ABSTRACT
Talking about statistics in connection with poverty can often provoke reactions of rejection: investing in constructing a means of observing poverty or exclusion would be a waste of resources, which would be better spent on relieving hardship. However, providing effective help for people in difficulty, and directing aid initially towards those who need it the most, requires that we can identify them and know the specific nature of their behaviour. This is exactly what we would expect from a statistical system. But do existing systems currently provide information that is relevant for this purpose? In a ‘rich’ country, statistical knowledge of populations in difficulty, although in constant progression, still leaves significant areas in shadow. In a ‘poor’ country, one of the ‘least-developed countries’ (LDCs),1 with a less comprehensive statistical system, is it possible to imagine an affordable system that could give useful information on the populations facing the greatest hardship, the ‘poorest of the poor’? It is clearly out of the question to suggest a system along French lines, but it is much more reasonable to review the experience of measuring poverty in European countries to see what lessons might be learned for the development of a system that would be appropriate for the LDCs: which basic concepts should be adopted? What are the pitfalls to be avoided?
