Title: Quantifying Lifetime Productivity Changes: A Longitudinal Study of 325,000 Late-Career Scientists

Abstract: This study focuses on persistence in research productivity over the course of an individual's entire scientific careers. We track 'late-career' scientists (N=324,463) in 16 STEMM disciplines (science, technology, engineering, mathematics, and medicine) from 38 OECD countries for up to five decades. We examine the details of their mobility patterns between the top, middle, and bottom productivity classes. Methodologically, we turn a large-scale publication and citation bibliometric dataset into a comprehensive, longitudinal data source for research on careers in science. The global science system emerges as highly immobile: 60% of global top performers continue their careers as top performers and half of global bottom performers as bottom performers. Jumpers-Up and Droppers-Down are extremely rare in science. Our regression analyses show that productivity is highly path dependent: for all disciplines examined, there is a single most important predictor of being a top performer: being a top performer at an earlier career stage.