Choosing a poverty measure: what absolute and relative thresholds capture across wealthy countries

Poverty in wealthy countries can be measured against an absolute threshold, counting people below a fixed standard of need, or a relative threshold, counting people below a moving fraction of national median income. This study provides empirical guidance on that choice for cross-national research. Drawing on 31 high-income OECD countries circa 2018 (29 for income-inequality analyses), it examines, at the country level, how well each kind of measure explains cross-national differences in a composite index of material deprivation and a composite index of economic inequality, across 25 statistical tests and extensive robustness checks. Absolute measures predict material deprivation strongly (r = 0.82) and inequality weakly; relative measures predict income inequality (r ≈ 0.67 to 0.72) while adding little unique information about deprivation beyond what absolute measures already capture. The deprivation result is more robust, holding against an external Eurostat benchmark; on that benchmark relative poverty is not null but tracks hardship roughly half as strongly as absolute poverty. The inequality result is weaker, is clearest for income inequality, and partly reflects relative poverty’s definitional overlap with the income distribution. Findings are cross-sectional and rest on a small sample. For tracking material hardship across countries, this paper supports the absolute measure as the better instrument, and labeling a relative figure simply as “poverty” can imply unmet need where it measures distributional position.