infonet-economy

Until the late 1980s, textbooks portrayed economics as a nonexperimental science because it was thought that “Economists…cannot perform the controlled experiments of chemists or biologists.…Like astronomers or meteorologists, they generally must be content largely to observe” (1). Since then, economics has experienced an experimental revolution (2–6). However, there has been a debate on the extent to which insights from economic lab experiments can be generalized to field settings (7–11). On page 545 of this issue, Herbst and Mas (12) show that the results of a class of lab experiments can be generalized to the field because they provide quantitatively precise descriptions of productivity spillovers between workers.