There is a widespread policy interest in poverty estimates at both national and global level. There has been an explosion of poverty measurement in the last two decades enabled by the growing availability of household survey data. These measurements are used by policy-makers to assess progress towards national and global goals for inclusive growth and poverty reduction. But the evidence base rests on shaky foundations, and policy-makers may have undue confidence in poverty and inequality estimates. Many household surveys are poorly designed to measure monetary living standards and poverty in an era of rising affluence and urban transition. Some key problems in measuring food consumption, housing services and the cost of living are discussed here. Alternatives to monetary measurement, such as using questions on life satisfaction and happiness, also rest on shaky foundations.