Chapter 8 | Invention, Knowledge Transfer, and Innovation
Invention: United States and Comparative Global Trends
Inventions and the pace of their emergence are critical features of a national innovation system. Invention is the creation of new, useful, and nonobvious goods, services, and processes and is an important source of the innovations that eventually emerge in the marketplace or other practical use. Some of these may be described in scientific papers, which provide a means for researchers to claim credit and disseminate the results of their discoveries.
Patents serve a different purpose. Inventors often have economic motivations to keep the details of their inventions secret. The patenting system provides the legal right for a limited time to exclude others from making, using, offering for sale, or selling the invention, in exchange for public disclosure of the technical information in the granted patent. Extensive publicly available administrative data exist for patents and their inventors, and extensive databases allow for systematic insights into these patents. In the absence of other comprehensive data on invention, patent data provide unique and useful insights into the inventions deemed valuable enough to patent. However, analysis of these data requires caution.
One caveat is that most patented inventions are never commercialized; they are neither representative of all inventions nor are they measures of innovation. Many valuable inventions that are commercialized are not patented. Companies choose a variety of strategies to protect their inventions and intellectual property. For example, U.S. companies rate trade secrets higher than patents in their importance for protecting intellectual property (Figure 8-1), which is true even for R&D-performing firms.
For companies that performed or funded R&D, shares rating intellectual property as being very or somewhat important: 2011
Note(s)
A mask work is a two- or three-dimensional layout or topography of an integrated circuit on a semiconductor that is protected under U.S. intellectual property law.
Source(s)
National Science Foundation, National Center for Science and Engineering Statistics, Business R&D and Innovation Survey (BRDIS), 2011.
Science and Engineering Indicators 2018
In addition, patent protection may be sought for reasons other than intended commercialization. Privately owned patents may be obtained to block rivals and negotiate with competitors, to use in lawsuits, or to build “thickets” of patents to impede or raise others’ costs of R&D and innovation (Cohen et al. 2000). Research suggests that some organizations and countries pursue “strategic patenting” to block competitors and to monetize patents through licensing and other activities (Ernst 2013:1–9). Other firms may respond by patenting defensively. New and emerging firms may seek patent protection to help obtain financing because investors perceive patents as potentially valuable for a firm’s assets and future profitability. Finally, cross-country analysis indicates that international differences in taxes on corporate and patent income influence the choice of patent location for multinational firms (OECD 2016:3). However, within these limitations, USPTO patent documents tell us when and in what technology areas inventors have decided to protect their intellectual property with patent protection. This rich detail, which also includes the name and address of the inventor and assignee, justifies their presentation. U.S. patents are issued to provide protection to inventions in the U.S. market. Foreign owners account for more than half of USPTO patents in recent years, 152,000 out of a total of slightly more than 300,000 patents granted in 2016 (Appendix Table 8-1). The USPTO reports the five organizations awarding the highest numbers of patents in 2015 as IBM, Samsung, Canon, Qualcomm, and Google (USPTO 2017).
USPTO Patenting Activity
As described previously, the purpose of patenting is to allow inventors to gain the economic benefits of their inventions in exchange for disclosure of technical information about the invention. Most patenting takes place in the business sector. Motivations differ substantially from the motivation of authors of peer-reviewed literature, where original contributions to publicly available knowledge may benefit reputation and career advancement without a direct financial benefit for the authors. Business researchers are also more likely to be engaged in experimental development activity than their academic and government counterparts (see Table 4-4 in Chapter 4), suggesting more opportunities for direct commercial applications of their work.
USPTO patents provide data on the inventor and the owner of the patent (known as the assignee). The data described in the next several paragraphs are based on the economic sector of the patent owner. In 2016, 151,000 USPTO patents were assigned to U.S. owners (Appendix Table 8-1). Among these U.S. owners, the private sector (for-profit companies) by far receives the most patents (85% share). Individuals receive the next largest share (9%), followed by the academic sector (4%). The government sector receives a small share of patents (1%), reflecting in part the focus of government entities on activities other than the protection of intellectual property, as well as a small number of U.S. government patents whose contents may reveal sensitive security information. The nonprofit sector, which is included in the “other” category, receives a very small share of patents (0.3% or less). Over the last decade, the private sector’s share of U.S. patents slightly increased from 82% to 85%. Although the individual share declined from 13% to 9%, continuing a long-term trend away from individual patenting, almost 13,600 patents were granted to individual U.S. owners in 2016.
Patenting by U.S. Industries
USPTO data provide information about the technology area for the patent but not the industry in which the inventor or assignee works. Industry-level measures of patenting are available for a more limited set of firms, those in scope for the National Science Foundation (NSF) National Center for Science and Engineering Statistics Business R&D and Innovation Survey (BRDIS), which focuses on the activity of R&D-performing firms. BRDIS data estimate that more than 91,000 patents were issued to R&D performing firms in the United States in 2015. The U.S. knowledge- and technology-intensive industries described in Chapter 6—high-technology manufacturing, medium-high technology manufacturing, and commercial knowledge-intensive services industries—have a far larger share of patents than other industries (Figure 8-2). U.S. high-technology manufacturing industries received 61% of the 61,000 USPTO patents granted to U.S. manufacturing industries in 2015. Medium-high technology manufacturing industries received almost a quarter of these patents. Together, these industries accounted for more than 80% of all patents granted to U.S. manufacturing industries in 2015.
U.S. commercial knowledge-intensive services received 87% of the 30,000 patents granted to nonmanufacturing industries in 2015 (Figure 8-2). The information services industry accounted for 16,000 patents, 62% of the patents granted to commercial knowledge-intensive services; the professional, scientific, and technical services accounted for almost 10,000 patents (37%).
USPTO patents granted, by selected U.S. industry: 2015
USPTO = U.S. Patent and Trademark Office.
Note(s)
High-technology manufacturing industries include aerospace, communications, computers and office machinery, pharmaceuticals, semiconductors, and testing, measuring, and control instruments. Medium-high-technology manufacturing industries include chemicals excluding pharmaceuticals, motor vehicles and parts, electrical equipment and appliances, machinery and equipment, and railroad and other transportation equipment. Detail may not add to total because of rounding. Industry classification is based on the dominant business code for domestic R&D performance, where available. For companies that did not report business codes, the classification used for sampling was assigned. Statistics are based on companies in the United States that reported to the survey, regardless of whether they did or did not perform or fund R&D. These statistics do not include an adjustment to the weight to account for unit nonresponse. For a small number of companies that were issued more than 100 patents by USPTO, survey data were supplemented with counts from https://www.uspto.gov/, accessed 20 January 2017.
Source(s)
National Science Foundation, National Center for Science and Engineering Statistics, Business R&D and Innovation Survey (BRDIS), 2015.
Science and Engineering Indicators 2018
Trends and Patterns in Academic Patenting
Compared with the production of S&E publications (as described in Chapter 5 section Outputs of S&E Research: Publications) patenting is a less-frequent event. For example, in 2016, 409,000 S&E publications were produced by U.S.-affiliated authors, almost 308,000 of these from U.S. academic authors (Appendix Table 5-41). By contrast, in the same year, 151,000 USPTO patents were assigned to U.S. owners (Appendix Table 8-1), and 6,600 of these patents were assigned to U.S. academic owners (Appendix Table 8-2).
These U.S. patents, together with 4,200 patents granted to foreign universities and colleges in 2016, account for just under 11,000 academic patents. Foreign universities have expanded patenting rapidly since 2008, when 900 were granted (Figure 8-3).
USPTO patents granted to U.S. and non-U.S. academic institutions: 1996–2016
USPTO = U.S. Patent and Trademark Office.
Note(s)
Patents are allocated according to patent ownership information. Patents are credited on a fractional-count basis (i.e., for articles with collaborating institutions, each institutions receives fractional credit on the basis of the proportion of its participating institutions). The sum of patents granted to non-U.S. and U.S. academic institutions is lower than the total number of patents granted to academic institutions as country affiliation of a few academic patents is unknown (data not presented).
Source(s)
National Science Foundation, National Center for Science and Engineering Statistics; SRI International; Science-Metrix; PatentsView; U.S. Patent and Trademark data, accessed April 2017.
Science and Engineering Indicators 2018
In a detailed examination of the USPTO data for 2009 to 2014, Leydesdorff, Etzkowitz, and Kushnir (2016) attribute the rapid growth in foreign university patenting to universities in Taiwan, South Korea, Japan, and China. They also found that universities in Saudi Arabia, Norway, and India experienced particularly rapid growth from a small base. The authors found that, unlike the long-term biomedical focus of European and U.S. university patenting, electronics patents are the focus of much of the recent growth in foreign university patents.
Patent data filings include detailed information on technology area, allowing for analysis of trends in patenting over time. The patent indicators described below are classified by technology areas from the World Intellectual Property Organization (WIPO), summarized into 35 technical fields shown in Appendix Table 8-2 for U.S. university patents for 1996–2016. In 2016, slightly more than half (54%) of all the patents granted to universities were in just 5 of the 35 technical fields: pharmaceuticals, biotechnology, medical technology, organic fine chemistry, and measurement (Table 8-1). For technical areas with more than 100 academic patents, the annual growth rate for 2016 was highest for digital communications (11.1%), microstructural and nanotechnology (9.3%), and computer technology (8.3%).
U.S. university patent awards, by technology area: 2002 and 2016
In 2016, just 5 of the 35 technical fields, pharmaceuticals, biotechnology, medical technology, organic fine chemistry, and measurement, accounted for slightly more than half (54%) of all the patents granted to universities (Table 8-1). Academic patenting data from USPTO are presented in 35 World Intellectual Property Organization (WIPO) technical fields shown in Appendix Table 8-2. The table shows patent awards for U.S. university patents for 1996–2016.
Although the pharmaceuticals field had the highest number of university patents in the most recent year for which we have data, 1,008 patents in 2016, this reflects a relatively recent trend. Over a longer period since the turn of the century, biotechnology patents had accounted for the largest number of U.S. university patents: Figure 8-4 shows the top five areas for university patenting in 5-year averages between 2002 and 2016. Of these five technical fields, medical technology and measurement, consisting of measurement instruments, have shown continual growth over all three 5-year periods.
U.S. academic patents, by selected technology area, 5-year averages: 2002–16
Note(s)
Patents are allocated according to patent ownership information.
Source(s)
National Science Foundation, National Center for Science and Engineering Statistics; SRI International; Science-Metrix; PatentsView; U.S. Patent and Trademark Office patent data, accessed April 2017. See Appendix Table 8-2, which includes data for 35 technology areas.
Science and Engineering Indicators 2018
Global Patent Trends and Cross-National Comparisons
Global and Cross-National Activity in USPTO Patents
The data described in this section are based on the geographic address of the inventor. The USPTO granted more than 300,000 patents in 2016 to inventors all over the world (Figure 8-5; Appendix Table 8-3 and Appendix Table 8-4). The United States received nearly half (47%) of them, followed by Japan (16%) and the member countries of the European Union (EU) (15%). Although several developed and developing economies, including South Korea, China, Taiwan, and India, have seen steep increases over time in their USPTO patenting activity, the United States, the EU, and Japan together still account for the clear majority of USPTO patents (Figure 8-5).
USPTO patents granted, by selected region, country, or economy of inventor: 2006–16
EU = European Union; ROW = rest of world; USPTO = U.S. Patent and Trademark Office.
Note(s)
Patent grants are fractionally allocated among regions, countries, or economies based on the proportion of the residences of all named inventors.
Source(s)
Science-Metrix; SRI International. See Appendix Table 6-37.
Science and Engineering Indicators 2018
After flat growth for most of the 2000s, the number of USPTO patents grew more than 80% between 2009 and 2016, led by growth of patents in information and communications technologies (ICT) (Figure 8-6; Appendix Table 8-4 through Appendix Table 8-10).
Faster growth of patents granted to non-U.S. inventors reduced the U.S. share from 51% in 2006 to 47% in 2016 (Figure 8-5; Appendix Table 8-4). The increase in foreign patents reflects globalization, as foreign firms file their existing patents in multiple jurisdictions (Fink, Khan, Zhou 2015). Large multinational companies, including those based outside of the United States, are increasingly seeking patent protection beyond their domestic borders.
The pattern of this globalization of USPTO patents has been uneven. Japan’s share fell, and the EU’s share remained steady between 2006 and 2016 (Figure 8-5; Appendix Table 8-4). Patenting activity in the Asian economies of South Korea, China, and India increased strongly over the last decade (Figure 8-7; Appendix Table 8-4). South Korea’s share doubled to reach 6%. China’s patenting activity grew the fastest, although from a low base, resulting in its share rising from 1% to 4%. India also grew from a low base, with its share reaching 1%. Although the number of patents more than doubled between 2006 and 2016, Taiwan’s global share remained at 4% during this period.
USPTO patents granted in selected broad technology categories: 2006 and 2016
ICT = information and communications technology; USPTO = U.S. Patent and Trademark Office.
Note(s)
Patents are classified under the World Intellectual Property Organization (WIPO) classification of patents, which classifies International Patent Classification (IPC) codes under 35 technical fields. IPC reformed codes take into account changes that were made to the WIPO classification in 2006 under the eighth version of the classification and were used to prepare these data. However, because PatentsView only provides the original IPC codes as they appeared on patents and not the IPC reformed codes, current Cooperative Patent Classification codes on patents were converted back to the most recent IPC classification to prepare these statistics. Fractional counts of patents were assigned to each technological field on patents to assign a proper weight of a patent to the corresponding technological fields under the classification. Patents are fractionally allocated among regions, countries, or economies based on the proportion of residences of all named inventors.
Source(s)
Science-Metrix; PatentsView; SRI International, accessed December 2016. See Appendix Table 6-37 through Appendix Table 6-48.
Science and Engineering Indicators 2018
USPTO patents granted, by selected country or economy of inventor: 2006–16
USPTO = U.S. Patent and Trademark Office.
Note(s)
China includes Hong Kong. Patent grants are fractionally allocated among regions, countries, or economies based on the proportion of the residences of all named inventors.
Source(s)
Science-Metrix; LexisNexis; SRI International. See Appendix Table 6-37.
Science and Engineering Indicators 2018
Patenting in selected technologies
This section discusses patterns and trends of four technology categories that are closely linked to science or the knowledge- and technology-intensive industries described in Chapter 6: ICT; testing, measuring, and control; chemistry and health; and materials and nanotechnology (Table 8-2). The patent count data by country for some of the 35 WIPO patent fields shown in Appendix Table 8-2 are reorganized into these four broader categories. The ICT category, consisting of six technologies, has the largest share of USPTO patents (37% of all USPTO patents in 2016) (Figure 8-6). Patents granted in these fields are shown in Appendix Table 8-5 through Appendix Table 8-10. Of ICT, computer technology is the largest in terms of USPTO patent share (14%), followed by digital communication (10%), semiconductors (6%), and telecommunications (4%). The next largest category is chemistry and health (16%), consisting of seven technologies; medical technology has the largest share (6%) among these technologies. Patents granted in these fields are shown in Appendix Table 8-11 through Appendix Table 8-17. The third largest (11%) is testing, measuring, and control, consisting of four technologies (Appendix Table 8-18 through Appendix Table 8-21). Materials and nanotechnology, consisting of three technologies, has a far smaller share (2%) (Appendix Table 8-22 through Appendix Table 8-24).
Selected technology areas of USPTO patents
The number of ICT patents nearly doubled between 2006 and 2016, the fastest growth of these four technology categories. The ICT share of all patents increased from 31% to 37% during this period (Figure 8-6). For example, the average smartphone, which uses a wide variety of ICT, is covered by around 250,000 patents, up from 70,000 patents in 2003 (Reidenberg 2015). In addition, patents of technical standards, guidelines, or specifications that govern the interaction of technologies in products, processes, and services, grew rapidly. Technical standards are used widely in ICT technologies, including smartphones (see sidebar Technical Standards, Invention, Innovation, and Economic Growth). The growth in ICT patents over the last decade was led by digital communication (195%) and computer technology (126%).
Patents in the chemistry and health category grew slightly faster (84%) than all patents (75%) between 2006 and 2016 (Figure 8-6; Appendix Table 8-11 through Appendix Table 8-17). Medical technology patents grew the fastest among this category (140%), resulting in its share of all patents rising from 4% to 6%. Two other technologies—pharmaceuticals (Appendix Table 8-13) and basic material chemistry (Appendix Table 8-14)—also had strong growth.
Patents in the testing, measuring, and control category grew significantly slower than all patents (43%) over the last decade (Figure 8-6; Appendix Table 8-18 through Appendix Table 8-21). Within this category, patents in analysis of biological materials grew the fastest (84%), albeit from a very low base. Patents in control technology grew modestly (74%).
Patents in the materials and nanotechnology category grew slower than all patents over the last decade (Figure 8-6; Appendix Table 8-22 through Appendix Table 8-24). Patents in materials and metallurgy and in surface technology and coating had modest growth; patents in microstructural and nanotechnology grew slightly (10%).
Country-level concentration in patenting technology areas
In contrast to growth rates, patent activity indexes provide insight into the areas where each country is concentrating its patenting activity. As noted previously, many factors, including industry-level propensity to patent and patent litigation, influence patenting activity. This section presents patent activity indexes of the United States, the EU, and several Asian economies in these technologies averaged for 2014–16, based on analysis of USPTO data. The Patenting Activity Index indicates the extent to which a country’s patents are concentrated in a particular technology. It is an output measure of specialization, assessing the share of a country’s patents produced in each technological area. The indicator is computed by comparing a country to the global average (see sidebar Patent Data Analytics and Terminology). Technologies with an activity index of 1.2 or more are defined here as relatively more concentrated.
Patenting in the United States is relatively more concentrated in six technologies (Figure 8-8; Appendix Table 8-25). Three of these are in the chemistry and health category—medical technology, pharmaceuticals, and biotechnology. The United States has a high concentration in analysis of biological materials, a technology classified in the testing, measuring, and control category, that is closely related to the chemistry and health category. The concentration of U.S. patenting activities in pharmaceuticals, analysis of biological materials, and biotechnology coincides with the strong U.S. market position in and considerable R&D investment in pharmaceuticals. The U.S. concentration in medical technology and control technologies coincides with a strong market position in testing, measuring, and control instruments. U.S. patenting is concentrated in one technology in the ICT category, information technology (IT) methods for management, which consists of business methods and software methods for data processing.
Patent activity index for selected technologies for the United States, the EU, and Japan: 2014–16
EU = European Union; IT = information technology.
Note(s)
A patent activity index is the ratio of a country’s share of a technology area to its share of all patents. A patent activity index greater (less) than 1.0 indicates that the country is relatively more (less) active in the technology area. Patents are classified under the World Intellectual Property Organization (WIPO) classification of patents, which classifies International Patent Classification (IPC) codes under 35 technical fields. IPC reformed codes take into account changes that were made to the WIPO classification in 2006 under the eighth version of the classification and were used to prepare these data. However, because PatentsView only provides the original IPC codes as they appeared on patents and not the IPC reformed codes, current Cooperative Patent Classification codes on patents were converted back to the most recent IPC classification to prepare these statistics. Fractional counts of patents were assigned to each technological field on patents to assign the proper weight of a patent to the corresponding technological fields under the classification. Patents are fractionally allocated among regions, countries, or economies based on the proportion of residences of all named inventors.
Source(s)
Science-Metrix; PatentsView; SRI International, accessed April 2017.
Science and Engineering Indicators 2018
The EU’s USPTO patenting is relatively more concentrated in nine technologies, with six that are in the chemistry and health category (Table 8-2; Figure 8-8; Appendix Table 8-25). The EU has a relatively high concentration in organic chemistry (1.7) and relatively high concentrations (1.3–1.6) in six other technologies: macromolecular chemistry, chemical engineering, biotechnology, pharmaceuticals, and basic material chemistry. The relatively high concentration in pharmaceuticals and biotechnology coincides with the EU’s strong market position in pharmaceuticals. In the testing, measuring, and control category, the EU is relatively more concentrated in measurement (1.4), which is consistent with the EU’s relatively strong market position in testing, measuring, and control instruments. The EU is relatively less concentrated in all technologies in the ICT category.
Japan’s concentration of USPTO patenting is far different from that of the United States or the EU. Japan has a very high concentration in optics (2.7), a technology in the testing, measuring, and control category, coinciding with dominance of Japanese-based companies in photography and imaging, including Canon, Fujifilm, Nikon, and Olympus (Table 8-2; Figure 8-8; Appendix Table 8-25). Japan has a moderately high concentration in two technologies in the ICT category—semiconductors (1.5) and telecommunications (1.2)—despite its considerable loss of market share in these two industries. Japan has a relatively high concentration in two technologies in the materials and nanotechnology category—surface technology and coating (1.4) and materials and metallurgy (1.5).
South Korea is relatively more concentrated in four technologies in the ICT category—semiconductors, digital communications, telecommunications, and basic communication processes (Table 8-2; Figure 8-9; Appendix Table 8-25). South Korea’s concentration in patenting of these technologies, particularly semiconductors, coincides with its strong market position in the ICT manufacturing industries of semiconductors and communications. South Korea, like Japan, has a relatively high concentration in optics.
Taiwan has a high concentration in two technologies in the ICT category: semiconductors, coinciding with its very strong market position in the semiconductors industry, and basic communication processes (Figure 8-9; Appendix Table 8-25). Taiwan, like Japan and South Korea, has a relatively high concentration in optics. Taiwan has a relatively high concentration in microstructural and nanotechnologies in contrast to the relatively low concentrations of the United States, the EU, Japan, and South Korea.
China has a relatively high concentration in four technologies, including two technologies in the ICT category—telecommunications and digital communications (Table 8-2; Figure 8-9; Appendix Table 8-25). China has a lower concentration in semiconductors. This is consistent with its technological development, where its industry lags behind firms based in South Korea, Taiwan, and other countries. China, like Taiwan, has a relatively high concentration in microstructural and nanotechnologies.
Patent activity index of selected technologies for South Korea, Taiwan, and China: 2014–16
Note(s)
A patent activity index is the ratio of a country’s share of a technology area to its share of all patents. A patent activity index greater (less) than 1.0 indicates that the country is relatively more (less) active in the technology area. Patents are classified under the World Intellectual Property Organization (WIPO) classification of patents, which classifies International Patent Classification (IPC) codes under 35 technical fields. IPC reformed codes take into account changes that were made to the WIPO classification in 2006 under the eighth version of the classification and were used to prepare these data. However, because PatentsView only provides the original IPC codes as they appeared on patents and not the IPC reformed codes, current Cooperative Patent Classification codes on patents were converted back to the most recent IPC classification to prepare these statistics. Fractional counts of patents were assigned to each technological field on patents to assign the proper weight of a patent to the corresponding technological fields under the classification. Patents are fractionally allocated among regions, countries, or economies based on the proportion of residences of all named inventors.
Source(s)
Science-Metrix; PatentsView; SRI International, accessed December 2016.
Science and Engineering Indicators 2018