<p>Consensual deprivation indicators assume that there is a broad consensus on what goods/services families should be able to afford, and that an inability to afford those items can measure deprivation. Using data from two British surveys in 1999, this paper makes two arguments. First, there is only limited agreement about which items families should be able to afford. Secondly, different social groups are more (or less) likely to say the absence of a 'necessity' is due to choice. Families who cannot afford two or more 'necessities' invariably have a number of 'nonnecessities', often many. Their patterns of preferences (and spending) are not typical and they are choosing to buy other goods Ö through preference rather than poverty. Simply checking whether people lack items for any reason provides results empirically as reliable, but subject to similar criticisms.</p>