Bottom incomes and the measurement of poverty and inequality

Incomes in surveys suffer from various measurement problems, most notably in the tails of their distributions. We study the prevalence of negative and zero incomes, and their implications for inequality and poverty measurement relying on 57 harmonized surveys covering 12 countries over the period 1995-2016. The paper explains the composition and sources of negative and zero incomes and assesses the distributional impacts of alternative correction methods on poverty and inequality measures. It finds that the main source of negative disposable incomes is negative self-employment income, and that high tax, social security withholdings and high self-paid social-security contributions account for negative incomes in some countries. Using detailed information on expenditure, we conclude that households with negative incomes are typically as well off as, or even better, than other households in terms of material wellbeing. By contrast, zero-income households are found to be materially deprived. Adjusting poverty and inequality measures for these findings can alter these measures significantly.