The poverty line figures given by the Planning Commission for 2009-10 have drawn strong criticism of foul play from politicians, social activists and some economists. Is the current poverty line justified? Why are there so many conflicting opinions? To understand this complex issue in simple terms, Indivjal Dhasmana interviewed Pronab Sen, principal advisor to the Planning Commission and former chief statistician. Edited excerpts:
There is so much confusion around the poverty line. Critics say the cut-offs of Rs 28 per person a day in urban areas and Rs 22 in rural areas are too low even in the Indian context. Is this valid?
It is very difficult to say what is high and what is low. If you are talking about a poverty line that will identify people who you can target, then you should not have the line so high that targeting becomes impossible. So far as this line for 2009-10 is concerned, 28 per cent of the people are living below that number. Now, 28 per cent in Indian context is large: it is about 320 million people. Does the government help these people, or those earning more? That is the question. Today, do I have the ability to handle even 320 million people? If I have an ability to handle 500 million people I will help 500 million. The question is how much can you handle?
But the Planning Commission’s position is that the government’s social welfare schemes are not linked to the poverty line.
You see, social welfare schemes are ultimately the responsibility of state governments. It is entirely their responsibility to determine who they wish to classify as the poor and who they don’t. It is not a new issue; this issue has been raised earlier too. In fact, the single largest social welfare scheme was the Rs 2 a kg rice scheme announced by N T Rama Rao. That scheme continued after the Rama Rao government. In Andhra Pradesh today, nearly 90 per cent of the people have a BPL [below the poverty line] card. However, the central government does not give Andhra Pradesh subsidy for 90 per cent of the people. We give them a subsidy only to the extent of our estimate of their poverty numbers. The state government takes care of the difference. Where the money is coming from is a different issue.
Does that mean that for the Food Security Bill, the Centre will provide money only for 320 million people and the state governments will contribute the rest?
No, the Food Security Bill is a somewhat different kind of Bill. It is an overriding Bill. Up to now, the states were free to give BPL cards to as many people as they wanted. But the targeted PDS [public distribution system] subsidy from the Centre was given to states on the basis of their share in the poverty estimates.
After the Tendulkar committee’s method was virtually discarded, the Planning Commission did not rule out the possibility of linking social welfare schemes to the poverty line.
It had to do so; whether to link it to the poverty line or some other method of identifying the poor, it has to be linked to something. Because you have to issue some kind of cards to the beneficiaries. If you don't wish that then there is no problem, it would become a universal issue. Then anybody as a citizen of India has the right to avail of the schemes. But if you are going to target, that target has to be based on something. So, now you have socio-economic caste census, which is underway. In this, no one is being asked about income or consumption expenditure, but about conditions of living. Even there, you have to determine who is to qualify as poor and who won’t. At the end of the day, you are doing a classification. It is easy for the people to say that Rs 28 a person a day is too low, but how are you going to react if I say on a scale of one to ten, 4.5 is the cut-off. I mean, it is very difficult to have an emotional debate around the index.
Many people have criticised the way the Planning Commission calculated the 2009-10 figures by including mid-day meals in household expenditure that year, an item that was not included in 2004-05. How will you react to that?
I don’t understand. The Planning Commission does not do the consumption survey. The Planning Commission uses NSSO [National Sample Survey Organisation] data. In NSSO data, historically, there has always been a category of free meals. Earlier, free meals were not evaluated because it was believed that free meals would be reflected in some other household expenditure. So you would be doing double counting, because you would be counting the household that is giving the free meal and the household that is receiving the free meal. This time, apparently, the mid-day meal scheme was evaluated, since it was coming from the government. I would emphasise this was not done by the Planning Commission, but by NSSO.
But is the methodology correct?
Yes, it is correct. Frankly, what it does is to introduce non-comparable data, perhaps. But it is theoretically correct, because if I am getting food, whether I am paying for it or not, I am eating food. It is a part of my consumption. Let me give you an example. If I have a PDS system through which I give you food free, you will say don't count this, then my single largest poverty alleviation scheme will simply not get counted.
No, but since the variable was introduced in between, would it not make the poverty line difficult to compare between the two years?
The point is the line is still the same. We did not have a mid-day meal scheme in 2004-05, by the way. The mid-day meal became universalised after that. It was not reflected in household consumption. The poverty line has nothing to do with it, it is calculated separately. The consumption expenditure survey gives you actual consumption of the people. So, Tendulkar has given a basket and a method by which the basket is valued. That basket has not been changed. It is the same, both for 2004-05 and 2009-10. It is only pricing that has changed to take care of inflation. What has changed is what is included in the consumption of households, which is what NSSO calculates.
But, it will give different picture of those below poverty line. Isn't it?
No, it won't, that is the point. If I have a scheme that exists now and did not exist then, can you say I should drop that scheme? If I am giving you a pension today, which you were not getting yesterday, do I say that in order to determine your economic situation today, I should deduct that pension? That would be wrong. Now, whether that mid-day meal scheme was properly evaluated or not is a different matter.
The government has virtually abandoned Tendulkar Committee recommendations, which the Planning Commission was earlier defending. Whose position is correct?
I can't comment on that.
NSSO household consumption expenditure was almost half of what was derived by the National Accounts for 2009-10. How do you explain that?
Nobody quite knows. Quite a number of committees were set up to go into the issue. It is not a new issue and is not unique to India. Almost every country has this difference, but our difference is larger. In most countries, it is 30 to 35 per cent. Ours is 50 per cent. One of the reasons could be that the list of expenditure in the NSSO survey has 560 items against which households are asked how much they spend. This list was last drawn up in the 1970s. Between the 1970s and now, the number of commodities that households consume has gone up manifold, which are not included in that list. If you ask on the basis of that list, it is likely that person would not give you response. National accounts capture it from production side, so anything that is produced is captured. In that sense, national accounts are capturing more new commodities and services that have come into the market.
But national accounts also measure from the expenditure side?
It does, but here, private consumption expenditure is a residual. What you get is total national income. From that, you deduct government expenditure and corporate expenditure and what is left is private expenditure.
Does that mean that NSSO survey hides inequalities in consumption expenditure?
Yes, it does. Consumption inequality that you get from NSSO data is probably understated. Income inequalities will be higher still.
How would you respond to a story of two Indians coming to India from abroad and living on Rs 32 a day which was a poverty line given by the Planning Commission to the Supreme Court for June, 2011. They found it virtually difficult to live.
Rs 32 per person per day translates into Rs 4,500 per household a month. If you come with me, I will show you any number of families living on less than Rs 4,500 a month. In fact, everyone reading this interview should ask themselves a question: how much do they pay their servants?They will get their answer.
But is that assessment of poverty by the two Indians from abroad correct?
No, it is wrong.
It’s simple, they were projecting a consumption pattern of their own. They wanted to buy a banana. The poor in India don’t eat bananas, they can’t afford to.
In a release, the Planning Commission had said any poverty line would essentially be an arbitrary line. What does it mean?
You know, you can set the poverty line anywhere you want. This does not make any difference, because a poverty line is used to measure what is happening to the people below the poverty line. It is an analytical device to track. Whether I set it at Rs 10 or Rs 30 or Rs 50 does not make any difference at all, so long as I hold it constant at current prices. If I have said it was Rs 10 in 1990, I have to apply an inflation factor to calculate it for today. What I cannot do is to say, for example, that Rs 10 is Rs 30 today on an inflation-adjusted basis, but I will set the poverty line at Rs 50. In that case, where I set it makes no difference.