Economy
Mint Primer: How to read SBI’s latest poverty estimates
Summary
- A recent SBI report has stirred the proverbial hornet’s nest with its claims of a sharp reduction in India’s poverty levels, particularly in rural areas, in the past decade. It has led to a flurry of counter claims, with economists questioning the methodology used to arrive at the findings.
A recent report by the State Bank of India has stirred up the proverbial hornet’s nest with its claims of a sharp reduction in poverty levels in India, particularly rural areas, in the past decade. It has led to a flurry of counter-claims, with economists questioning the methodology used to arrive at the findings. Mint explains what the matter is about.
What does the SBI report say?
A brief factsheet of the latest annual Household Consumption Expenditure Survey (HCES), held in August 2023-July 2024, was released in late December. Based on that, the SBI report says the poverty ratio in rural areas declined to 4.86% in the period (i.e. 4.86% of rural Indians are below the poverty line), down from 25.7% in 2011-12, whereas in urban areas, it fell to 4.09% from 13.7%, indicating that rural poverty has fallen faster. This puts all-India poverty levels somewhere between 4% and 4.5%, with prevalence of extreme poverty “almost minimal", the report claims.
The report attributes this decline to government initiatives in rural India such as direct cash transfers and investment in physical infrastructure, which, it said, have improved consumption growth in the bottom 5% of the population.
How does the report arrive at the conclusion?
The most commonly used method to estimate poverty in India is through measuring consumption levels. If a household’s consumption is below a minimum monetary threshold, it is seen as being ‘below poverty line’. Data is captured through Household Consumption Expenditure Surveys (HCES), which gather information on households’ expenditure on goods and services.
The Suresh Tendulkar Committee calculated the poverty line at ₹816 per capita per month for rural areas and ₹1,000 for urban areas in 2011-12. The SBI report adjusts these figures for inflation to determine the 2023-24 poverty line at ₹1,632 (rural) and ₹1,944 (urban).
The respondent-level data of the HCES, which could tell us exactly how many Indians spend less than these levels, is still awaited. But the factsheet released by the government did say that the bottom 5% of the population spend an average of ₹1,677 (rural) and ₹2,376 (urban) per capita per month.
The report’s poverty headcount figures of 4.86% and 4.09% rely on the simple assumption that these 5% Indians have uniform consumption patterns, and so the figure can be interpolated. In that sense, it’s a premature back-of-the-envelope estimate, but it’s clear that if we do assume the SBI report’s adjusted poverty line, the headcount would indeed be less than 5%. But the question is: is that adjusted poverty line reliable?
Why are the findings being contested?
For starters, the consumption surveys from 2022-23 onwards have seen revamps in data collection methods, such as an increase in the number of questionnaires and household visits, making the data difficult to compare to past ones. The surveys are using a modified mixed reference period (MMRP) method, in which households are asked about their consumption in the past week for some items, in the past 30 days for some, and in the past 365 days for the rest.
The Tendulkar committee’s poverty line is based on an older survey that used a mixed reference period (MRP), which was based on 30-day and 365-day recall periods. It’s unclear how adequately the SBI report adjusts the Tendulkar poverty line for this, if at all. Experts also contend thatconsumption patterns have undergone changes in the past decade, therefore to use a decade-old poverty line to compute the present one could lead to underestimation.
How does India estimate poverty?
India has had several committees since independence that were tasked to gather poverty estimates. TheTendulkar committee (2009) had estimated India’s povertyratio at 21.9% (25.7% rural; 13.7% urban) in 2011-12. However, the findings faced criticism from some quarters for using a ‘low’ poverty line, leading to underestimating the number of poor. Another committee, chaired by C. Rangarajan, was set up, and it suggested a new method (using MMRP method), estimating much higher poverty levels (29.5%). The report remains untapped.
The lack of a decennial census since 2011 has also posed a hurdle in understanding the extent of poverty, P.C. Mohanan, former head of the National Statistical Commission, told Mint. “Many larger villages are almost urbanized now, and those things are not accounted for when we discuss rural-urban dichotomy because we have had no census," he said.
Are there other economic indicators to understand India’s poverty trend?
The Niti Aayog’s multidimensional poverty index is based on a global index by the same name. Moving beyond income and consumption, it considers deprivation of services across health, education and living standards, using data from national family health surveys (NFHS). The last iteration of the report in 2023 (based on NFHS 2019-2021) concluded that the share of India’s population who are “multidimensionally poor" fell to 11.28% in 2022-23 from 29.17% in 2013-14.
Growth in wages can be another indicator of measuring poverty, an indication of purchasing power. Data from the government Labour Bureau shows that wage growth in rural areas in real terms (after deducting annual inflation in rural India) has been stagnant at 0.8% for agricultural workers in the past decade.
To understand poverty levels, India needs a new estimation methodology for a poverty line, which could then be applied to the HCES findings.
