative justice is accurate decisionmaking carried on
through processes appropriately rationalized to take account of costs.”).
177. BAUMOL, supra note 16, at 22 (emphasis omitted).
178. Id. at 22–23 (observing that, in comparison to manufacture of automobiles,
surgery and car repair “are inherently resistant to standardization” and that, for various
services, “quality is—or at least is believed to be—inescapably correlated with the amount
of labor expended on their production”).
179. See id. at 23 (noting some “self-deception by the providers of [stagnant] services
and the customers who obtain them” and the possibilities for greater use of computers in
health care, “a filmed lecture by an extraordinarily talented teacher,” and “recorded
music”).
180. See Golden, supra note 122.
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
499
Such substantially fact-based rulemaking seems a model that the
USPTO should be able to replicate with respect to other
technologies. Why can’t the USPTO chronicle, for the benefit of
its examiners, various already standard ways of combining
communication, doing business, or finding amusement with
existing electronic, computer-related, or more specifically
Internet-related technologies? Couldn’t the USPTO have
generated an account of the replacement of various mechanical
control technologies with electronic control technologies that
would have better enabled the examiner of the patent in KSR
International Co. v. Teleflex Inc. to reject that claimed invention
as obvious?181 Although catalogs of standard “moves” will
necessarily be a moving target as knowledge and practice
develop, even a not-fully-up-to-date recorded catalog can remain
useful for the class of potential combinations to which it applies.
By explaining and ordering knowledge in a way that is
specifically tailored to help examiners, USPTO guidelines can
effectively arm the USPTO to make fact-supported rejections on
various grounds of unpatentability without having to reinvent
the wheel.
Such USPTO rulemaking need not be “substantive
182
rulemaking” that possesses binding force of law in the courts.
As indicated above, the USPTO can work effectively through
183
guidelines. This approach might even be preferable to truer
USPTO “lawmaking” because guidelines tend to be easier to
amend184 and also tend naturally—on judicial review if nowhere
else—to leave room for rebuttal or distinction based on the facts
of the individual case.
More radical savings in individual examination could come if
we could trust machines to do some significant part of the review
for us—effectively using rules encoded in their programming
181. See KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 424 (2007) (concluding that, at
the time of invention, “[t]echnological developments made it clear that engines using
computer-controlled throttles would become standard”); cf. Michael Abramowicz & John
F. Duffy, The Inducement Standard of Patentability, 120 YALE L.J. 1590, 1677 (2011)
(“The occurrence of an exogenous shock that decreases the cost of an invention or
increases the benefit of an invention should count as a secondary consideration of
obviousness.”).
182. See generally Ryan Vacca, Acting Like an Administrative Agency: The Federal
Circuit En Banc, 76 MO. L. REV. 733, 744–47 (2011) (describing “substantive
rulemaking”).
183. See supra text accompanying note 180.
184. See Kathryn A. Watts, Regulatory Moratoria, 61 DUKE L.J. 1883, 1923 (2012)
(noting that “more informal guidance documents, general policy statements, and
interpretive rules . . . are generally exempt from notice-and-comment requirements”
(footnotes omitted)).
Do Not Delete
500
11/24/2013 11:28 AM
HOUSTON LAW REVIEW
[51:2
either to filter patent wheat from application chaff or, at least to
flag for significantly more searching review, patent applications
that appear particularly problematic or otherwise sensitive. With
the merits of software to grade student examinations already
subject to debate,185 is software that serves as a check or
substantial aid to the evaluation of patentability conceivable?
Even if the answer is “Yes,” there are likely to be limits. Dangers
of applicants figuring out the nature of the software’s algorithms
and then “gaming the system” would loom large. As the Internal
Revenue Service has found and as courts of equity have long
appeared to know, rules with some flex in them can be the best
way to combine a desirable degree of predictability, justice, and
robustness against opportunistic behavior.186 Such flex might be
best implemented through human examiners. A bit more humanoriented algorithmic formulation—in the shape of fact-based
guidelines directed at relatively specific technological issues—
could improve examination efficiency and quality in the face of a
rising patent flood.
V.
CONCLUSION
The current crush of patent applications appears unlikely to be
a passing phenomenon. For centuries, both the number of patent
applications flowing into the patent office and the number of
granted patents flowing out have been growing at a generally
accelerating rate. This continual acceleration exerts constant
pressure on patent law’s administrative institutions and forced
repeated, dramatic change in the U.S. patent system’s early
decades. A “more hands on deck” approach has been the dominant
response to this pressure since the generation of a patent office with
a professional examining corps under the Patent Act of 1836. But
even as the USPTO opens four regional offices, this strategy seems
to come ever closer to Sisyphean futility. Indeed, the very opening of
the new offices might suggest how desperate the search for patentbureaucracy personnel has become.
A mix of additional strategies can limit the need to rely on
the “more hands on deck” approach. Drawing on the efforts of
185. John Markoff, Software Seen Giving Grades on Essay Tests, N.Y. TIMES, Apr. 5,
2013, at A1 (“[T]he use of artificial intelligence technology to grade essay answers has not
yet received widespread endorsement by educators and has many critics.”).
186. See Mark P. Gergen, John M. Golden & Henry E. Smith, The Supreme Court’s
Accidental Revolution? The Test for Permanent Injunctions, 112 COLUM. L. REV. 203, 237–
38 (2012) (“It is foundational that equity must be open textured in light of the ability of
parties to opportunistically evade legal obligations. This problem is at an extreme in tax
law, where anti-evasion doctrines are aimed at discouraging opportunism.” (footnote
omitted)).
Do Not Delete
2013]
11/24/2013 11:28 AM
PROLIFERATING PATENTS
501
foreign patent offices and increasing the contributions of private
parties to examination can mitigate current and even mediumterm pressures. But there are likely limits to the capacity of
these approaches to keep up with continually accelerating growth
over the long term. Another approach—incorporating more
generalized fact-finding and rule-informed adjudication in the
examination process—might help fill any remaining breach. The
USPTO likely can provide more fact-based but nonetheless
generalized guidance to examiners on what sorts of innovations
and innovation disclosures should be considered to meet patent
law’s basic patentability requirements. Over time, greater
experience with such rulemaking might enable the USPTO to be
more adept at it, in particular better at using it to facilitate
processing an ever-growing deluge of applications. Whether the
USPTO can and should evolve into an institution that delivers a
socially acceptable form of “mass justice” as opposed to more
individualized adjudication remains to be seen. But some move to
experiment with more systematic rulemaking to expedite the
work of USPTO examiners and appeals boards seems well worth
a try.