Karel van den Bosch - Dynamic microsimulation of income inequality and poverty risk for government
Presenting author: Karel van den Bosch (Belgian Federal Planning Bureau)
Authors: Gijs Dekkers, Karel van den Bosch
Session: C02A - Dynamic / Long term [3] - Wednesday 11:00-12:30 - Ceremonial Hall
Slides: PDF
MIDAS is a dynamic microsimulation model (DMSM) developed at the Belgian Federal Planning Bureau with the aim to produce long-term (up to 2070) forecasts of income inequality and poverty risk among pensioners, the elderly and the population in general. The results show the future consequences of current pension and social security policies, given projected demographic and economic developments. MIDAS is a DMSM of the cross-sectional type, runs on a starting data set from administrative sources of 553K individuals and encompasses demographic, labour-market, earnings, social benefits and tax modules. To achieve the purpose of credible projections which are relevant for policy making, two main strategies are used. First, the model is aligned to external projections of the size and age structure of the population, immigration and emigration, household composition, employment, unemployment (as well as other socio-economic categories, broken down by age and sex) and average wage growth. Second, within this demographic and macro-economic framework, existing social gradients as observed in the starting data are maintained as much as possible. While the model is in operation since 2009, it has recently been thoroughly revised to enhance its validity, credibility and scope. Changes include the modeling of migration, alignment on projections of household composition, the revision of the labour market module and its behavioural equations, the inclusion of region-specific child benefits, and, finally, a number of benefits. The revision has made it possible to produce projections of poverty risks for the whole population and more detailed age groups, while earlier versions were focused on the elderly and pensioners. The presentation will offer more details as well as key results. It will also discuss the relevance of projections of income inequality and poverty risk over a period of nearly 50 years.