The ratio of total academic R&D expenditures to publications increased from the late 1980s through 2006 and remained more constant after that time. In contrast, the ratio of total publications to researchers remained relatively stable during 1994–2011. With some variation, these trends prevailed across all types of academic institutions. These trends do not necessarily indicate diminishing returns for research expenditures, as this paper will discuss.

Trends varied by field in the ratio of R&D spending to publications. Life sciences—which has long accounted for the majority of academic R&D expenditures, a sizeable share (about one-third) of the academic doctoral workforce, and more than one-half of publications—shows a trend close to the overall pattern, with an upward trajectory in the spending-to-publications ratio in most years. After adjustment for inflation, life sciences showed the greatest increase in this ratio from 1994 to 2011 (81%). In contrast, the spending-to-publications ratio for physical sciences was nearly flat (11%). Social sciences and engineering both saw a moderate increase in the spending-to-publications ratio (roughly 20% in each field).

Overall, the ratio of articles to researchers was largely flat during the period of analysis, although there was some variation by field. There were slight increases in the publications-to-researcher ratio in social sciences, engineering, and physical sciences, particularly at public institutions. The publications-to-researcher ratio in the life sciences dropped slightly during this time, especially at private institutions.

The trends reported here do not necessarily indicate changes in researcher productivity. Many other interpretations of these trends are possible. Further research is needed to explore the broad array of factors that have affected academic S&E research and publication practices and to analyze their impact on the relationships reported in this paper.