Himanshu: The World Bank’s revision of its poverty estimates is befuddling

For the record, the April estimate was based on the $2.15-per-day poverty line using the 2017 PPP index derived from the International Comparison Programme (ICP) 2017 cycle. For India, it suggested a decline in poverty from 16.2% in 2011-12 to 2.3% in 2022-23. 

The latest estimates use the updated 2021 ICP cycle for the PPP index with the extreme poverty line revised to $3 a day. Based on this new poverty line and PPP exchange rates, poverty in India was found to have declined from 27.1% in 2011-12 to 5.3% in 2022-23. 

Also Read: The poverty line has moved but have basic vulnerabilities in India eased?

However, in India’s case, there are several issues. First, take the contentious issue of comparability of the 2022-23 Household Consumption Expenditure Survey (HCES) data with earlier rounds. The methodology note acknowledges this, but then goes ahead and uses it. While the absence of comparable consumption data is a major issue in the case of India, the World Bank’s decision to use Indian data for global poverty estimation could raise questions on the credibility of its estimate for poverty across the world.

Similarly, its use of ICP 2021 also raises doubts over PPP estimates. The PPP exchange rate for household final consumption expenditure in India suggests that a PPP exchange rate of ₹20.15 to the dollar was taken for 2017 and just ₹19.47 per dollar for 2021. This drop is surprising, given that prices in India increased by roughly 20% during this period, faster than the 8% increase in the US. The implication in nominal rupee terms is that the World Bank international poverty line does not effectively increase much between 2017 and 2021, even though the line in dollar terms has been revised upwards. 

The World Bank does improve on how it arrives at its state and sectoral poverty lines. Unlike its previous ad-hoc adjustments, it now uses the Tendulkar panel methodology to adjust prices spatially. It has also revised its earlier poverty estimates to align with  ‘recall period’ changes in the 2022-23 HCES. Both these moves are welcome. 

Also Read: Mint Quick Edit | Poverty isn’t widespread but prosperity needs to be

The most important change, however, is an adjustment in the reported consumption expenditure in the 2022-23 HCES. 

This is motivated by a decision to move away from the concept of consumption expenditure to that of a real welfare aggregate. Consequently, the World Bank now adds the value of free and subsidized items received to household expenditure. So free foodgrains received by households as part  of the Pradhan Mantri Garib Kalyan Yojana are accounted for, as also school uniforms and footwear. 

The inclusion of free items increases the average  consumption expenditure only by 2.8%,  but it does so by almost 10% among the poorest decile. As a result, poverty estimates are lower. It also excludes expenditure on durables, hospitalization and other household items. Since most of these are consumed by better-off households, it suggests a sharper decline in inequality than otherwise observed with unadjusted data. 

Also Read: Himanshu: India needs official poverty data for effective policymaking

While there is no doubt that free goods—food or otherwise—improve household welfare, it also opens up a Pandora’s Box on which goods and services should be included and excluded. Household welfare also improves on account of free and subsidized health, education, law and order, security, freedom, a clean environment and so on.  

In fact, almost all services provided by governments can be treated as welfare-augmenting. By adding these to private consumption, the World Bank has revised the definition of poverty. It has also created scope for governments to include all their budgetary spending in household expenditure. This also opens up the possibility of data manipulation if governments raise subsidies as a quick way to reduce poverty. Some of these are basic questions and have been debated earlier, but perhaps we need a deeper debate now. 

Also Read: India must redraw its poverty line to reflect economic progress

It is not clear whether the same rules apply for all countries, since many do not have any estimate of these benefits accruing to households. The issue of comparability of consumption surveys and an appropriate PPP index may be unique to India, but the move towards a completely different concept of private consumption that includes government spending is relevant for many countries. The World Bank must resolve these issues for the credibility of its poverty estimates. 

The author is associate professor at Jawaharlal Nehru University and visiting fellow at the Centre de Sciences Humaines, New Delhi.