Patents: Information about the Publication Provisions of the
American Inventors Protection Act (20-MAY-04, GAO-04-603).
The U.S. Patent and Trademark Office (USPTO) receives over
300,000 patent applications each year. Before the American
Inventors Protection Act of 1999, USPTO was required by law to
keep the information on patent applications confidential until a
patent was granted. The act modified this requirement and
mandated that USPTO publish most patent applications 18 months
after filing. One exception to this requirement is available to
patent applicants filing only in the United States. These
applicants can request that their application not be published.
The act required GAO to provide information on how patent
applications have been affected by the 18-month publication
provisions. GAO was required to determine (1) the number of
patent applications filed only in the United States, and (2)
whether certain differences exist between published and
unpublished patent applications. Specifically, GAO examined
differences relating to whether the applicant was a large or
small entity, the percentage of patents issued by USPTO and
applications abandoned by the applicants, and the length of time
between filing an application and when USPTO issued a patent or
the application was abandoned.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-04-603
ACCNO: A10054
TITLE: Patents: Information about the Publication Provisions of
the American Inventors Protection Act
DATE: 05/20/2004
SUBJECT: Agency proceedings
Information disclosure
Intellectual property
Patents
Reporting requirements
Patent applications
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
******************************************************************
GAO-04-603
Report to Congressional Committees
May 2004
PATENTS
Information about the Publication Provisions of the American Inventors
Protection Act
Contents
Table
Figures
May 20, 2004Letter
The Honorable Orrin G. Hatch Chairman The Honorable Patrick Leahy Ranking
Minority Member Committee on the Judiciary United States Senate
The Honorable F. James Sensenbrenner Jr. Chairman The Honorable John
Conyers Jr. Ranking Minority Member Committee on the Judiciary House of
Representatives
The U.S. Patent and Trademark Office (USPTO) administers the nation's
patent laws to protect inventions and encourage innovation by examining
applications and granting patents to qualified inventors for a period of
time. A patent provides inventors with certain property rights to the
invention.1 Between fiscal years 1998 and 2003, USPTO received an average
of about 317,000 patent applications annually. Following the enactment of
the American Inventors Protection Act of 1999,2 USPTO significantly
changed its processes for publishing patent applications, bringing U.S.
patent policy into greater harmony with the policies of most other
nations. Effective November 29, 2000, USPTO generally must publish patent
applications 18 months after the application is first filed anywhere.
Previously, USPTO published patent applications when patents had been
issued--for fiscal year 2003, a patent was typically issued about 28
months after the filing of the application.3 With this new publication
requirement, other inventors, in the United States and abroad, may be able
to improve upon the invention in the original application and submit their
own patent application more quickly. While the new publication requirement
may accelerate the pace of innovation, it also raised concerns among some
smaller entities-individual inventors, nonprofit organizations, and
enterprises with 500 or fewer employees-that larger competitors might be
able to take advantage of the published information and use their greater
resources to reap benefits in the marketplace before USPTO has granted a
patent to the original inventor.
The 1999 act contains two provisions to address the concerns of these
smaller entities. First, it allows all inventors to avoid publication by
certifying, when they file their application, that they are, in essence,
filing for a patent only in the United States.4 Generally, applications
that do not include such certification are automatically published 18
months after they are first filed. Second, inventors whose applications
are published, and who have been issued patents, may be able to collect
compensation5 retroactively to the date of publication from those who use
the invention without approval. Since inventors are responsible for
enforcing their patent rights, without the aid of USPTO, the potential
litigation costs could deter some inventors from seeking to recover
reasonable compensation.
While inventors may certify that they are seeking a patent only in the
United States, they may rescind their certification at any time. Inventors
may rescind their requests for several reasons, such as deciding to seek a
patent abroad or to become eligible to collect compensation from others
who are using their invention without approval. An applicant who makes a
certification and later files in another country directly or under an
international agreement must notify USPTO of such filing within 45 days.
According to USPTO officials, the agency records in its database both
those that specifically rescind their certification and those that provide
notice as having rescinded their certification.6 In both cases, USPTO
publishes information on the application as it would for any other
application. USPTO does not require patent applicants who do not certify
to indicate whether they intend to file in only the United States or
whether they also intend to file abroad.
The American Inventors Protection Act required GAO to provide information
on patent applications affected by the 18-month publication provisions. In
response, we analyzed USPTO's database for applications submitted to the
agency during a 3-year period-November 29, 2000, through November 28,
2003--to determine (1) the number of applications seeking patents only in
the United States, the number that requested at the time of filing not to
be published, and the number that later rescinded that request; and (2)
whether certain differences exist between published and unpublished patent
applications. Specifically, GAO examined differences relating to whether
the applicant was a large or small entity, the percentage of patents
issued by USPTO and applications that were abandoned, and the length of
time between the applicants' filing of an application and either USPTO's
issuing of a patent or the application being abandoned.7
To address these questions, we examined USPTO records of the patent
applications filed or published between November 29, 2000, and November
28, 2003. We interviewed program officials to determine how these records
are maintained and to interpret the data the records contain. We assessed
the completeness and accuracy of the data we used in our review by (1)
performing electronic tests of relevant fields and (2) reviewing related
program and system documentation, as well as audit and system reviews and
reports. We determined that the data were sufficiently reliable for the
purpose of our review. For about 44 percent of patent applications,
USPTO's data allowed us to determine whether the applicants sought patent
protection only in the United States or in both the United States and
other countries. For the remaining 56 percent of applications, we relied
on USPTO officials' best estimates of how many applicants would also seek
patent protection outside of the United States. To obtain a better
understanding of patent law, we discussed patent issues with USPTO
economists and attorneys. We conducted our work from June 2003 through
April 2004 in accordance with generally accepted government auditing
standards.
Results in Brief
USPTO's database does not provide information for many applications
regarding whether the applicant intends to file only in the United States
or in both the United States and abroad. Of the approximately 805,000
patent applications USPTO received between November 29, 2000, and November
28, 2003, about 88,000 were filed by applicants certifying they would
apply for a patent only in the United States. Subsequently, 8,000 of these
applicants rescinded this certification. For about another 275,000
applications, the applicants indicated they had already applied for a
patent in another country. For the remaining 450,000 applications, USPTO's
database does not provide the information needed to determine whether the
applicants intend to file in the United States alone or also in another
country.
We found four differences between the published and unpublished patent
applications USPTO received between November 29, 2000, and November 28,
2003:
o USPTO has published or plans to publish applications from about 85
percent of the applicants qualifying as large entities compared with only
about 74 percent of those qualifying as small entities.
o USPTO issued patents to about 22 percent of the applications it had
published and considered about 8 percent abandoned, although resolution
for most of the approximately 456,000 published patent applications is
still pending. For applications it did not publish, USPTO issued patents
to about 25 percent and considered about 19 percent abandoned, although
resolution for most of these approximately 117,000 applications is also
still pending.
o The median length of time for the agency's review of an issued patent
was about 20 months for applications that had been published and about 15
months for those not published.
o The median length of time between the filing and abandonment of an
application was about 18 months for published applications and about 8
months for those not published.
In commenting on a draft of this report, USPTO provided technical comments
that we have incorporated, as appropriate.
Background
USPTO administers the nation's patent and trademark laws to protect
inventions, encourage innovation, and advance science and technology
through the preservation and dissemination of patent information. The
principal effect of patent law in a market economy is that it provides the
economic incentive for inventors to devote resources to technological
innovation. Granting the inventor exclusive rights to control the use of
the patented invention produces this effect.
Patent rights do not rise automatically, however. Inventors who seek
patent protection must first prepare applications that fully disclose and
clearly claim their inventions. Before issuing a patent, USPTO examines
the application to verify that the patent is new, useful, and nonobvious.8
See figure 1 for a description of the application examination and
publication process. If the agency issues a patent, the inventor has, in
most instances, exclusive rights to the invention until 20 years from the
date the application was first filed. Patents typically fall into one of
three categories: (1) utility-for useful inventions, such as processes,
machines, articles of manufacture, or composition of matter; (2)
design-for changes in configuration, shape, or surface ornamentation that
do not involve changes in function; or (3) plant-for asexually
reproducible plants.9 The publication provisions of the American Inventors
Protection Act apply only to applications for plant and utility patents,
which constitute most of the applications filed.10
Figure 1: USPTO's Patent Application Examination and Publication Process
Until 1999, USPTO was required to keep in confidence submitted patent
applications, and applicants had no enforceable patent rights before the
issue date of the patent. The American Inventors Protection Act of 1999
removed this confidentiality, so that, with certain exceptions, such as
applications subject to a secrecy order for reasons of national security,
patent applications filed on or after November 29, 2000, will be published
18 months after they are first filed. In exchange for the loss of
confidentiality and the public dissemination of this proprietary
information, the applicant may be able to obtain compensation
retroactively to the date of publication from those who use an invention
without approval.11
USPTO may publish some applications earlier than 18 months. Some
inventors, for example, request that their applications be published
early. Applications that are first filed abroad will also normally be
published in fewer than 18 months after being filed in the U.S. This is
because USPTO designates an application's publication date as 18 months
from when it was first filed, regardless of whether that occurred in the
United States or abroad. Applicants who file first in another country and
wish to obtain a patent in the U.S. must file with USPTO within a year.12
If such an applicant took 12 months to file with USPTO, the agency would
then publish the application within 6 months. If, for example, such an
application was abandoned 9 months after it was published, then USPTO's
database would show that the application was abandoned in 15 months--the
number of months that had passed since the application had been filed in
the U.S.--rather than the 27 months that had passed since the application
had been first filed anywhere.
Proponents of the act believe that the changes brought U.S. patent policy
into greater harmony with the policies of nearly all other nations and
will increase the rate at which dissemination of new technology occurs,
thus speeding the diffusion of knowledge and avoiding the duplication of
research expenditures by competing firms. Some inventors, however,
objected to the early disclosure requirement of the act. For example, a
group of 26 Nobel laureates claimed that this practice would be most
harmful to small inventors and would discourage the flow of new
inventions. They pointed out that the patents most affected by early
disclosure would likely be breakthrough inventions, often made by
independent inventors, that take time to progress from application to an
issued patent. Independent inventors and smaller entities have also
expressed concern that the new law aids larger entities that have greater
resources, at the expense of smaller entities. They contend that smaller
entities are less able to obtain legal redress if larger firms use their
inventions without approval, and are less able to keep pace if a larger
firm attempts to invent around the patent-that is, to improve the patented
device enough to apply for a separate patent.
The 1999 act allows all inventors not intending to seek patent protection
abroad to request that USPTO publish the application only when it issues
the patent. Such inventors also have the option of rescinding the original
request before USPTO issues a patent and asking the agency to publish the
application as it would other applications.
USPTO's Database Does Not Include Information on the Status of Many
Applications
USPTO's database does not provide information for many applications
regarding whether the applicant intends to file only in the United States
or in both the United States and abroad. Our analysis of USPTO's database
indicates that of the approximately 805,000 applications USPTO received,
applicants certified they would apply for a patent only in the United
States in about 80,000 applications. Although about 88,000 applications
originally included this certification in their applications, about 8,000
applicants later rescinded this certification.13 Consequently, these
80,000 applications are exempt from the publication provisions of the
American Inventors Protection Act because they are filing for patent
protection only in the United States. In about 275,000 applications, the
applicants indicated they had applied for a patent in another country. As
a result, these applications are subject to the publication provisions of
the American Inventors Protection Act, and USPTO will publish these
applications within 18 months of when they were originally filed.
USPTO's database does not provide the information needed to determine
definitively whether the applicants for the remaining 450,000 applications
intended to apply for a patent only in the United States or in both the
United States and abroad. However, USPTO officials estimated that for
about one-third of such applications, the applicants generally seek patent
protection in both the United States and abroad. Applying this proportion
to the 450,000 applications for which no information is available, about
300,000 were likely to seek patents only in the United States and about
150,000 were likely to seek patents in both the United States and
abroad.14 By combining these estimates with the data for the applications
that we could definitively identify from the database, we calculated that
about 380,000 applications (about 47 percent) likely were filed only in
the United States, and about 425,000 (about 53 percent) likely were filed
in both the United States and other countries.15 See figure 2.
Figure 2: Domestic and Foreign Applications Filed with USPTO, November 29,
2000, through November 28, 2003
Some Differences Exist between Published and Unpublished Applications
We found four differences between the published and unpublished
applications filed with USPTO between November 29, 2000, and November 28,
2003.16 First, USPTO published, or plans to publish, applications filed by
inventors qualifying as large entities at a somewhat higher rate (85
percent) than those filed by inventors qualifying as small entities (74
percent). Of the approximately 580,000 applications filed by large
entities, USPTO published about 60 percent and listed another 25 percent
as eligible for publication-that is, the applications will be published 18
months after they were first filed. About 12 percent of the applications
filed by inventors qualifying as large entities are ineligible for
publication-typically because the applicants certified that they will file
only in the United States. USPTO's database does not provide the
information needed to determine the publication status for the remaining 3
percent.
For the approximately 225,000 applications filed by inventors qualifying
as small entities, USPTO published about 46 percent and lists another 28
percent as eligible for publication. About 22 percent of the applications
filed by inventors qualifying as small entities are ineligible for
publication. USPTO's database does not provide the information needed to
determine the publication status for the remaining 4 percent. See figures
3 and 4.
Figure 3: Status of Applications Filed by Inventors Qualifying as Large
Entities, November 29, 2000, through November 28, 2003
Figure 4: Status of Applications Filed by Inventors Qualifying as Small
Entities, November 29, 2000, through November 28, 2003
Second, although the rate at which USPTO issued patents to published and
unpublished applications differed only slightly, the rate at which
applicants abandoned published applications and unpublished applications
differed significantly. Of the approximately 456,000 applications USPTO
published, about 70 percent are still pending, about 22 percent were
issued patents, and 8 percent were considered abandoned. Similarly, of the
approximately 117,000 applications USPTO did not publish, about 56 percent
are still pending, about 25 percent were issued patents, and 19 percent
were considered abandoned. For a summary of these differences, see table
1.
Table 1: Comparison of Published and Unpublished Applications, November
29, 2000, through November 28, 2003
Number
Published (% of total) Unpublished (% of total)
Applications
Issued as patents 102,200 (22) 29,300 (25)
Abandoneda 37,300 (8) 22,300 (19)
Pending 316,200 (70) 65,500 (56)
Totalb 455,700 (100) 117,100 (100)
Median pendency in months
Filing to patent issue 20 15
Time to abandonment 18 8
Source: GAO analysis of USPTO data.
Note: Numbers are rounded to the nearest 100.
aAccording to USPTO officials, the difference in the rate at which
published and unpublished applications are abandoned occurs primarily
because nearly all patent applications abandoned within about 18 months of
filing are listed as unpublished.
bThis analysis excludes approximately 206,000 applications USPTO considers
eligible for publication but that have been neither issued nor abandoned,
and approximately 26,000 applications that have not yet reached the stage
at which publication status can be determined.
Third, USPTO took longer to issue patents that had been published (a
median time of about 20 months) than those that had not been published
(about 15 months). According to agency officials, the difference occurs
primarily because examiners sometimes complete their review and decide to
issue a patent in fewer than 18 months. In these cases, USPTO does not
publish the application because the issued patent will itself be published
and become public information. As a result, the median length of time for
unpublished patents is less.
Fourth, for applications considered abandoned, the median length of time
between filing and abandonment was longer for published applications
(about 18 months) than for those not published (about 8 months). According
to USPTO officials, this difference occurs primarily because nearly all
patent applications abandoned within about 18 months of filing are not
published.
Agency Comments and Our Evaluation
In commenting on a draft of this report, USPTO provided technical comments
that we have incorporated, as appropriate.
We are sending copies of this report to the appropriate House and Senate
committees; interested Members of Congress; the Secretary of Commerce; the
Director, USPTO; the Director, Office of Management and Budget; and other
interested parties. We are also providing copies to others upon request.
In addition, the report is available at no charge on GAO's Web site at
http://www.gao.gov.
If you or your staff have any questions about this report, please contact
me at (202) 512-3841. Key contributors to this report are listed in
appendix II.
Anu K. Mittal Director, Natural Resources and Environment
Scope and MethodologyAppendix I
To provide information on the patent applications affected by the American
Inventors Protection Act of 1999, we analyzed data obtained from the U.S.
Patent and Trademark Office's (USPTO) Patent Application Location and
Monitoring system (PALM).1 We used these data to determine the number of
applications USPTO received, the number of applications certifying that
they would be filed only in the United States, and the number that
rescinded this certification. PALM also contained data we used to
determine whether applicants had filed an application for the same
invention in another country before filing with USPTO. In addition, we
used PALM data to analyze whether differences existed between published
and unpublished applications in terms of the relative size of the applying
entity, the length of time USPTO reviewed the application before it was
issued or abandoned, and the rate at which applications were issued or
abandoned.
In about 44 percent of the cases, the USPTO data provided information
about whether applicants sought patent protection only in the United
States or in both the United States and elsewhere. For the remaining 56
percent of the cases, USPTO's database does not provide the information
needed to determine whether the applicants have filed or intend to file in
the United States alone or also in another country. We discussed other
possible sources of information about these cases with USPTO program
officials, attorneys, and economists. These USPTO officials noted that
while they had conducted one study that provided some data as to whether
applicants filed in the United States alone or also in another country,
the limitations of the study's methodology and results precluded any
statistically valid use of the information. Three other studies we
examined had similar methodological limitations that precluded their use.
The USPTO officials with whom we discussed these issues estimated that,
based on their professional experience, for about one-third of the
applications filed in the United States, applicants may subsequently file
in another country after filing with USPTO. Although we used this estimate
to determine the potential number of applications that may have been filed
in the United States only and in both the United States and abroad, use of
this estimate should be considered in light of the limitation that it
could not be corroborated by other methodologically sound independent
sources.
We assessed the reliability of USPTO's PALM data that we used in our
analysis by (1) performing electronic testing of relevant data elements,
(2) reviewing existing information about the data and the system that
produced them, (3) interviewing agency and contractor officials
knowledgeable about the data, and (4) reviewing audit and system reports
on accuracy of data entry performed by contractors. We determined that the
data were sufficiently reliable for the purposes of this report.
To determine if studies analyzing the impact of the American Inventors
Protection Act had been conducted by any other entity, we interviewed
officials from the Intellectual Property Owners Association, the American
Intellectual Property Law Association, and the National Intellectual
Property Researchers Association. To explore the practicality of
conducting a survey of patent attorneys to determine how many patent
applications were filed only in the United States or in both the United
States and abroad, we contacted patent attorneys around the United States
representing private legal firms, corporations, academia, and government.
After discussions with these officials, we concluded that a survey was not
feasible because of concerns about attorney/client confidentiality.
We conducted our work from June 2003 through April 2004 in accordance with
generally accepted government auditing standards.
GAO Contacts and Staff AcknowledgmentsAppendix II
GAO Contacts
Anu K. Mittal, (202) 512-3841 Cheryl Williams, (404) 679-1991
Staff Acknowledgments
In addition to those named above, Oliver Easterwood, Judy Pagano, Elsie
Picyk, Paul Rhodes, Stuart Ryba, Carol Herrnstadt Shulman, and Mindi
Weisenbloom made key contributions to this report.
(360369)
*** End of document. ***