The life-cycle research productivity of mathematicians and scientists.

Declining research productivity with age is implied by economic models of life-cycle human capital investment but is denied by some recent empirical studies. The purpose of the present study is to provide new evidence on whether a scientist's output generally declines with advancing age. A longitudinal data set has been compiled for scientists and mathematicians at six major departments, including data on age, salaries, annual citations (stock of human capital), citations to current output (flow of human capital), and quantity of current output measured both in number of articles and in number of pages. Analysis of the data indicates that salaries peak from the early to mid-60s, whereas annual citations appear to peak from age 39 to 89 for different departments with a mean age of 59 for the 6 departments. The quantity and quality of current research output appear to decline continuously with age.