The Statistical Analysis of End of Working Life: Issues raised by the Average Effective Age of Retirement Indicator

The paper focuses on the Average Effective Age of Retirement (AEAR) indicator allowing policy makers and scientists to evaluate potential gaps between the age of the end of the working life and the pension age in European countries. Used at the European level after the Barcelona European Council of March 2002, this indicator provided by the Organisation for Economic Co-operation and Development raises a number of problems that are reviewed in this article. First, we show that employment rates of older workers dropped continuously in all European countries since 1970. From the mid-1990s and much more in the last 10 years, following European targets designed to increase older workers employment rates, the AEAR increases. Second, we analyse the formula of the AEAR as such. We show that the “employment rate” is the main component of its calculation. Using data from the European Social Surveys a new indicator based on individual data is provided. Last, using Labour Force Surveys, we develop at a macro-level a dynamic reading of labour market taking into account transitions into and outside employment. In addition, we focus particularly—at a micro-level—on individual factors via a multinomial logistic regression performed for Belgium, France and the Netherlands, particularly characterised by high rates of early retirement since the 1970s. We show that uses of exit arrangement vary depending on to gender, the level of education, and the professional and marital status.SCOPUS: ar.j