Just as there are many ways to define poverty, there is now an a la carte menu of poverty estimates to choose from. Arjun Sengupta, the economist and member of the Rajya Sabha, declared a few years ago that an overwhelming 836 million people in India lived on a per capita consumption of less than Rs 20 a day (or Rs 600 in a month). This startling figure, which meant that 77 per cent of the country was below the poverty line, has been much loved by the Left after it appeared in a report on the Conditions of Work and Promotion of Livelihood in the Unorganised Sector.
Now, a new set of numbers on the poor has been worked out by NC Saxena, a retired civil servant acting as a commissioner appointed by the Supreme Court. He has reportedly said that half the country’s population of 1.15 billion is below the poverty line, which he apparently defines as a monthly per capita income of Rs 700 in rural areas and Rs 1,000 in urban areas. In other words, using a higher income cut-off, Mr Saxena comes up with a much lower poverty figure than Dr Sengupta.
But even Mr Saxena’s number could be far too pessimistic, and therefore the real poverty figure even lower. His assertion that half the people live on a rural-urban weighted average monthly income of Rs 775 or less, would mean that their combined expenditure would total about 15 per cent of all private consumption (which is two-thirds of GDP). This cannot be true, since other figures tell us that the lowest 20 per cent of the population accounts for 8.9 per cent of consumption. The lowest 50 per cent would then account for at least two-and-a-half times as much, ie 22.25 per cent (and not Mr Saxena’s 15 per cent).
The World Bank has a third set of figures, based on an international poverty line of $1.25 per day of income (adjusted for each country’s purchasing power parity, which for India would make it slightly more than Dr Sengupta’s Rs 20 per day); the World Bank’s figure for those below this line in India is 42 per cent, based on 2005 data. The government’s poverty number, based on a different definition, is 28.6 per cent of the population.
When different numbers float around, people will choose to go by the number that fits their theory, personal predilection or political position. In some senses, this is a game of definitions. It could reasonably be argued, for instance, that an official poverty line worked out four decades ago, on the basis of what a basic diet would cost, needs to be updated to include some other essentials, and therefore that the official assessment is no longer very useful. On the other hand, it could be argued that an absolute poverty line serves usefully as a long-term marker and hence should not be discarded even if poverty itself gets redefined to include (say) lack of access to education or health services.
The immediate problem, though, is that the different numbers that have been put out do not sit well with each other, being mutually inconsistent. If one estimate is correct, then one or more of the other estimates have to be wrong. And it does not help that yet other data contradict all these assessments. The National Council of Applied Economic Research has conducted household surveys since the 1980s. Based on the last one, it has projected family incomes for 2009-10, according to which about half of all Indian families should now be living on more than Rs 7,500 per month (at 2001-02 prices). Adjusted for inflation, this works out to Rs 11,000 a month today. Given that the average Indian family has about five members, 50 per cent of Indians live on a per capita figure of Rs 2,000 per month or more, whereas Mr Saxena’s finding is that it is Rs 1,000 or less. If you are inclined to believe NCAER, then the actual prevalence of poverty is closest to the official poverty figure, and much less than the other three poverty estimates.