Should purchasing power parity of $3.2 be made India's formal poverty line?

Experts have some reservations about the methodology used to arrive at the number, and that the poverty figures under the existing PPP of $1.9 per person per day have been understated

A recent International Monetary Fund (IMF) working paper suggests that $3.2 on purchasing power parity (PPP) basis per person per day could be made an official poverty line since India has almost eliminated extreme poverty at $1.9 PPP.

"The low level of extreme poverty–-around 0.8 per cent in both 2019 (0.76 per cent) and 2020 (0.86 per cent)–-is suggestive of the need for the official poverty line to now be PPP $3.2," said the paper, co-authored by Surjit Bhalla, executive director for India at the Fund, Arvind Virmani, former chief economic advisor and Karan Bhasin, a policy researcher.

The paper 'Pandemic, Poverty, and Inequality: Evidence from India' uses two methods, both of which lean on the consumer expenditure survey by the National Statistics Office (NSO) in their calculations. However, the problem is that the last such survey came in 2011-12, and the latest one in 2017-18 was junked by the government.

The suggestion assumes significance since India does not have any official poverty line. The last such line was calculated for 2011-12.  

IMF clarified that the paper represents the views of authors and not of the organisation.

The authors used the real private final consumption expenditure (PFCE) growth on the consumer expenditure survey to arrive at poverty estimates. The paper also used real state gross domestic product (SGDP) growth rate to arrive at the poverty figures. Virmani said the SGDP method is more appropriate.

On the basis of the PFCE method, the paper showed that poverty at $1.9 PPP declined to 0.86 per cent in 2020-21 from  10.8 per cent in 2011-12, after factoring in food transfers by the government. In fact, poverty had declined to 0.76 per cent during 2019-20, but rose a bit in Covid-affected FY20.

On the SGDP methodology, the paper showed that poverty at $1.9 PPP declined to 1.42 per cent during 2020-21 from 10.8 per cent in 2011-12. It had, in fact, declined to 1.3 per cent in 2019-20.

When asked whether the paper's suggestion was appropriate, former chief statistician Pronab Sen said advice is contingent on the other statement by the study that at $1.9 PPP, poverty does not exist in India.

"Now, I contest that argument. I think it is wrong. They have done the usual Surjit (one of the authors) game, which is they have taken distribution from the NSS and applied it to the PFCE numbers from the national accounts. Essentially, what they have done is that they have multiplied all consumption by about 60 per cent. I seriously object to that particular procedure. Everyone other than Surjit objects to that procedure," Sen said.

Sixty per cent in the study is the difference between the total NSS consumption and the national accounts estimates of consumption.

Sen wondered whether at $1.9 PPP, poverty is actually 0.8 per cent. "The people who work on it, other than Surjit, say that poverty is close to 10-12 per cent when you are at $1.9 PPP. If that is the case, there is no reason to shift to $3.2 PPP," Sen said.

For instance, a recent World Bank paper, authored by economists Sutirtha Sinha Roy and Roy van der Weide, showed that poverty at $1.9 PPP stood at 10.2 per cent in 2019.  

When reminded that the paper used two methods, the other being on the basis of real SGDP growth rates, Sen said it essentially comes to the same thing.

Pointing out that there are plenty of issues with this, he said price differences among states in India are huge, at 30-40 per cent. "When we do poverty calculations in India, we calculate the poverty line for each state separately," he said.

When this query was posed to Virmani, he said the $1.9/day poverty line is approximately equal to the Tendulkar poverty line for the rural poverty line of the base state.

"Our paper uses all the urban and rural poverty lines for every state determined by the Tendulkar Committee--the same poverty lines, separately for urban and rural areas of each state," he said.

He further explained that for $3.2, the corresponding poverty lines for the urban and rural areas of each state is arrived at by multiplying each by $3.2/$1.9 = 1.684.

"We use these detailed, state-wise (urban, rural) poverty lines in deriving our results, not just the base lines of $1.9 and $3.2," Virmani said.

However, Sen said India does not have a national poverty line, and that it is derived from the state levels. The IMF paper did it the other way round--deducing the state level poverty lines from the national one.

India stopped coming out with a poverty line ever since the Narendra Modi government came to power at the Centre.

Earlier, the erstwhile Planning Commission used to measure the number of the poor on the basis of the poverty line. According to one such report, 21.9 per cent of the population was poor in India in 2011-12 against 29.8 per cent in 2009-10. That report was based on the Suresh Tendulkar methodology. Under that, those who spent more than Rs 33 a day in urban areas and Rs 27 in rural areas were not regarded as poor. This had triggered a controversy.

Ultimately, the C Rangarajan-led panel came up with another report that said the number of the poor in India was much higher in 2011-12 at 29.5 per cent.

The report took a person spending less than Rs 47 a day in cities and one spending less than Rs 32 a day in villages as poor. This poverty line approach was abandoned by the NITI Aayog, which replaced the Planning Commission on January 1, 2015.

Table: Indian poverty line over the years

Year	Rural	Urban
2004-05	14.89	19.29
2009-10	22.43	28.65
2011-12	27.20	33.33
2011-12*	32.00	47.00

* Rangarajan panel; Figures are rupees spent per person; Source: Derived from PRS, erstwhile Planning Commission