ABSTRACT
This chapter compares two approaches for making forecasts about the future of old-age poverty in three exemplary countries: one approach relies on conventional methods of index-building, the other on microsimulation. Both are using the same International Social Survey Programme (ISSP) data and the same parameter estimates, which describe the causes of vulnerability by retirement. The comparison of the two methods shows that the returns form microsimulation are worth the method-specific extra-effort. Social policy aims at fighting social problems like unemployment, work-accidents or poverty after retirement, which typically mark planned or unplanned transitions in the individual life course. For demographic variables referring to different marital statuses, the assumption of the stability of the composition of cohorts is less realistic and thus only acceptable if the forecasted time period does not exceed 20 years and the interpretation of the derived early warning indicators is rather qualitative than quantitative.
