Changing Times, Testing Times: A Bootstrap Analysis of Poverty and Inequality using the PACO Database

Conclusions about poverty and the distribution of incomes are typically based on information obtained from sample surveys. However, sample surveys are subject to sampling and non-sampling errors. Statistical inference allows us to deal with sampling errors. In this paper, we demonstrate the usefulness of bootstrapping techniques for carrying out statistical inference for poverty and inequality measures. We analyse poverty and income inequality among pensioners in Hungary, Luxembourg and the United Kingdom. We carry out comparisons of the living standards of pensioners across countries and over time. Our results have far-reaching policy implications for the reforms of public pension systems currently under way.