Effect of the Career on Premature Retirement. Using Propensity-Score-Matching to Estimate Causality

Markus Thiede, University of Marburg
Maria Weske, Deutschen Rentenversicherung Hessen and University of Marburg
Martina Schmidt-Stolte, University of Marburg
Ronny Westerman, University of Marburg
Ulrich O. Mueller, University of Marburg

The Effect of the career on Premature Retirement (PMR) will be analyzed with administrative longitudinal data from the Research Data Center of German Federal Pension Insurance. The data include the whole biography till retirement not only for the retired, there is also a sample of still working persons and there biography included. To show the effect of the career for the premature the propensity score matching (PSM) is used to select treatment and control groups. The PSM is an inexact matching method were the conditional probability of receiving the treatment is used. This is necessary for causality analysis with the Rubin-Causality-Model (RCM). The longitudinal data make it possible to show the Average Treatment Effect (ATE) as well as the Average Treatment Effect of the Treated (ATET). These two Methods are helpful instruments for analysis longitudinal datasets.