Meaning versus measurement: why do 'economic' indicators of poverty still predominate?

It is virtually undisputed that poverty is multi-dimensional. However, 'economic' or income/expenditure/monetary measures of poverty still maintain a higher status in key development indicators and policy: The number one Millennium Development Goal (MDG) is the dollar-a-day; the Human Development Index (HDI) and Gender Development Index (GDI) of UNDP both use weightings that very strongly favour GDP per capita; and the income poverty headcount is generally the main focus in Poverty Reduction Strategy Papers. This article is concerned with this apparent contradiction between the consensus over the meaning of poverty and the choice of methods with which to measure poverty in practice. A brief history of the meaning and measurement of poverty is given and it is argued that while 'economic' determinism has gradually retreated from centrality in the meaning of poverty it has continued to dominate the measurement of poverty. This is followed by a section that contrasts the relative merits of 'economic' and 'non-economic' measures of poverty. The question is posed: why do 'economic' measures of poverty still have a higher status than non-economic measures?