The recent debate on the extent of poverty in India has been generating a lot of heat, but very little light. The debate was sparked off by the latest poverty estimates released by the Planning Commission. These show quite a dramatic fall in the level of poverty. The number of people below the poverty line — the threshold level of per capita expenditure below which a person is deemed poor — has declined from 37 per cent to 22 per cent in the seven years between 2004-05 and 2011-12. This would, undoubtedly, be the fastest rate of decline in poverty in India.
Members of the Opposition parties as well as others are up in arms, arguing that there is little basis for the claim that the incidence of poverty has gone down. In particular, their criticism has focused on the fact that the poverty line, set at Rs 33 per person per day in urban areas, is absurdly low. “Who can live on so little money?” they ask. Not surprisingly, Congress politicians have responded with some ridiculous statements. An official spokesman claimed that Rs 12 was enough for a hearty meal in Mumbai. This was trumped by another who said that even Rs 5 was enough for the same purpose in Delhi. The media had a rollicking time going round dhabas in Mumbai and Delhi, demonstrating the preposterous nature of these figures.
So, is the poverty line set at too low a figure? Should it be set higher? These questions themselves are somewhat ill-framed because there is no objective way in which a poverty line can be specified. If the reader tries to find the answer to “what is the poverty line?” on the internet, the typical answer will be that it is the level of income required for a socially minimally acceptable level of living. This is, of course, completely subjective since it is impossible to pin down what is a minimal acceptable level of living.
Sometimes, some objective basis is sought to be given to the specification of the poverty line by using a calorie norm. Nutrition experts recommend minimum calorie norms required for sustenance, these norms depending on age, sex and activity levels of individuals. The age, sex and occupation structure of rural and urban populations can then be used to determine the average calorie norm for rural and urban areas. For instance, a task force appointed by the Planning Commission in 1979 recommended the figures of 2,400 and 2,100 calories per day for rural and urban areas. National Sample Survey data on household consumption expenditure in 1973-74 were used to calculate the monetary equivalents of these calorie requirements. So, the data showed that on an average, those spending approximately Rs 49 per person per month in 1973-74 satisfied the calorie norm. The corresponding figure was roughly Rs 56.50 in urban areas. These then became the rural and urban poverty lines for 1973-74. These figures are then inflated by appropriate consumer price index numbers for urban and rural India to derive the poverty lines for later years.
Another committee with the late Suresh Tendulkar as chairman was appointed in 2009 to suggest modifications. The committee stated its desire to move away from a calorie-based norm. However, it kept the 2004-05 urban poverty line unchanged, although this was based on the earlier calorie norm, but increased the rural poverty line. This has come to be called the Tendulkar poverty line and forms the basis for the latest poverty estimates of the Planning Commission. This line was Rs 16 per person per day in 2004-05. This was comparable to the World Bank specification of one-dollar-a day benchmark when differences in purchasing power across countries were taken into account.
The calculation of the number of people spending less than the poverty line is purely mechanical. So there is no question that there has been a sharp decrease in poverty between 2004-05 and 2011 if the Tendulkar poverty line is used. Critics who feel that the poverty line should be much higher in 2011 (and hence the level of poverty higher too) must, of course, concede that consistency requires that the poverty line should also have been set higher in 2004-05. In other words, the only way to measure inter-temporal trends in the incidence of poverty is to use a common poverty line in real terms. Of course, the threshold level in nominal terms will change because of inflation. But, if a higher poverty line was used for 2004-05, then the base figure of poverty would also have been higher, and so we would have almost certainly observed a similar rate of decrease in poverty during the seven years.
So, the entire debate about the “correct” level of the poverty line has been quite meaningless. In fact, this is particularly true if there is an attempt to base the poverty line on a calorie norm. Naturally, consumption baskets change when relative prices change over time. Hence, even when a household has an income level which is sufficient to allow it to consume the required calories, it may choose not to do so. Moreover, even poor households spend small fractions of their total expenditures on tobacco, alcohol and perhaps even on an occasional egg. This expenditure may not give them the required level of calories but it surely gives them satisfaction.
Where do we then go from here? Since any specification of a poverty line is essentially arbitrary, a better picture of changes in the living conditions of the poor must move away from counting the number of persons with expenditures below any threshold. Instead, it is better to focus on changes in mean expenditures of the bottom deciles. Have the mean expenditures of the bottom 20 per cent, 30 per cent and 40 per cent of the population in terms of per capita expenditure increased or decreased between 2004-05 and 2011-12? If the living conditions of all these deciles improved, then surely the incidence of income poverty has gone down.
Given the large reduction in the incidence of poverty reported by the Planning Commission, it is very likely that all deciles of the population have recorded some increase in living standards. Of course, this is entirely consistent with an increase in inequality simply because the proportional increase in expenditure of the highest deciles may have been significantly greater than that of the bottom deciles. Perhaps the poor care only about improvements in their own absolute standard of living and are less concerned about relative inequality.