BPL Census and Alternative

In view of clear indications now that the proposed Food Security Act will not be adopting a universal approach, the recent piece by Jean Drèze and Reetika Khera (EPW, 27 February 2010) assumes particular significance. The authors start their paper by pointing out that the proposed below-the-poverty-line (BPL) census is not, strictly speaking, to identify BPL people and hence the term BPL census is actually a misnomer for the exercise. They then discuss in considerable critical detail the ways in which BPL households have been identified for different types of social assistance so far. They go on to examine the alternate ways in which households can be identified for social assistance. The alternate ways considered by them are based on the exclusion criteria, the inclusion criteria and different types of hybrids of these two. They have also shown clearly the implications of these in terms of the percentage of different categories of households covered for social assistance if we follow the different methods of identifying such households. The categories of households considered in this regard include SC/STs, the landless, those with no educated adult, those headed by single women, agricultural labour, those with a low standard of living, and all rural households. Households are also classifi ed into five quintiles based on the wealth index compiled in the National Family Health Survey 2005-06 carried out by the International Institute for Population Sciences. While agreeing almost entirely with the views contained in the paper, one is constrained to point out one or two little refinements to the arguments contained mostly in section five, which has the major chunk of the logic expounded.

The underlying argument regarding the exclusion, inclusion and hybrid approaches to identify households for social assistance can be conveniently put in terms of attributes of households. The attributes under consideration are two – the satisfying of the criteria for exclusion and the satisfying of the criteria for inclusion. We can, in a simple and non-conventional manner, denote by E, the possession of the attribute of satisfying of exclusion criteria with e, denoting the non-possession of the attribute. We denote by I, the possession of the attribute of satisfying of inclusion criteria with i, denoting the non-possession of the attribute. The population N of households can be split into four groups based on these two attributes.