How do Indians divide their consumption expenditure across categories? How many are below the poverty line? How has the picture changed over time? Answers to such questions have always been drawn from the quinquennial consumption expenditure surveys. After a decade-long hiatus, a factsheet based on the revamped and much awaited Household Consumption Expenditure Survey (HCES) for 2022-23 has been released.
Based on the factsheet data, can we decisively answer these and other such questions? Not really, for reasons explained below. This piece comments on a few key takeaways.
The factsheet gives a useful glimpse into present patterns of household consumption. There have been quick comparisons with earlier rounds of the consumption expenditure survey, especially with the last publicly available 68th round in 2011-12, to deduce a dramatic reduction in poverty over time and/or drastic changes in consumption patterns. Such comparisons are premature — the full data has not been released and there are several changes in item coverage, survey design and methodology.
The 2011-12 round covered 347 items of consumption compared to 405 in HCES. Such revisions have been done in the past too and are par for the course as consumption habits change over time.
There are three major changes that make the current survey non-comparable with its predecessors. First, the earlier rounds had one single questionnaire to cover expenditure on all items. The new HCES has three separate surveys on food items, consumable and service items and durable goods respectively. Second, the HCES involved multiple separate visits to households.
As data collection methodologies, both these changes are welcome because it is well-known that respondent fatigue with long questionnaires in a single sitting leads to poorer responses. Shorter and more focused questionnaires elicit more precise answers. However, these changes have implications for comparability — these are likely to yield higher estimates compared to previous rounds.
To check for the magnitude and direction of this upward bias, ideally the HCES should have chosen a subset of households to which the older survey with its single-visit design should have been administered. A comparison between the older and revised methodologies would have given us the magnitude of the difference.
The third difference between HCES and earlier rounds relates to imputed values for items received or consumed free of cost through various social welfare programmes. These items range from staples like rice, wheat to footwear, laptops, and motorcycles. We can’t assess the validity of this until unit-level price and quantity data are released.
We can examine the HCES data for within-survey indicators, that is, inequality.
Average all-India urban monthly per capita consumption expenditure (MPCE) at Rs 6,459 in 2022-23, is roughly 72 per cent higher than in rural areas (Rs 3,773). The corresponding figure was 84 per cent in 2011-12. This seems to indicate a decline in rural-urban inequality over the decade.
However, there are two caveats. This does not account for the rural-urban price differential, which can fluctuate. Additionally, a longer view reveals that the rural-urban gap tends to fluctuate. From 75.9 per cent in 1999-2000, it rose to 90.8 in 2004-05, and then declined to 83.9 in 2011-12. We don’t know if the ratio increased or decreased in 2017-18.
How has the distribution of consumption expenditure changed over the decade? In 2011-12, the ratio of the 10th percentile of the rural expenditure distribution (Rs 710) to the 90th percentile (Rs 2,296) was 0.31. This ratio is 0.33 in 2022-23 (Rs 1,782/Rs 5,356). In other words, the consumption expenditure of the bottom 10 per cent of the rural distribution is roughly one-third of the top 10 per cent — a ratio that has not changed substantially over the decade. For urban areas, the corresponding ratio was 0.21 in 2011-12, which has risen to 0.27 in 2022-23, indicating a reduction in urban inequality.
What about inter-caste differences? The ratio of average rural Scheduled Caste (SC) MPCE to that of the higher-ranked castes was 0.73 in 2011-12. This has remained unchanged at 0.7 in 2022-23. For rural Scheduled Tribes (ST), this ratio has improved marginally from 0.65 to 0.69 and for Other Backward Classes (OBC), from 0.83 to 0.87.
While rural India does not show a marked narrowing of inter-caste gaps in MPCE, the urban figures suggest a narrowing of inter-group MPCE gaps: For SCs, the ratio increased from 0.63 to 0.72; for STs, from 0.68 to 0.74 and for OBCs, from 0.7 to 0.84.
The factsheet reveals that in rural India, the percentage share of cereals — a food staple — in average MPCE is now 4.91 per cent and 3.64 per cent in urban India, compared to 10.75 per cent and 6.66 per cent respectively in 2011-12. This is accompanied by an increase in the share of processed foods and beverages.
Overall, HCES indicates a decline in the share of food expenditure over the decade, which stands at 46 per cent in rural and 39 per cent in urban India. What are the items whose share in total expenditure has increased? Medical expenses on hospitalisation, conveyance, consumer services and durable goods in rural India. In urban India, the share of paan, tobacco and intoxicants, conveyance and durable goods has increased.
The release of the survey findings, however limited, is a welcome development. For data to inform policy and analysis of well-being, this should be followed up by a quick release of the full price and quantity unit-level data for 2022-23, as well as from the follow up survey for 2023-24.
As the economy diversifies and grows, it is only natural that consumption habits change and our surveys need to be able to capture that. However, as the adage goes, for surveys to measure change, the change in the survey instrument must be minimal. Combining these two imperatives is not impossible, as long-standing longitudinal panel surveys from other countries show us. The HCES has introduced welcome changes in survey design, but missed the opportunity to build the concordance bridge to previous rounds.
The writer is professor of economics and founder director, CEDA, Ashoka University. Views are personal
login
who are we?
contact us
qosheapp
Toi Staff
Gideon Levy
Belen Fernandez
Rami G Khouri
Mort Laitner
Rachel Marsden
Bradley Blankenship
Michael Kwet
visit website
