This paper looks at BRAC’s Rural Development Programmes’ (RDP) interventions and consumption based poverty using household expenditure data collected from 3518 households in fourteen villages in Matlab between April and August 1995. Poverty lines and measures are constructed to compare the socioeconomic status of BRAC members relative to non-members. These poverty measures are also used to compare BRAC members according to membership length, type of inputs received and loan size. The bivariate analysis suggests that the poorest members of the village are selected towards the beginning of RDP’s operations with newer members coming from the moderate poor group. The multivariate analysis focuses on the relative impact of RDP’s inputs, particularly credit, and caters for the crucial issue of selectivity bias. The results suggest that whilst borrowing beyond a certain loan threshold (10000 taka) is significantly associated with improvements in welfare for the ‘typical’ BRAC member this result does not hold for the poorest members.
Authors: Zaman, Hassan
Type: Working Paper
Year: 1997