T N Ninan: A fresh poverty line

The Rangarajan committee, set up two years ago to take a fresh look at poverty issues, has according to a newspaper report recommended a radically new way of deciding who is officially poor. New in India, that is, because the approach that the committee has reportedly recommended is already in practice in the US and Europe: poverty is defined, not as an absolute level of income, but relative to the average levels of income. The US says all those whose income is 40 per cent or less than the average (median, not mean) are poor; Europe is more generous, and puts the cut-off at 60 per cent of the median income.

Such an approach has been advocated repeatedly in these columns, and would allow sanity to enter the poverty debate. Given that incomes are low in India, the poverty line should be in line with the more generous European model: 60 per cent of median income (properly measured, not what is reported by the National Sample Survey). This would place India's official poverty line much higher than it has been so far. An informed hunch would be that a five-member family would be officially poor if its monthly income fell below Rs 6,000 in a rural area, and Rs 11,000 in a big city.

India's official poverty line has been re-defined quite a few times. By the 1993 definition, the officially poor in India would have been about 11 per cent of the population in 2011-12. Such a tall claim would be hard to support. By the 2005 definition (which was broadly equivalent to the internationally accepted benchmark of $1.25 per day), the figure was 22 per cent. The new Rangarajan formula is likely to take the number into the 30 per cent-plus range - which sounds more real. The new method has two additional merits. First, the poor by definition will be much less than half the population, so targeted handouts will go to the needy and not to others. Second, as national income rises, it automatically raises the poverty line. So, at any level of prosperity, there will be poor who need to be taken care of.

There is another issue. Poverty-level income (or expenditure) should be measured on the basis of households, not individuals - because the spending unit is the household, not the individual. A five-member family would find it easier to manage on Rs 10,000 per month than a two-member family on Rs 4,000, though both households have the same level of per capita income. But data matching income or expenditure to size of household do not exist, and will have to be worked out. The subsequent task of enumerating which household falls into what category is even more fraught with problems - because every state wants to exaggerate the numbers of its poor. Some states have issued more BPL (below poverty line) cards than their total households.

The finance minister has talked of tackling the subsidy problem. The key to doing this is to have a sensible poverty line, and to make it the basis for deciding who gets government handouts and who does not. It is ridiculous, for instance, that the food security law stipulates a 90 per cent grain subsidy for half the urban population and three-quarters of the rural population (that's 850 million people). Again, the 29 per cent households who use cooking gas are likely to be better off than the 69 per cent households who use firewood, crop residue or cowdung cake. None of these latter gets any subsidy, so why massively subsidise cooking gas? Proper identification and targeting are the real challenges, if the subsidy bill is to be contained. This won't be a cakewalk, but who said governance was easy?