Home Insights Consumption Survey Analysed – IMPRI Impact And Policy Research Institute

Consumption Survey Analysed – IMPRI Impact And Policy Research Institute

IMG 1203 e1709360941837

Arun Kumar

The methodology of data collection for the latest Household Consumption Expenditure Survey has changed in this round. Therefore, the survey results raise a lot of questions since alternative data is at variance with it.

The government has released the Household Consumption Expenditure Survey (HCES) for the period of August 2022 and July 2023 earlier this week.

The survey generates estimates of households’ Monthly Per Capita Consumption Expenditure (MPCE) and its distribution. However, only the summary results have been released as of now, in the form of a factsheet (along with Annexures). The detailed report is expected to be brought out subsequently.

The survey report has been released after 2011-12. In between, a survey was done in 2017-18 but its report was not released by the government because it did not suit its narrative. This gives rise to suspicion that reports are released selectively. It also gives rise to the possibility of manipulation of data collection and the way it is officially presented.

HCES 2022-23 does show consumption rising across the board while the leaked report of 2017-18 showed a decrease in consumption. This has enabled officials to claim that poverty in India has declined and less than 5% of the population is poor. At current prices, the increase in average consumption looks impressive. It increased 164% in rural areas to Rs 3,773 and in urban areas by 146% to Rs 6,459. But most of it is due to inflation. Adjusting for inflation, the real increase is 40% and 33.5% for rural and urban areas, respectively. This is in 11 years.

During this period, officially, the net national income has risen from Rs 77.42 lakh crore to Rs 133.48 lakh crore, which is an increase of 72.34%. In other words, consumption growth is way behind the growth in incomes.

Inference on inequality? 

If the poor were getting higher incomes, they would spend most of it in consumption since they are unable to fulfil their basic needs. Their consumption should have risen at a much faster rate than the new data shows. It can be inferred that most of the income increase has been for the higher echelons of the income ladder. These well-off save a large fraction of their incomes since they do not need to spend much more on increasing their consumption. The implication is that inequality in incomes is increasing in the country.

What the HCES data shows is the consumption inequality. The more relevant factor is income inequality. And, that is much higher than the consumption inequality since those with higher incomes save more.

This is consistent with the argument that the organised sector of the economy is growing while the unorganised sector is declining. The unorganised sector consists of agriculture and micro and small sectors of the economy. The agricultural sector has been in crisis for a long time as evidenced by the repeated protests staged by farmers. The micro and small sectors’ growth has been declining since at least demonetisation in 2016. This is consistent with the growth pattern of the economy being K-shaped.

The latest taxation data also suggests the same. Officially, it is said that there is high buoyancy – that is, tax collections are rapidly rising. But, in the Indian economy, the vast bulk of the tax is paid by the organised sector. Corporation tax and income tax are paid by the well-off. The prime minister has said that only 1.5 crore individuals (1.1% of the population) are effective taxpayers. GST is collected largely from the large- and medium-scale units since the unorganised sector is largely exempt or in the composition scheme paying a tax of 1%.

Inference on poverty 

The NITI Aayog has been quick to claim that the data supports the argument that poverty has decreased in India from 2011-12 to 2022-23. The data shows a drop in the share of expenditure on food.

However, several factors must be taken into account when saying that poverty has declined.

According to Engel’s Law, the percentage of income allocated for food purchases decreases as a household’s income rises, while the percentage spent on other things (such as education and recreation) increases.

Firstly, the volume of consumption is not given. It could be that the increased absolute expenditure on food items is a reflection of an increase in the prices of those items. When the full data is released, then only one would understand this better.

Second, what is the poverty line that is being used to claim that poverty has declined? Poverty has to be defined as ‘minimum social necessary consumption’. This is space and time specific. So, it keeps changing. The World Bank has changed its poverty line recently from $1.9 to $2.15 per person per day. This amounts to about Rs 26,000 per family of five per month. Even adjusting for nominal dollars, it would be about Rs 10,000 per family per month. If this poverty line is considered, then the number of poor would be much more than the 5% being quoted by officials.

Third, 300 million people have registered on the e-shram portal, and 90% of them have said they earn less than Rs 10,000 per month. Given the level of unemployment, this would imply that most of these people would fall in the category of being poor.

Fourth, the share of expenditure on health and transportation etc. has increased. For the poor, an increase in the cost of travel would be necessary to earn their livelihood, and would not mean that they are better off. Health expenditures are likely to be even higher than stated, given the recovery from the pandemic, and it would not represent a better off family but a poor family struggling to survive.

Fifth, the increase in demand for work under MGNREGS means that people’s incomes have not recovered to pre-pandemic levels. They are still struggling.

Sixth, the level of malnourishment remains high among women, as per the NFHS data. Would these families be spending a lower percentage of their income on food? Would they be buying more of the prepared food (packaged) from the market?

Seventh, high unemployment, under-employment and low labour force participation ratio suggests that incomes could not have recovered, and correspondingly, consumption could not have recovered.

Eighth, could there be higher consumption due to the government welfare measures. The data released provides information on the increase in consumption on account of these programmes giving free items (called ‘revdi’ by the prime minister) like food. On an average, this increases consumption by Rs 87 in rural areas and Rs 62 in urban areas. It is even lower for the poorer sections. It amounts to an increase in consumption of 2.3% for the rural areas and 1% for urban areas. Thus, either the argument that the government is giving a lot of support to the poor is incorrect or the benefits are not reaching the poor.

In brief, more detailed data would be needed to say something definitive about poverty declining.

Conclusion: Premature comparison

The official document mentions that the methodology of data collection has changed in this round compared to earlier. It is no more a consumer survey but a consumption survey. Consumption is split up into three broad categories and the data is collected separately on each one of them. In surveys when the method of collection changes and the questionnaires change, responses become non-comparable.

There has always been a problem in collecting the data of the wealthy and the poor. At both ends of the spectrum, the enumerators do not get a proper response. Thus, even inferring about inequality becomes difficult.

In brief, it is good that the data on consumption has finally come out. But it raises a lot of questions since alternative data is at variance with it. There is also a controversy about the follow up survey being done to test if the 2022-23 survey is robust. This is necessary since the methodology has been changed compared to the earlier consumer surveys. Maybe the full survey results will help clear some of the doubts.

Arun Kumar is a Retired Professor of Economics at the Jawaharlal Nehru University. He is the author of `Demonetisation and Black Economy’ (2018, Penguin Random House). 

This article was first published in The Wire as Which Data Can We Rely On? on February 29th, 2024.

Read more by the author: Beyond Electoral Bonds: Tackling Black Money in Indian Elections.

Disclaimer: All views expressed in the article belong solely to the author and not necessarily to the organisation.

Acknowledgment: This article was posted by Puspa Kumari, a research Intern at IMPRI.

Previous articleChild Marriage Curb On Socio-Economic Development In India – IMPRI Impact And Policy Research Institute
Next articleUnlocking Opportunities: India And Mauritius As Catalysts For Investment And Development In Africa – IMPRI Impact And Policy Research Institute
IMPRI, a startup research think tank, is a platform for pro-active, independent, non-partisan and policy-based research. It contributes to debates and deliberations for action-based solutions to a host of strategic issues. IMPRI is committed to democracy, mobilization and community building.


Please enter your comment!
Please enter your name here