A platter of choices….Abhijit Banerjee
Why it is better to give the poor access to cash than PDS foodgrain
Why it is better to give the poor access to cash than PDS foodgrain
One of the more charming aspects of being in India is the seeming obsession that everyone has with feeding people: the stewardess on Jet Airways seemed genuinely concerned that I did not finish my murg methiwala. In my half-an-hour at the Planning Commission at least four people asked me if I would like some tea; and the guy who cuts my hair was baffled, as always, by my refusal to take him up on his offer of a cold drink.
It is when it turns into public policy that this fixation with feeding starts to be a problem. I have nothing against the government guaranteeing a minimum standard of living for its citizens quite the contrary. But why does that necessarily have to take the form of the government delivering food at peoples doorsteps? On current evidence, the Indian state has shown little aptitude for getting food to the right people. The governments own Programme Evaluation Organisations recent report on the Targeted Public Distribution System tells us that 36 per cent of the grains intended for the poor somehow vanish along the way. Of this 20 per cent is just somehow lost in transit, while the other 16 per cent is given out against ghost BPL (Below Poverty Line) cards. Not all of the remaining 64 per cent of the grains reach the poor, either. This is because the BPL cards, which entitle one to be targeted, are only slightly more likely to end up in the hands of the non-poor. In 2007, a team from Harvard University and the University of Chicago surveyed 21 households in each of 173 villages in Raichur district in the state of Karnataka. In each of these households they collected the data used for BPL classification by the government and based on that data they constructed their own BPL list. They concluded that only 57 per cent of BPL card-holders would qualify if the official criteria were properly applied. This would imply that only slightly more than a third (57 per cent of 64 per cent) actually reach the eligible. The governments own assessment is only slightly less negative: the same report that already came up puts the fraction reaching the eligible at 42 per cent.
One might be willing to overlook all of this waste, if at least all the poor got reached. In fact the governments report claims that only 57 per cent of the poor have BPL cards, a number confirmed by an eight-village study in Rajasthan by Ritika Khera.
So why does the government insist on feeding everyone? And why are we discussing an expanded PDS under the new Right to Food (RTF) legislation? Why not give people money, so they can buy themselves the food?
One answer that we hear from the RTF supporters is that a lot of the problems will disappear if we stop trying to target. The BPL will all get cards, simply because everyone will get a card. Equally importantly, the middle classes will use their political capital to make the system work better the ration shop owner will not be able to get away with saying that the food somehow didnt show up.
I agree. But what does this have to do with the government doling out food? Wouldnt the same apply to a universal cash transfer, which would relieve the government of the responsibility of moving millions of tonnes of grain around, storing them and fighting off a variety of potential intruders, both small and furry but also of the more biped variety?
Moreover, my sense is that the politically influential middle classes these days may be more willing to fight for their right to cash, than for some grain that is unlikely to be of the highest quality. This was already true when I was a child and we had universal PDS a lot of households would either not pick up their grain or sell them to the poor and now the middle classes are just much richer.
But dont we want people to eat more? Isnt South Asia the malnutrition capital of the world? Absolutely, but we know from the work of Angus Deaton and Jean Dreze (and many others) that this is not primarily because people cannot afford enough nutritious food. Even the children of families that are in the middle of the wealth distribution (based on the admittedly imprecise measures of wealth in the National Family Health Survey) have malnutrition rates that are twice what we find in sub-Saharan Africa. It has something to do with diet, with disease, micronutrients and, unfortunately, the fact that proper nutrition is not enough of a priority for families. For that reason dumping food in the hands of families who would rather have a cellphone may not get us very much further if they dont want the food, they will sell it. And in any case, we could presumably achieve the same goal by giving people food-stamps, which is cash that can only be spent on food (or food and education and healthcare).
The final argument against cash is the one that does worry me. Because everyone likes cash, wouldnt the incentive to defraud the system be that much stronger? With a universal transfer, we avoid the targeting issues, but how about many more ghost BPL cards? This is where tying the whole thing to the UID makes a lot of sense: that would be an obvious way to avoid duplication. Unfortunately the supporters of the RTF dont want that, for the most part it seems, because the UID will take time (though I have heard other, more purely ideological arguments) and they are in a hurry. Given how costly it is to undo a system (especially a system of subsidies) once it has been set up, this seems extraordinarily cavalier.
The writer is Ford Foundation International Professor of Economics and director, Abdul Latif Jameel Poverty Action Lab, MIT express@expressindia.com