The working poor as a statistical category#Methodological diffiulties and exploration of a notion of poverty in earned income

The concept of working poor may seem easy to grasp, but its statistical implementation doesn’t go without difficulties, one of them being that it requires a specific approach to workers. In addition, given that being a worker is an individual feature while poverty is defined on the basis of variables measured at the household level, the category is defined at the intersection of two statistical units, the individual and the household. The first part of this working paper proposes a comparison of the main definitions of workers used in the statistical approach to the working poor. The objective is to analyse the impact of a change in the criteria used to identify workers on the size and characteristics of the population of working poor, as well as on the analysis that can be done of the causes of their poverty. The second part of the paper focuses on the specificity of the statistical construction “individual/worker poor/household”, which results in many difficulties in the analysis of the phenomenon, since it obliges to disentangle the role of the individual and the household factors. In order to avoid this difficulty, we propose an approach which goes from the individual to the household, based on an indicator of “poverty in market income” defined at the individual level, then by studying whether this “poverty” is counterbalanced by the other private incomes at the household level and by social transfers. The whole study is based on data from EU-SILC 2006 and compares 10 countries of the European union.