Skip menu navigation Home > Listing of PTMT Viewable Reports > U.S. Government Patents Report Table of Contents > Fractional Counts By NAICS / Whole Counts By NAICS > This Page

Patent Technology Monitoring Team (PTMT)


- this report has been prepared with support from the National Science Foundation -





Patents Included in These Tables

The patents used to prepare these tables are U.S. utility patents (i.e., patents for inventions) granted during the period 1963 to 2008. Plant patents and design patents are not included.

Patent Counts By NAICS-Based Product Fields

Each U.S. patent is classified in one or more subclases of the U.S. Patent Classification System (USPC) dependent on the information claimed and disclosed in the patent. One of these subclasses is designated as the primary classification of the patent (called the 'original' classification in U.S. Patent and Trademark Office-specific terminology) and the remainder, if any, are designated as 'cross-reference' classifications.

The U.S. Patent and Trademark Office (USPTO), Patent Technology Monitoring Team (PTMT), has established a concordance between the USPC and 30 product fields based on the 1997 North American Industry Classification System (NAICS). The Concordance has been used to provide patent data, organized by product field, to the National Science Foundation (NSF), Science Indicators Unit. The report, "Patenting Trends in the United States, 2008," produced for the NSF, contains patent counts by product field and is the source of the data appearing in these tables. Because the USPC is continually being updated and because all existing patents are reclassified to reflect changes to the USPC, NAICS-based or Standard Industrial Classification System (SIC) based product field data from reports produced for NSF in previous years may not exactly match the data presented in these tables.

In the tables that display patent counts by NAICS-based product fields, only a patent's primary classification, as of December 31, 2008, is considered.

Product Fields

These tables profile patent activity for the years 1963 to 2008 in each of 30 NAICS-based product fields. The 30 product fields consist of 26 unique product fields and 4 product fields that are combinations or 'roll-ups' of the unique product fields. The tables also include a product field titled 'All Industries', which is a roll-up of the 26 unique product fields.

Each product field patent count in these tables is generated by identifying USPC subclasses which are pertinent to that product field. The patents in those subclasses (based each patent's primary classification) then are included in that profile. This procedure results, in most cases, in the inclusion of the majority of patents relevant to the NAICS-based product field and few, if any, irrelevant patents. However, since the match between the NAICS system and the USPC system is imperfect, the profiles cannot be considered to include all relevant patents, nor to exclude all irrelevant patents. More detail on the Concordance may be found in "Review and Assessment of the OTAF Concordance between the U.S. Patent Classification and the Standard Industrial Classification System: Final Report," OTAF, 1984, available from PTMT.

Fractional Counts and Whole Counts

There are two sets of tables with patent counts by NAICS-based product field. One set of tables uses a 'Whole Counts' methodology to count patents while the other set uses a 'Fractional Counts' methodology for patent counting.

With the 'Whole Counts' methodology, a single patent may be fully counted in as many as seven unique NAICS-based product fields in the corresponding tables. With the 'Fractional Counts' methodology, this multiple counting of patents across product fields is eliminated. Further explanation follows.

The USPC to NAICS Concordance assigns USPC patent subclasses to all (up to seven) identified NAICS-based product fields to which they are pertinent. With the 'Whole Counts' methodology, a patent is counted in a NAICS-based product field if the patent's primary USPC subclass (called the 'original' subclass, in USPTO-specific terminology) is matched, via Concordance, to that product field. With the 'Whole Counts' methodology, for example, if a patent has primary classification in a USPC subclass which is matched to 3 unique NAICS-based product fields, that patent would be counted once in each of those three associated NAICS-based product fields. Consequently, that single patent would result in an effective total count of 3 patents if the patents in the three product fields are summed together (i.e., for the 'Whole Counts' tables, the patent count for multiple product fields summed together would total more than the actual number of patents).

The 'Fractional Counts' methodology eliminates this multiple counting of patents by dividing each patent equally among the NAICS-based product fields to which the patent's primary USPC subclass has been matched, via the Concordance. Using the example above, one-third of a patent count would be added to each of the 3 product fields matched with the patent's primary USPC subclass.

Note that listed 'Fractional Counts' are rounded to the nearest whole number. A zero represents a count of less than 0.5, which has been rounded down. Blanks indicate a zero count. For a given year displayed in a fractional counts table, the patent counts for the unique product fields, when summed, should equal the total counts for that year (except for minor differences due to rounding). Because of rounding, the patent counts across rows may not add exactly to the totals shown in the columns on the right.

'Patent Grant' Data and 'Patented Application' Data

The patent counts in these tables are distributed both by the year in which the patents were granted (Table 1) and by the year in which the applications for those patents were filed (Table 2). When distributed by year of patent grant, the data are referred to as 'patent grant' data. When distributed by year of application, the data are referred to as 'patented application' data. The tables do not contain data for patent applications that were filed but did not result in a patent grant during the period 1963 to December 2008. Roughly 35% of all applications filed never result in a patent grant.

Patent grant data are affected to some extent by PTO operational factors. There is always some delay (known as 'pendency') between the filing of a patent application and its subsequent issuance as a patent. The length of this delay is determined by many factors, including PTO workload, budget and manpower levels, and patent printing schedules. Pendency varies between technologies and from year to year.

The 1986 and 2005 patent grant counts are lower than would have normally been expected due to such administrative factors. This is an example of how PTO operating procedures can affect patent grant data.

Patented application data also can be affected by external factors, though usually to a lesser degree than for patent grant data. For example, the number of patented applications for 1995 is higher than would normally have been expected and the number of patented applications for 1996 is lower than would have normally been expected due to the implementation of patent legislation (GATT) in June of 1995 as applicants hurried to submit patent applications prior to the implementation of the legislation.

Please note that patented application data for the years 2000 to 2008 are incomplete because many applications filed from 2000 to 2008, which will ultimately become patents, were still pending in December of 2008 (patent counts included in these tables are limited to patents granted only through December 31, 2008). For utility patents granted through December 2008, patent grant data, as distributed by year of application are approximately 98% complete for utility patent applications filed in 2000, 93% complete for applications filed in 2001, 90% complete for applications filed in 2002, 78% complete for applications filed in 2003, 60% complete for applications filed in 2004, 37% complete for applications filed in 2005, and 17% complete for applications filed in 2006; data are essentially complete for applications filed prior to 2000.

Patent Ownership

The patent counts in the fractional and whole counts NAICS-based product field tables are limited to patents owned by the U.S. Federal Government. Ownership is determined by the assignment of ownership at the time of grant, and corresponds to the information that appears on the front page of the patent. Information relating to subsequent mergers, name changes, or ownership changes is not incorporated into the ownership information that is used to produce these tables.

Please note that ownership information, as contained in these tables, cannot be directly compared with the ownership information contained in "Patenting Trends in the United States" reports published prior to December of 1993. In those prior reports, ownership information was split into two primary categories, 'U.S.-Origin' and 'Foreign Origin', based on the residence of the first-named inventor. In these tables, no such U.S./Foreign 'Origin' distinction is made with regard to ownership; a patent identified as owned by the U.S. Government at the time of grant is not necessarily of 'U.S. Origin'.


PTMT Contacts

Questions regarding these reports should be directed to:

U.S. Patent and Trademark Office
Office of Electronic Information Products - Patent Technology Monitoring Team
P.O. Box 1450
Alexandria, VA 22313-1450

tel: (571) 272-5600
fax: (571) 273-0110

address of PTMT Internet pages :
selected PTMT files available for download at :


Skip menu navigation Home > Listing of PTMT Viewable Reports > U.S. Government Patents Report Table of Contents > Fractional Counts By NAICS / Whole Counts By NAICS > This Page