Getting down to small areas: estimating regional income and wealth in Australia using spatial microsimulation

Regional policy makers and researchers or national policy makers concerned with the regional impact of their decisions rely on the availability of detailed and current small area data to inform their decision making. However, in the past it has been difficult to assess income levels and inequality, wealth, poverty, disadvantage and other socio-economic characteristics at a detailed regional level. The main source of small area socio-demographic data in Australia is the five yearly Census of Population and Housing conducted by the Australian Bureau of Statistics (ABS). However, the Census contains only limited socio-economic information, is only readily available as a pre-defined series of tables at the Collection District level, and small cells within the tables are randomised by the ABS to protect confidentiality. In addition to the Census, the ABS conducts surveys to collect detailed information on incomes, expenditures and other individual and household characteristics, such as the Household Expenditure Survey, the Survey of Income and Housing Costs and the National Health Survey. However, these surveys typically suppress geographic detail, with 'State or Territory' of residence often representing the most detailed geographic descriptor available. In the past few years NATSEM has developed spatial modelling techniques to create synthetic small area socio-economic data. This is achieved by combining the Census data with that from the ABS national sample surveys. This paper describes the techniques used to create the synthetic small area data. It then provides some examples of how the new methods and data are being used for estimation and analysis of income, wealth and other socio-economic characteristics, as well as the spatial distribution of policy change.