A Dynamic Socio-Demographic Microsimulation Model for Austria: General Framework and Application for Educational Projections
Macro-level projections of future population trends are limited to a few variables. This makes microsimulation an interesting modeling option, especially as it allows for modeling of the interaction of demographic with social, environmental and economic variables. Microsimulation can produce useful projections for the analysis of different socio-demographic phenomena considering additional dimensions including educational composition, rural/urban differentials, household structures and family networks, which become increasingly important in the context of the ongoing demographic change. The aim of this thesis is twofold. First, it aims at the development of a general conceptual and technical framework for a dynamic socio-demographic microsimulation model for Austria. Based on a literature review and a survey of existing microsimulation models, conclusions regarding the development for an Austrian model are drown and a general computational microsimulation modeling platform is developed and implemented in object oriented C++. The second aim regards the development of the core behavioral modules of a socio-demographic model for Austria synthesized to a model that is applicable for educational projections. These core modules include models for educational careers, partner matching and the quantum, timing and spacing of births. The models are based on the retrospective event history data collected in the special program of the 1996 micro census that was also used to generate the starting population for projections. The analysis of school choices reveals a very strong influence of the educational attainment of parents leading to strong intergenerational transmission mechanisms within families. In contrast to the ongoing educational expansion at the population level, very stable behavioral relationships on the micro-level can be found when accounting for the parental educational attainment. Education also has a high impact on fertility behavior both regarding quantum and timing of birth and plays a key role regarding partner matching. Based on this analysis, a comprehensive microsimulation model is developed and used to study the effect of intergenerational educational transmission processes within families and fertility differentials between educational groups on the future educational composition of the population. Beside this application for educational projections, both the technical modeling platform and the behavioral models are also seen as main building blocks for a wide range of future applications and extensions ranging from the projection of kinship networks to dynamic tax-benefit, pension and health care models.