Microsimulering

Microsimulering, or microsimulation, is a modeling approach used to analyze policy impacts at the individual or household level by applying policy rules to a large synthetic population. It contrasts with aggregate macro models that measure average effects, and it emphasizes distributional outcomes and heterogeneity.

In static microsimulation, each unit’s characteristics are unchanged over time and policy rules are applied to current data to produce outcomes such as after-tax income. Dynamic microsimulation follows individuals over time, updating characteristics (employment, income) as policies and incentives affect behavior and lifecycles.

Data come from microdata sources such as household and tax surveys, census, and administrative records. A synthetic population is created by reweighting or matching data to reflect the target population; models may be calibrated to reproduce known aggregates.

Applications include evaluation of tax-benefit systems, social welfare programs, pensions, labor-market interventions, and health or housing policies. Microsimulation is widely used by government agencies, research institutes, and international organizations to study distributional effects and poverty.

Advantages include detailed insight into how policies affect different groups; limitations include data demands, assumptions about behavior, computational requirements, and potential privacy concerns. Some models are publicly available (for example EUROMOD for EU tax-benefit analysis).

Related topics include tax-benefit microsimulation, dynamic microsimulation, and synthetic populations.