Falling poverty claims in SBI report spark concerns about methodology

‘Before jumping to a conclusion that India has been able to end poverty, we should interrogate the labour markets.’

IMAGE: A scavenger, surrounded by a flock of Greater Adjutant birds, collects plastic for recycling at a dump site in Guwahati. Photograph: Utpal Baruah/Reuters

State Bank of India (SBI), in a report on Friday, claimed a significant decline in the headline poverty ratio in rural areas on account of enhanced physical infrastructure, higher consumption growth in the bottom fractile and direct benefit transfers (DBTs).

Using the latest annual Household Consumption Expenditure Survey (HCES) data for August 2023-July 2024, the report claimed that poverty ratio in rural areas declined to 4.86 per cent in FY24 from 7.2 per cent in FY23.

Similarly, the poverty ratio in urban areas fell to 4.09 per cent from 4.6 per cent during the same period.

The SBI report estimated the poverty lines in FY24 at Rs 1,632 for rural areas and Rs 1,944 for urban areas by adjusting for decadal inflation and the imputation factor in the poverty line set by the Suresh Tendulkar Committee in 2011.

However, economists raise doubts about the veracity of these claims.

R Ramakumar, professor, Tata Institute of Social Sciences, said, the SBI report uses a flawed poverty line as the Tendulkar poverty line is not a poverty line, but just a “destitutional line.”

And, the previous government had to appoint the C Rangarajan committee to address criticism around it.

“It uses a method for updating the poverty line that does not account for changes in the consumption basket of households.

“Due to this, they end up using a very low poverty line.

“The poverty lines used for 2023-24 in the SBI report amount to just Rs 55 per day in rural areas and Rs 65 per day for urban areas.

“Unsurprisingly, they have underestimated the poverty line to obtain a lower estimate of poverty,” he added.

Echoing similar views, Pronob Sen, former chief statistician of India, said that poverty lines are usually estimated by basing them on certain developmental norms, which keeps on improving as time progresses.

He said to still use the norms prescribed more than a decade ago does not exude confidence in the fairness of the exercise.

“Poverty has no doubt reduced, but the question is the extent to which it has happened.

“Simply adjusting the Tendulkar line for inflation is completely wrong on account of the changes in consumption basket and the difference in the methodology of earlier consumption surveys and current HCES.

“At least use the same principles enunciated by professor Tendulkar and calculate the poverty line afresh,” he added.

Methodologically, the current HCES uses a modified mixed reference period (MMRP) in which the recall period for expenditure on items like edible oil, egg, meat, milk, vegetables and fruits is ‘last 7 days’.

For items like clothing, bedding, footwear, education, durable goods and medical, it is ‘last 365 days’.

For all other items, the recall period is ‘last 30 days’.

Earlier consumption expenditure surveys used the mixed reference period (MRP) method in which only two categories of recall periods existed.

Besides, the sampling methodology has also undergone a significant change in the current HCES.

For example, second stage stratum classification is now made on the basis of land for rural areas and car ownership for urban areas.

This is unlike previous exercises where the divisions were made based on household status and household activity for rural areas and monthly per-capita income for urban areas.

Santosh Mehrotra, visiting professor at University of Bath, UK, said such a sampling frame gives disproportionate representation to affluent sections of the society.

This results in recording of higher consumption expenditure, which then distorts the picture.

“How can poverty have fallen when job growth and real wages came down with the inflation eating away all the resources of the poor.

“Nearly 190 million workers (2021-22) in India are earning just up to Rs 100 per day (in real terms at 2010 prices) which can be categorised as “absolutely poor.”

This compares to just 106.1 million workers in 2011-12.

“Before jumping to a conclusion that India has been able to end poverty, we should interrogate the labour markets,” he added.