|How rich are India's poor?|
|Tuesday, 29 December 2009 00:00|
The 37.2 per cent poverty number put out by the expert group headed by Suresh Tendulkar last week sounds a lot less pessimistic than the 78 per cent figure of Indians living on less than Rs 20 per day put out by Arjun Sengupta. But the question is whether it is more reliable, and if so, by how much. Another report, by NC Saxena, put the figure at around 50 per cent some months ago. And at 42 per cent, the World Bank has yet another number!
All the estimates, it should be obvious, are based on what you take as the poverty line. The existing government method used an expenditure poverty line which Tendulkar said was too low as it didn’t take into account expenses on education and health, and so he sensibly added on a cost for this. He also felt the distinction between urban and rural consumption norms was outdated and decided to use the same urban consumption basket for rural areas, but obviously with different prices to reflect the difference in costs.
He came out with a poverty line expenditure of Rs 483.60 per capita per month — Rs 446.70 for rural areas and Rs 578.80 for urban areas — and this is what ensured his poverty estimates are 37.2 per cent as compared to the government’s 27.5 per cent. The World Bank’s poverty line expenditure is marginally higher than Tendulkar’s, and Saxena’s is higher than the Bank’s — unlike the others who used expenditure data, Saxena used an income of Rs 700 for rural areas and Rs 1,000 for urban areas for his poverty estimates. So, the higher the poverty line, the greater the number of poor in the country. It doesn’t quite require a rocket scientist to figure out that choosing of the correct poverty line is critical.
This, of course, is where Tendulkar’s numbers also come unstuck, just like the official 27.5 per cent poverty number he seeks to replace. As Surjit Bhalla has pointed out, the expenditure data provided by the National Sample Survey (NSS) captures less and less of India’s expenditure with each passing year. In 1960, according to Bhalla, the all-India expenditure estimate given by the NSS was around 87 per cent of the expenditure that the National Accounts or the GDP data told us was taking place in the country.
By 2004-05, this was down to just 48 per cent — in other words, the NSS data, which the government/World Bank/ Tendulkar used, reflects less than half the total consumption in the country! Given this, these exercises will always indicate that India has more poor people than is really the case — Bhalla estimates that, using the pre-Tendulkar expenditure poverty line, just 11 per cent of Indians are poor once a correction is made for this under-reporting. Even if you increase the expenditure that comprises the poverty line like Tendulkar has done, it’s unlikely the figure will rise to more than 15-16 per cent, a far cry from Tendulkar’s 37.2 per cent.
The National Council of Applied Economic Research’s (NCAER’s) annual household survey of income is another useful tool to measure the Tendulkar estimates by, and this also reinforces the view that Tendulkar’s poverty numbers are way too high. According to NCAER 2004-05, 30.5 per cent of Indians — 35 per cent in rural India and 19 per cent in urban India — are “poor”, based on Tendulkar’s expenditure poverty line.
While this is significantly lower than the 37.2 per cent figure put out by Tendulkar, even more important is what these people own. Around one in 13 households in this group of people in rural areas own a two-wheeler — the figure rises to one in five in urban areas. Which means these households earn enough to afford to buy petrol to use these vehicles. One in four owns a ceiling fan in rural areas and seven in 10 own the same in urban areas; one in 20 in rural India owns a colour TV and the figure is one in 4 in urban India (the figure for black and white TVs is many times this) — which means they live in houses that have electricity. In other words, this knocks a serious hole in the view that these are households below the poverty line.