The impact of many public policies depends on how they affect populations in all their diversity (socio-economic status, location, demographic composition, type of housing, etc.). The micro-simulation approach makes it possible to statistically construct a population in all its diversity using a representative sample and information on each individual or household within that population. Public policies are then simulated using models representing their current state or potential changes, in order to test their impact on the population as a whole.
The micro-simulation approach to public policy analysis has led to the development of three areas of research:
- Statistical analysis, to construct a comprehensive and detailed representation of the population under analysis. This may include data matching, data fusion techniques, or the incorporation of information from household surveys into administratively collected data.
- Public policy modelling, which in this context involves developing algorithms to simulate various public interventions. This most often concerns the social security and tax system, but public expenditure, regulations and standards may also be the subject of such simulations.
- Modelling behavioural responses to policy changes. This involves modelling the extent of reactions to incentives or obligations introduced by public policies. The current state of the art involves using past experience to measure the sensitivity of responses (i.e. elasticities) to policy changes, and then incorporating a behavioural module into the microsimulation model.
The micro-simulation approach was initially developed in the research community in the 1970s, after which these tools were widely adopted by public administrations due to their need to estimate the cost of socio-fiscal reforms, and the cost of maintaining such tools. Nevertheless, research continues to rely on these models to understand how complex public policies affect populations in different ways. Macroeconomic models cannot capture the immense heterogeneity of situations, and statistical empirical approaches struggle to measure the complexity of public policy interventions.
Today, there are numerous models in use around the world that enable international comparisons; to name but a few, these include TAXSIM at the NBER for the United States, TAXBEN at the Institute for Fiscal Studies for the United Kingdom, and EUROMOD at the European Commission. This principle extends beyond socio-fiscal systems to encompass a wide range of dimensions (health, geography, the environment, housing, etc.).
