BS Number Wise: Here's why India's poverty estimates are problematic

Traditional estimates of poverty in the country rely on indicators that are outdated or inaccurate

Separate reports say India’s poverty reduced by 50 million or 140 million in the last five years, strengthening the government’s claims about its policies. The feel-good numbers don’t hide this though: the way India counts the poor is problematic.

The World Bank’s measurement shows that extreme poverty — people living on less than $2.15 daily in terms of purchasing power parity (PPP) — declined by 57 million between 2015-16 and 2019-21. On the other hand, United Nations Development Programme (UNDP) data indicates that multidimensional poverty—a holistic indicator that considers parameters like sanitation, health, and education—declined by 140 million (see chart 1).. The number of multi-dimensionally poor declined more than the number of poor, vindicating the government's health, sanitation and education schemes.

Such reports also vindicate part of an argument advanced by economists Surjit Bhalla, Karan Bhasin and Arvind Virmani in an International Monetary Fund (IMF) paper. Their analysis showed that between 2015 and 2020, the number of people in extreme poverty (living on less than $1.9 daily in PPP terms) declined by 79 million if government transfers were not taken into consideration and by 84 million if one incorporated the effect of welfare measures.

Estimates of how many people fall below the poverty line vary. As per the UNDP report, the number of multi-dimensionally poor declined 10.7 percentage points from 27.1 per cent in 2015-16 to 16.4 per cent in 2019-21. Bhalla estimated that the number without transfers declined by 5.9 percentage points and 6.3 percentage points if transfers were considered. But the World Bank report puts the decline at 4.7 percentage points  (see chart 2).

The various reports do not take away from India’s lack of a cogent and official measure of poverty. Any academic calculation fails to reflect the true nature of the problem, given the outdated estimates or gaps in new measures like the CMIE’s Consumer Pyramids Household Survey used by the World Bank in its latest report. Poverty estimates rely on consumer expenditure surveys from 2011-12, but consumption patterns have changed drastically. So, the older estimates, like the Tendulkar poverty line or even the World Bank’s $2.15 PPP dollar measure, could be outdated (see chart 3).

The government needs to update its poverty measure for regular and accurate assessment. Indeed, the multidimensional poverty reports by Niti Aayog can be a barometer. Income- or consumption-based poverty lines are necessary too.

A reliable estimate of the poverty line would help the government decide how much support it needs to extend to families.