Can reference budgets be used as a poverty line?

The most common indicator to measure and compare the extent of poverty within and across European countries is the well-known at-risk-of-poverty indicator. Although the relative income-based measure is widely used, over time it has been the target of considerable criticism. In this paper, reference budgets are introduced as a valuable complementary indicator, since they illustrate the cost of baskets of goods and services that are essential to participate adequately in society. When constructed in a comparable way, they show which standard of living can be achieved at the level of the at-risk-of-poverty threshold in different countries, taking account of out-of-pocket costs of public goods and services. In this paper, we draw on data from cross-nationally comparable reference budgets in three reference cities (Antwerp, Barcelona and Helsinki) to illustrate how RBs can be used to evaluate other poverty indicators and to construct complementary poverty thresholds. At the same time, we explain that there are important challenges to address, including (1) the limited number of specific household types for which reference budgets are developed, (2) problems of robustness and comparability, and (3) the lack of important information in the EU-SILC microdata for our purposes. Acknowledging these limitations, this paper provides a first illustrative attempt to estimate of the number of people with a disposable income below the RB threshold for densely populated areas in Belgium, Finland and Spain.The results show how the at-risk-of-poverty threshold does not represent the same level of living standard across EU Member States, and probably underestimates the out-of-pocket cost of an adequate living standard in Spain. First estimates indicate that families renting on the private market, families with children and young people are relatively worse off when poverty is measured with the reference budget indicator as compared to the at-risk-of-poverty indicator. If these results are confirmed in future research, this may have important implications for anti-poverty policies.