As described in the introduction, R&D activities from all parts of society create new knowledge (Image INV-1, see Introduction). A large share of this new knowledge is intended as a public good and disclosed to the scientific community and to the public via scientific literature. This activity is described in the Indicators report “Publications Output: U.S. Trends and International Comparisons.” A subset of new knowledge takes the form of inventions, which are the development of new and useful processes, machines, manufactures, or compositions of matter, or any new and useful improvements. Examples of invention include mechanical devices, new materials, new processes, and software programs. Inventors include researchers in government, academia, private nonprofit institutions, business firms, and individuals working independently.

The distinction between invention and innovation is important and often overlooked. Just as invention represents a subset of newly created knowledge, innovation represents a subset of invention. It is innovation that brings new products and processes into our lives. While all inventions have the potential for practical use, few are ever put to use in operational settings (Shepherd 1979). A new product or process becomes an innovation when it has been made available to potential users or brought into use (OECD Eurostat 2018). The mechanisms by which inventions are refined into innovations are complex and nonlinear, subject to feedback loops, reversals, and unforeseen, often surprising areas of application (Rosenberg 1983). A single invention can have far-reaching, multiplicative impact, not just on innovation but on further invention and scientific knowledge creation.

Governments grant exclusive rights to inventions to create incentives for organizations and individuals to conduct R&D and other activity that produces inventions and brings them into use. To qualify for the legal protection of a patent, inventions must be more than merely useful; they must also satisfy the criteria of being novel and non-obvious. Patent applications, patents granted, and citations in patent documents represent the most common set of indicators of invention. Applications are statements of potential inventions for official review, and patents granted represent official protection that is subsequently granted to that invention, after review. In this report two dimensions of patent activity are shown, patenting based on the residence of the inventor, and based on the institutional owner of the patent rights.

The U.S. federal government grants three types of patents: (1) design patents, which protect the visual characteristics of an invention; (2) utility patents, which protect the way that an invention functions; and (3) plant patents, which are issued for the discovery or invention of new and distinct asexually reproducing plants. In addition to patents, the U.S. Copyright Office provides for the legal protection of integrated circuit layout designs by granting mask works, which are images representing the predetermined, three-dimensional pattern of materials used to form the layers of a semiconductor chip product. Only utility patents, the largest component of patenting, are presented in this report. Trends in USPTO design and plant patents are covered in the Indicators 2020 “Invention, Knowledge Transfer, and Innovation” report section "Invention: U.S. and Comparative Global Trends."

The expense associated with filing, maintaining, and defending a patent can be substantial, and patenting is not the only way that inventions are protected. In some cases, inventors do not seek patent protection because they intend for their creations to be immediately shared widely, or they choose a different form of protection. Many inventions are not patented because they do not qualify as patentable under patent law or because the disclosure required by patent applications places more information in the public domain than companies would prefer. Additionally, inventors may not consider the period of patent protection, which is 20 years from date of application, sufficient to protect their inventions. For these reasons, many commercial entities choose to protect their inventions through nondisclosure agreements and trade secrets rather than via the legal protection conferred by patenting.

Despite the many reasons inventors often do not patent their inventions, the scope of patent data available from the USPTO, the European Patent Office’s PATSTAT database, and WIPO provide extensive and detailed indicators of invention, including a historical record of changing patterns of the organizations that have patented inventions. For example, although the post-World War II period in the United States has been one of dominance for business patenting and a relative decline of individual patenting, USPTO data indicate the absolute number of patents granted to individuals each year remains steadily above 10,000 patents, including 14,373 in 2020, about 8% of patents granted to U.S. owners (Figure INV-1 and Table SINV-1).

USPTO, PATSTAT, and WIPO data and analysis are used in this report to examine U.S. and global trends in patenting by contributor, industry, geography, and technology area. Analysis of global trends reveals how the United States compares internationally as a source of invention. Additional data from the 2018 Business Research and Development Survey, conducted jointly by NCSES and the U.S. Census Bureau, provide indicators of the use of both patenting and trade secrets by businesses, including cross-industry differences in these indicators.

USPTO patent records provide industry, technology areas, and geographical inventor information for all patents registered in the United States, including patents granted to individuals and to foreign entities. More information on the patent data used in this report is available in the Technical Appendix provided with this report.

The USPTO awarded 353,701 utility patents in 2020, nearly equally divided between foreign and domestic owners (Table SINV-1). Obtaining and defending a patent incurs significant costs, and among U.S. assignees (patents that assigned the rights of ownership to a U.S. entity), businesses received by far the most patents (150,558 or 85%) (Figure INV-1 and Table SINV-1). Utility patents assigned to individuals represented 8% of the total in 2020 (Table SINV-1). Academia received 4% of utility patent grants, and the government sector (primarily federal) received 1% of utility patents in 2020 (Table SINV-1). The persistently small contribution of the government is not surprising given that not all government labs emphasize filing patent applications for their internal research staff’s discoveries; instead, federal agencies and institutions are encouraged to openly share the results of their research (Holdren 2013).

The number of patents awarded annually to U.S. patent owners dropped 1% from 2019 to 2020 after increasing 16% in the prior year. This 2020 drop is primarily accounted for by a drop in business patents. In contrast, university patents increased by almost 1% from 2019 to 2020 (Table SINV-1). The number of patents awarded to individuals also increased from 2018 to 2020; however, the share continues a long-term decline (Table SINV-1).

While international patent family data provide an illustration of global trends, USPTO patents focus on inventions, both foreign and domestic, that are granted exclusive use rights in the U.S. market. Of the 353,701 utility patents granted in 2020 by the USPTO, 164,074 were granted to U.S. inventors, or 46%. Foreign inventors were granted 54% in 2020, up from 46% in 2000, continuing the trend of increased internationalization of U.S. patents (Table SINV-44).

Large multinational companies, including those based outside of the United States, have increasingly sought patent protection beyond their domestic borders. As commercial activities increasingly cross international borders, rising rates of foreign patenting reflect, among other factors, the desire of foreign firms to seek patent protection in multiple international jurisdictions (Fink, Khan, and Zhou 2015). The European Union (EU) and Japan continue to account for the largest numbers of foreign USPTO patent grantees (Figure INV-2). Although inventors from China account for a relatively small number of USPTO patents, over the last decade, these have increased more than 10-fold from 1,606 in 2008 to 22,500 in 2020 (Figure INV-2). Patents related to electrical engineering accounted for the majority (63%) of USPTO patents granted to inventors from China (Figure INV-3).

According to a recent USPTO report, the growth in USPTO patents granted to China is driven by both market forces and nonmarket factors. University and government subsidies for patenting can exceed the cost of filing the patent and these subsidies are often higher for patents granted in foreign jurisdictions (USPTO 2021a).

To accelerate the patent approval process for inventions related to COVID-19 that are also subject to Food and Drug Administration approval, the USPTO instituted a program early in 2020 that would allow for prioritized examination without an additional prioritization fee. As of August 2021, 120 patents have been issued under the pilot program (USPTO 2021b).

As a market-driven activity, invention tends to take place where skilled human capital and other inputs are relatively abundant and low cost. As new knowledge spreads, it produces growth benefits or spillovers with a localized dimension that drive additional growth (Grossman and Helpman 1991). States that have relatively high proportions of workers with a bachelor’s degree or higher in science, technology, engineering, and mathematics (STEM) are located primarily on the east and west coasts of the United States and in the Midwest (see Indicators report “The STEM Labor Force of Today: Scientists, Engineers, and Skilled Technical Workers”). As skilled labor and other inputs accumulate, the result can be that a location’s initial advantages grow over time relative to other locations, leading to different growth rates and economic outcomes. Patenting provides an indicator of the location of inventive activity.

County-level patenting intensity is measured here by the location of the owner of the patents granted per 1,000 population in 2020;^{​Population data come from U.S. Department of Commerce, Census Bureau 2021.} counties with high patenting intensity are concentrated in the United States along the coasts and in parts of the Great Lakes region, Texas, and the Rocky Mountains (Figure INV-4). This geographic distribution corresponds with indicators of knowledge- and technology-intensive (KTI) industries, in which California is the single largest producer of KTI industry output, followed by Texas and New York (see forthcoming Indicators report “Production and Trade of Knowledge- and Technology-Intensive Industries”).