October 11, 2015

Once More on Poverty Estimates

Prabhat Patnaik

A REPORT in The Hindu (October 6) quotes a World Bank paper to the effect that the head-count poverty ratio has declined for the world as a whole from 14.2 percent of the global population in 2011 to 12.8 percent in 2012, which presumably also implies a decline for India. (In fact the report in The Hindu is headlined “World Bank Estimates Show Fall in India’s Poverty Rate”).

How the World Bank came to this conclusion, at least in the case of India, is a mystery, since the National Sample Surveys are undertaken for a large sample in India only once in a quinquennium; for the other years we only have surveys of small samples which are not very reliable and hence not much used. The last large sample survey was undertaken in 2011-12; so the claim of a decline in poverty ratio between 2011 and 2012 becomes particularly intriguing. And if the conclusion is intriguing for India, then it is equally intriguing for the world as a whole since the bulk of the world’s poor, even according to the World Bank, are to be found in India. But let us leave this issue aside and see what actually happened to poverty in India in 2011-12  compared to the earlier years

The reason why a large sample survey was carried out in 2011-12 is itself interesting. Since a large sample survey had already been undertaken in 2009-10, the next one in the normal course of things should have been in 2014-15. But, so dismal was the picture of hunger painted by the 2009-10 survey that the government of the time, despite constantly claiming in public a steady decline in the poverty ratio in the country, decided to order a special extraordinary quinquennial survey in 2011-12. It hoped that since 2009-10 had been a drought year and had therefore shown an “abnormal” degree of food deprivation for the people at large, if a “good” crop year like 2011-12 were surveyed then a “better-looking” picture would emerge. And not surprisingly it did; for the extent of nutritional deprivation does of course differ between drought and good harvest years.

But two crucial points must be noted here. The first is that when the successive quinquennial surveys, until and including 2009-10, were showing growing nutritional deprivation among the people, government spokesmen generally pooh-poohed that particular conclusion. They claimed that people were voluntarily cutting down their food intake, and hence their calorie intake, and choosing instead to spend on other things like education and health; and such a change in their spending pattern was interpreted as a symptom of their growing affluence rather than growing poverty. Growing nutritional deprivation in other words was taken to be the voluntary decision of a people actually becoming more affluent.




This of course was an absurd claim. Its absurdity is obvious from the fact that all over the world, as real incomes rise, the per capita foodgrain consumption, taking both direct and indirect consumption together (the latter in the form of processed food and animal products into which food grains enter as feed grains) increases, as does the per capita calorie intake. In India on the other hand not only was the opposite happening, but it was happening at an alarming rate. The per capita daily calorie intake for rural India which was 2156 in 1993-94, had dropped to 2038 in 2004-05 and 2018 in 2009-10 (the 1999-2000 figures are excluded because they are not comparable with these figures because of a different methodology used in that year). Likewise for urban India the per capita daily calorie intake had declined from 2072 in 1993-94 to 2007 in 1994-95 and 1981 in 2009-10.

Poverty in India is officially defined on the basis of daily calorie intake. Those who do not have access to 2100 calories per person per day in urban India and 2200 calories per person per day in rural India are considered “poor”; and the “poverty line” is defined as that level of per capita total expenditure at which people access exactly these levels of calories. If we take rural India then the percentage of persons falling below 2200 calories per person per day had increased from 58.5 in 1993-94 to 69.5 in 2004-05 to 76 in 2009-10; and in urban India the percentage of persons accessing less than 2100 calories per person per day had increased from 57 in 1993-94 to 64.5 in 2004-05 to 73 in 2009-10.

This obvious evidence for growing poverty, by the government’s own criterion, was deliberately misinterpreted to suggest that people were actually becoming better off and therefore consuming less calories! But when the 2011-12 data showed that compared to 2009-10 the calorie intake had increased, the same government agencies claimed that this finding vindicated their position and that poverty had actually declined. But if the earlier decline in calorie intake had been voluntary as the government had claimed, and indicative of people becoming better off, then the increase between 2009-10 and 2011-12 should have meant the opposite, by the same logic; ie, that people had become worse off during this period, that the increase in foodgrain consumption had been involuntary. By a process of absurd reasoning, however, the same government which had claimed the earlier decline as voluntary, now claimed that the 2011-12 increase too was voluntary.

The truth was that the very increase between 2009-10 and 2011-12 showed that people who had been getting squeezed earlier and hence reducing their food intake, increased this intake the moment the squeeze got a little less tight. Why it got a little less tight needs of course to be investigated: the year 2011-12 being a good crop year obviously had something to do with it, as had the existence of programmes like the MGNREGS. But the point is that the increase in 2011-12 vindicated the position that calorie intake reflects quite accurately, and can serve as a proxy for, the real incomes of the people.

And here I come to the second point. Even in 2011-12 the per capita daily calorie intake, both in rural and urban India, though higher than in 2009-10 and even 2004-05, was still considerably below what it had been in the year 1993-94. For rural India the per capita daily calorie intake had been 2156 in 1993-94 and had fallen to 2090 in 2011-12; for urban India the corresponding figures were 2072 and 2049. And in view of the fact that the government’s own claims, as just suggested, vindicate taking calorie intake as a proxy for real income, it follows that in the period of liberalisation which roughly coincides with the years between 1993-94 and 2011-12, there has been a worsening of the real incomes of the majority of the Indian people, whether in urban or in rural areas.

To say that the real incomes of the majority of the Indian people have declined over this period is not synonymous with saying that they are worse off today than they were in the beginning of the nineties; and this is so for one obvious reason. While the commodity composition of their consumption has obviously changed dramatically over this period, one aspect of this change has to do with the introduction of commodities which are beneficial for life to an extent far greater than is measured by their price, such as the introduction of new drugs and new medical practices unknown earlier. This may improve the quality of living, even when real incomes, measured by their proxy, the calorie intake, are declining.

But liberalisation per se does not contribute to the introduction of such new innovations, especially in the medical sphere; these would have come into the country even if the public sector had a monopoly over medical facilities. Hence the contribution of liberalisation per se has been a decline in food grain intake which in turn reflects a decline in real income for the bulk of the people.

Among the many reasons for this decline, viz, the process of “primitive accumulation of capital” that dispossesses petty producers without absorbing them into what Marx had called the “active army of labour”; the process of squeezing the real incomes of petty producers, even when they are not dispossessed outright, through the curtailment of subsidies and the offer of un-remunerative prices; one in particular stands out. And this has to do with the privatisation of a range of essential services, like education and health, which increases the prices paid for these services even when there is no change in service quality.




Such an increase in the price paid does not get counted statistically as a rise in the cost of living. For instance, if in the initial period there were only government hospitals, offering, let us say, a particular surgery for Rs 100, while, now, government hospitals, though still charging Rs 100 for the same surgery, are few and far between, which forces people to go to private facilities that charge a lot more for it, the price index would still show a zero increase in the cost of this surgery. Privatisation in short imposes a squeeze on the people which is not reflected in the price index. The people who are so squeezed, cut back on their food intake to pay for their hospital bills that get correspondingly inflated.

What has really happened in this case is a squeeze on the people through a de facto price increase, effected through privatisation, which has forced them to cut down their food intake; ie, a clear decline in their real income. But what appears to have happened, because of the assumption of no increase in the cost of living, is a mere gratuitous shift in consumption expenditure from food to medical facilities; and this is then interpreted as a voluntary shift.

All poverty estimates that ignore the palpable fact of a decline in foodgrain intake per capita, taking both direct and indirect consumption into account, are fundamentally wrong. The World Bank’s reported estimates cannot be otherwise.