Modelling the end of mandatory retirement for U.S. medical school faculty

The author used data taken in mid-1989 from the Faculty Roster of the Association of American Medical Colleges to calculate the age distributions of U.S. medical school faculty active at any time from 1978 to 1988 and their age-specific probabilities of separating from their academic positions (e.g., to take another position, to enter the private sector, or to retire). A simple Markov model was constructed to predict the future faculty age distributions, isolating the factors of growth rate, separation probability, and potential delays in retirement. Age projections proved most sensitive to assumptions about future growth rate, less sensitive to the choice of separation probabilities within the range of those observed in the previous decade, and least sensitive to modelled delays in retirement of as much as five years for half the retiring faculty. The same pattern of sensitivity was true of projected turnover. The conclusion is that the end of mandatory retirement alone will have a negligible effect on medical faculty demographics.