ON THE 'PRO-POORNESS' OF GROWTH IN A MULTIDIMENSIONAL CONTEXT.

This paper represents a first attempt to bring together the issues of multidimensional poverty and growth 'pro-poorness' assessments. More specifically, we suggest the use of sequential dominance procedures to test the 'pro-poorness' of observed growth spells when poverty is measured on the basis of income and another discrete well-being attribute. Sequential procedures are also used to obtain graphical tools that are consistent with the spirit of Ravallion and Chen's growth incidence curve and Son's poverty growth curve. Contrary to traditional unidimensional tests, our method makes it possible to take into account the importance of deprivation correlations at the individual level and thus may reverse results observed with the traditional tools used to check the 'pro-poorness' of growth. An illustration of our approach is given using Turkish data for the period 2003-05. [ABSTRACT FROM AUTHOR]