Microsimulation & Econometric Data Analysis

More and more people are continuing to work after reaching their retirement age. Proposed reforms such as the German Aktivrente are intended to reinforce this trend, but their impact is difficult to predict. In a feasibility study for the Forschungsnetzwerk Alterssicherung we analyzed if modern microsimulation models can help to forecast retirees’ decisions to stay employed or take up work post-retirement, and can thus provide a sound basis for evaluating policy measures. 

Demographic change is a major challenge for labor markets and pension systems. As the baby boomers retire, pressure on pay-as-you-go pension schemes is growing, as is the shortage of skilled workers in many industries. As a consequence, designing efficient incentives for continued employment in retirement becomes a political challenge.

A key example is the Aktivrente scheme introduced in 2026, which allows for tax-free additional earnings during retirement. It is intended to motivate pensioners to remain in the workforce longer or to return to work after retirement. Whether and to what extent such measures are effective can be assessed in advance only if appropriate analytical tools are available.

On behalf of the Forschungsnetzwerk Alterssicherung of Deutsche Rentenversicherung, we investigated how the available labor market microsimulation models can be enhanced to more accurately reflect the changing behavior of retirees. Our focus was on decisions regarding working hours post-retirement and how they can be systematically captured and integrated into models. Until now, retirement has often been treated as a definitive exit from the labor market – a view that no longer corresponds to the desired reality of growing labor force participation among retirees.

The Socio-Economic Panel data allowed us to develop a model that accounts for different working time options – ranging from non-employment to mini-jobs to extensive employment. The discrete-choice concept let us analyze individual preferences for income or leisure time and derive utility function parameters suitable for simulating the decisions of retirees.

Initial simulations show plausible results that are consistent with findings regarding labor supply effects among the younger working-age population. Our feasibility study demonstrates that microsimulation models can provide a basis for evaluating pension and labor market policy measures, such as Aktivrente, and for assessing their long-term consequences for employment, public finances, and social security systems.