The Bureau of Economic Analysis (BEA) now treats expenditures in research and development (R&D) as investment when estimating U.S. gross domestic product (GDP) and other national income and product accounts statistics.
BEA and the National Science Foundation’s National Center for Science and Engineering Statistics (NCSES) collaborated in developing methodology to use NCSES R&D expenditure statistics for the purpose of estimating U.S. GDP. Changes in GDP methodology, including but not limited to the treatment of R&D, increased historical GDP levels. In turn, U.S. R&D-to-GDP ratios produced by NCSES are now slightly lower, as discussed in a new report from NCSES.
R&D is now capitalized (treated as investment) by BEA in the U.S. GDP and related statistics for all sectors of the economy and included in a new asset category called “intellectual property products.” Private sector R&D expenses, for example, were treated previously as an intermediate production cost.
BEA’s changes in methodology highlight the importance of R&D. GDP is the widest measure of goods and services across economies. Including R&D as an investment in GDP estimates recognizes the long-term benefits and value of new knowledge.
The BEA’s capitalizing of R&D expenditures was part of a set of methodological changes made during a comprehensive revision of its publications released in 2013. Those changes are consistent with the latest international statistics standards adopted by the United Nations Statistical Commission, known as the 2008 System of National Accounts.
A nation’s R&D-to-GDP ratio is often reported as a measure of the intensiveness of its overall research and development efforts and is widely used internationally to compare countries’ overall R&D systems.
As a result of GDP changes, NSF’s NCSES revised its ratio of U.S. R&D expenditures to GDP going back to 1953. Due to higher U.S. GDP estimates and unchanged R&D expenditure levels, this ratio is now modestly lower than previously reported by NCSES.
For more information and statistical products please visit NSF’s National Center for Science and Engineering Statistics.