Health capacity to work at older ages in France

Health status is one potentially limiting barrier to policies that aim at increasing retirement ages. This paper examines the application to France of two methods that have been proposed in the literature for evaluating how strong is this constraint. Both methods provide measures of a so-called « health-related additional work capacity », i.e. the gap between effective employment rates and the ones that would prevail according to some conventional reference relationship between employment and health status. The first method is based on historical changes in employment rates at given mortality levels, under the strong hypothesis that changes in mortality rates are an adequate proxy of changes in health conditions (Milligan and Wise, 2012). The second one is based on the employment-health relationship observed at a given reference age, using both subjective and objective measures of health status : this relationship is used for inferring the health-related work capacity at other ages, for the same time period (Cutler, Meara and Richards-Shubik, 2013). Both methods suggest that health would not be constraining on the average, but under assumptions whose limits are discussed. We show in particular that messages become different when results are disaggregated by social groups or education. This emphasizes the idea that policies aiming at activating any estimated additional work capacity should take into account, when possible, the heterogeneity of health conditions in the population. Moreover, additional work capacity cannot be a general indicator of how much more seniors should work as the methods used here indeed leave aside many factors that determine the employment rate of older workers.