“THE issue of poverty estimates has been the subject matter of public debate in and outside parliament. Government respects and is fully conscious of the sensitivities of the people on this issue…The government has also taken a decision to set up a technical group to revise/revisit the methodology for estimating poverty in a manner which is consistent with the current realities.” This extract is from the government’s press note in March 2012 on the appointment of C Rangarajan as the chairman of the technical group. The Rangarajan group was appointed to contain a massive public outrage in 2011. The reason: a Planning Commission affidavit to the Supreme Court had stated that the poverty line for rural areas was Rs 29.3 per capita per day and urban areas was Rs 32.5 per capita per day. If such an affidavit was submitted in the 1980s or the early-1990s, the outrage may not have been comparable. Till then, the interest in poverty and poverty lines was largely academic; estimates of poverty were primarily used to understand whether benefits of five year plans did indeed trickle down to the masses. However, this was not the case in the 2000s. By then, economic reforms had led to major shifts in social policy. The most significant was the shift from universal provision to targeted provision in a range of social services, especially food, health and housing. The eligibility for these service provisions were linked to the poverty rates obtained using the official poverty lines. These poverty rates were just estimates from a sample, and identification had to follow estimation. For identification, another faulty survey was used combined with a rather bizarre scoring method. As a result, there was little correlation between the levels of living of a household and whether the household was eligible to receive benefits. Public anger over the massive exclusions from targeted schemes was growing when the affidavit was submitted. The poverty line, from then on, was no more an academic matter. It had begun to directly touch the everyday lives of common people. Media covered the controversy widely. The report on Tushar Vashisht and Matthew Cherian, both MIT graduates, who travelled to India and attempted to live on Rs 32 a day, became a hit. Neo-liberal economists, who till then prided on the self-proclaimed statistical prowess of poverty statistics, ate the humble pie and tried to explain. Montek Singh Ahluwalia said that the official poverty line was “not comfortable but…not all that ridiculous in Indian conditions”. According to him, the poverty line, by definition, “implied considerable stress”. He was adding fuel to the fire. The result: the then official poverty line became irrevocably and permanently discredited in the public sphere. What were the earlier methods of estimating poverty? A task force of the Planning Commission in 1979 defined the poverty line as that per capita expenditure at which the average per capita per day calorie intake was 2400 calories in rural areas and 2100 calories in urban areas. Thus, for 1973-74, the poverty lines arrived at were Rs 49.09 per capita per month in rural areas and Rs 56.64 per capita per month in urban areas. These poverty lines were not freshly estimated for every future year; instead, the poverty lines for 1973-74 were simply updated by accounting for changes in consumer price indices. INCREASING DISCONNECT Such a method turned out faulty. It did not, for instance, account for temporal changes in the consumption baskets of people. As a result, simple updating of the old poverty line led to an increasing disconnect between the poverty line and the normative calorie consumption expected at that level. As the Tendulkar Committee report itself revealed, individuals at the all-India urban poverty line, who were expected to be consuming 2100 calories per day, were actually consuming only 1776 calories per day. Left economists like Utsa Patnaik, who pointed to such errors, were curiously derided as “calorie fundamentalists”. The Tendulkar Committee report, submitted in 2010, substituted the old faulty method with a new faulty method. The new method was the following: we take the all-India urban poverty line as the basis for estimating every other poverty line. This all-India urban poverty line was to be estimated (a) by using the same old calorie norm of 2100 calories per day; (b) by updating Rs 56.64 (i.e., the per capita monthly expenditure at 1973-74 prices, adequate to buy 2100 calories per day) for inflation; and (c) with no regard for temporal changes in the consumption basket after 1973-74. From this all-India urban poverty line, using a parity index, we estimate what it takes for people of different states to afford the same level of consumption. While Tendulkar’s new method was faulty on many grounds, two issues need stress. First, the poverty line was hardly raised by the new method. In per capita daily terms, for 2004-05, the poverty line was raised from Rs 12 to Rs 15 in rural areas and from Rs 18 to Rs 19 in urban areas. Secondly, the new poverty line represented a huge fall of the normative calorie requirement from 2100 calories to 1776 calories in urban areas, which was carried over to all the other poverty lines estimated from it. As Madhura Swaminathan pointed out, Tendulkar had wrongly used a report by the Food and Agriculture Organisation (FAO) while downwardly revising the normative calorie requirement. The poverty estimates that became controversial after 2011 were based on this Tendulkar report. What does the new Rangarajan report have to offer? First, the Rangarajan report has rejected the arbitrary method that Tendulkar report suggested, which was to take the all-India urban poverty line as a benchmark to determine all other poverty lines in the states. The report recommends that the government revert to the method of using separate rural and urban poverty basket lines. Secondly, the Rangarajan report also rejects the recommendation of the Tendulkar report to delink poverty lines from calorie norms, though Tendulkar himself had ended up using it for estimating the all-India urban poverty line. The report adds proteins and fats to the nutrient-norm, in addition to calories. The new protein norm is 48 gm (rural) and 50 gm (urban) per capita per day. The new fat norm is 28 gm (rural) and 26 gm (urban) per capita per day. Thirdly, however, the Rangarajan report has reduced the calorie requirement in rural areas to 2155 calories (from 2400 calories) and in urban areas to 2090 calories (from 2100 calories). When the normative calorie norms were fixed in 1979, it was assumed that a worker in rural India required at least 300 calories per day more than his/her urban counterpart. This was due to the lower levels of mechanisation in rural areas, as well as the relatively poor facilities of transport and infrastructure. In the Rangarajan report, there is little difference between normative calorie norms in the rural and urban areas. If the difference between the rural and urban calorie norms was 300 calories earlier, it is now just 65 calories. While the calorie requirement in urban areas has fallen by just 10 calories (from 2100 to 2090), the calorie requirement in rural areas has fallen by 245 calories (from 2400 to 2155). Has infrastructural facilities in the urban areas not changed at all? Conversely, is it that there are now no differences between the rural and urban areas with respect to transport and other forms of infrastructure that influence energy requirements? Fourthly, in addition to nutrients, the Rangarajan report claims to have added two more items of expenditure in estimating the poverty line: (a) clothing expenses, rent, conveyance and education expenses; and (b) all non-food expenses that meet nutrition requirements. Thus, for 2011-12, the Committee estimates the following: (a) a per capita monthly expenditure of Rs 554 in rural areas and Rs 656 in urban areas to meet calorie, protein and fat requirements; (b) a per capita monthly expenditure of Rs 141 in rural areas and Rs 407 in urban areas for clothing expenses, rent, conveyance and education expenses; and (c) a per capita monthly expenditure of Rs 277 in rural areas and Rs 344 in urban areas for non-food items that meet nutrition requirements. Together, they add up to the new poverty line: a per capita per month expenditure of Rs 972 in rural areas and Rs 1,407 in urban areas. In other words, Rs 32 per capita per day in rural areas and Rs 47 per capita per month in urban areas. PROBLEMATIC CONSTRUCTIONS It is in the construction of these expenditure requirements that the problem lies. For instance, both Tendulkar and Rangarajan use the expenditure of the median fractile in the NSS sample to estimate the expenditure on clothing expenses, rent, and conveyance and education requirements. In both reports, the observed median is considered as the normative requirement! How much is this median expenditure considered as adequate? Just Rs 141 per capita per month in rural areas and Rs 407 per capita per month in urban areas. Or, Rs 705 per month per five-member family in rural areas and Rs 2035 per month for a five-member family in urban areas. Or, Rs 23 per day per five-member family in rural areas and Rs 68 per day for a five-member family in urban areas. Or, Rs 5 per capita per day in rural areas and Rs 14 per capita per day in urban areas. For clothing, rent, conveyance and education, all together! Despite setting the poverty line at destitution levels, the report shows that 30.9 per cent of the rural population and 26.4 per cent of the urban population were poor in 2011-12. In absolute terms, there were 26.1 crore poor persons in rural India and 10.3 crore poor persons in urban areas. In all, India had 36.3 crore poor persons in 2011-12. If the use of a destitution line can show that 36.3 crore persons were poor, what would be the real extent and depth of poverty in India? The outrage that the 2011 affidavit triggered has in no way been ameliorated by the Rangarajan Committee. That debate raised important questions: is the analysis of poverty a statistical issue or a livelihood issue? Is not adequacy an important concern? Or, should we be content with the view that the poverty line should necessarily reflect “considerable stress”? Three years down, and yet another report in hand, these questions have remained.