What the long-due consumption survey data shows—and what it doesn’t
- The 2022-23 survey was conducted using a vastly different methodology in an effort to capture consumption trends better. That means comparisons with past surveys are inappropriate, and any poverty estimation using it is premature
On Saturday, the statistics ministry unveiled preliminary findings from the consumption expenditure survey conducted between August 2022 and July 2023. It’s only a matter of time before the government releases more data, that will enable deeper, more credible analysis.
Yet, this teaser has sparked numerous analyses, with some experts suggesting the rural economy may be more robust than anticipated and that less than 5% of Indians now live below the poverty line. But with limited data available, drawing such conclusions can be premature, especially ahead of a hot election campaign.
The data brief shows that in 2022-23, average monthly per capita spending reached ₹3,773 in rural areas and ₹6,459 in urban India. Prima facie, this indicates that household budgets have risen an average of 9.2% annually in rural areas and 8.5% in urban centres since 2011-12, the last year for which official data was available. However, when adjusted for inflation, these rates come down to 3.1% and 2.7%, respectively. While not a dramatic improvement in living standards, it does suggest that rural consumption saw faster growth.
That said, it’s crucial to note that the two surveys are not comparable due to wholesale changes adopted in data collection methods after the government junked the results of the previous survey held in 2017-18, citing “data quality" issues (critics claimed it was due to adverse results). The new methodology tries to capture data better: the number of consumption items covered was increased from 347 to 405, the number of visits to each household was increased from one to three to avoid long interviews, and the pen-and-paper survey was replaced with an electronic one. Comparisons, therefore, are inappropriate, experts pointed out. (The problem of outdated definitions used to define rural and urban areas, based on Census 2011, remains.)
Work in progress
The improved methodology helps in better coverage of data, and, as a result, could have led to a slight increase in expenditure, thus implying lower poverty, according to P.C. Mohanan, former acting chairperson of the National Statistical Commission. However, given the changes, the survey should have canvassed a small sample of households to collect data based on the old methodology to estimate the impact of better coverage, he said. He cautioned against any comparison with earlier years, and said the new survey should ideally be treated as a “stand-alone" report.
Pronab Sen, former chief statistician, who spearheaded the survey, concurred with Mohanan on the need to canvas a smaller sample using the old methodology “at some point". The plan is to test the new methodology in two back-to-back surveys (2022-23 and 2023-24) to check its robustness, he said. “We will have to wait for the results of the (ongoing) 2023-24 survey as well to see if this particular methodology is appropriate," he added.
The survey also captured the impact of free goods given by the government, which lower a household’s actual spending. This is estimated by “imputing" the value of these items, such as free food grain, laptops, two-wheelers, and school uniforms. It raises effective consumption by ₹87 per capita per month (2.3%) in rural areas, and by ₹62 (1%), in urban areas. In urban areas, richer groups get less of these benefits, but in rural areas, the bottom 5% spenders got the least: ₹68 per capita, against ₹80-93 for all other classes. According to Sen, this suggests there may be less coverage of welfare schemes among the poorest 5%.
Poverty of data
The two consumption surveys are expected to lay down the foundation for revisions in several critical datasets such as the consumer price index (CPI) basket and poverty headcount. However, these would take time.
From an inflation standpoint, the 2022-23 survey showed that food items now make up less than 50% of rural household spending. It should lead to lower weightage for food in the inflation basket, which could make headline inflation less volatile. However, the long-due overhaul in the CPI basket will have to wait until the results of the 2023-24 survey is also out.
It’s the same for poverty estimates. B.V.R. Subrahmanyam, chief executive of the government think tank Niti Aayog, has reportedly claimed, citing the survey, that poverty may now be less than 5%, likely based on the 2011-12 poverty line adjusted for inflation. But experts caution against an approach based on limited data with an old poverty line derived from consumption patterns assessed using a vastly different methodology.
“It is not appropriate to apply the same poverty line," Sen said. “Poverty estimates may require a more careful analysis of the data by the experts in the field."
A more detailed report and household-level raw data, which is expected in the coming months, may offer a better analysis. But that, too, won’t be enough to draw conclusions on the changes in living standards since 2011-12. We need to wait for the results of the 2023-24 survey to make any comparative analysis. Meanwhile, it will be important to fend off election-driven celebratory narratives based on limited information.
topics
