There has been a great deal of debate about the extent to which the Indian economy has recovered after the Covid-19 pandemic. The first advance estimates of Gross Domestic Product for 2023-’24 announced early in January by the Ministry of Statistics and Planning indeed point to demand side weaknesses in the economy. Private final consumption expenditure is expected to grow at a mere 4.4%, the slowest since 2002-’03, if we ignore the Covid-induced contraction in 2020-’21.
A narrative has emerged about a “K-shaped recovery”, arguing that after the pandemic, the poor have got poorer while the rich have got richer. This refers to a situation in which the income increases for a section of the economy (often the rich) but deteriorates for another section (often the poor), showing directionality of growth similar to the letter K.
On the other hand, a report by the State Bank of India on January 8 titled “Debunking K Shaped Recovery” has questioned this narrative. The study is largely based on claims of increased upward mobility and the reduced inequality among income tax payers and data on Micro, Small and Medium Enterprises.
The study claims that 36.3% of people who file income tax have moved from earning less than Rs 3.5 lakh to higher income groups and so concludes that “growth is seen in all income classes and its skewness has been decreasing with convergence of income towards the middle from both top as well as bottom”.
This should be understood in context: merely 66.5 million individuals pay personal income tax in India. They constitute 4.8% of the total population and 6.3% of the adult population.
The claim of upward mobility within the same income groups is hardly convincing given that during the same period, nominal GDP has grown by more than 45%, leading to nominal growth in income. This would automatically have lifted some taxpayers into higher income brackets.
We do not have any information on those who could have moved in the reverse direction. In addition, the claims that inequality among taxpayers has declined is not particularly useful since we have no data on the vast majority whose incomes are not taxable.
On the other hand, the study’s finding that Micro, Small and Medium Enterprises expansions have led to the expansion at the base of the pyramid is not very convincing evidence for across-the-board improvements in incomes.
‘K or not-K’ debate
To investigate the K debate, we start by examining the changes in household incomes from the Income Pyramids database of the Centre for Monitoring Indian Economy. These data have been criticised for the sampling frame the centre uses.
However, it is the only dataset that gives household incomes at a monthly frequency. The major problem with the data is that they miss out on poor, homeless and mobile households, thus under-stating poverty. We will be mindful of this shortcoming while interpreting the results.
A true interpretation of the K-shape recovery process would mean that post-pandemic incomes have fallen on the lower quantiles of the income distribution while increasing over the top quantiles. We simplify the issue by looking at household incomes at the bottom 20th and top 90th percentiles. We did this analysis for all India as well as for a few select states.
A true K-shaped recovery would characterise those states where incomes have fallen at the bottom 20th percentile and has increased at the top 90th percentile. We distinguish between rural and urban parts of states, since the recovery process is likely to be quite different across the two regions
We treat January 2020 as the last pre-Covid month, since Covid cases were first detected in India in that month. Given the seasonal nature of incomes in India, we compare January 2020 with January 2023.
For rural India as a whole, the incomes of the lowest 20 percentile increased from Rs 8,000 per month to Rs 8,900 per month. The income of the top 90 percentile however, actually fell from Rs 39,476 to Rs 38,800. This is not indicative of a K-class recovery, but a movement towards the middle, as hinted at in the State Bank Report. However, it is not a substantial reduction in inequality, with the rural Gini coefficient dropping marginally from 0.4577 to 0.4516.
Urban India saw a similar movement too. The income of the lowest 20 percentile moved from Rs 12,000 per month to Rs 13,989 per month, while that at the top 90 percentile fell from Rs 50,800 to Rs 48,400. Inequality declined from 0.3926 to 0.3385.
Given that the Indian economy is large and spatially diverse, and that many of its states can be considered large economies in their own right, we studied the issue by examining what is happening at the level of states that make a significant contribution to the Indian GDP. Maharashtra, Uttar Pradesh, Tamil Nadu, Gujarat, and West Bengal together account for almost 50% of India’s GDP.
We start with Maharashtra, which accounts for 15.7% of India’s GDP. Unlike rural India, rural Maharashtra shows a clear case of K shape recovery. In rural Maharashtra, incomes of the bottom 20 percentile households actually fell from Rs 6,200 per month to Rs 6,000 per month over January 2020 to January 2023. On the other hand, incomes at the top 90 percentile increased from Rs 42,100 to Rs 50,890.
The urban pattern however , was in consonance with the rest of urban India, with income at the bottom 20 percentile rising from Rs 12,150 to Rs 14,000, while those at the top 90 percentile falling from Rs 65,000 to Rs 45,120. The rural inequality increased from 0.48 to a substantial 0.52, while urban inequality fell from 0.41 to 0.33.
Uttar Pradesh and Tamil Nadu are the next two states, making a very similar contribution to Indian GDP (around 9.2 % for Uttar Pradesh and 9.1% for Tamil Nadu).
As far as rural Uttar Pradesh goes, there is an across-the-board improvement, with incomes rising at the bottom 20 percentile as well as the top 90 percentile , but with a higher increment at the bottom. Consequently, rural inequality fell from 0.49 to 0.47. However, the urban story is similar to the rest of India, with income gains at the bottom 20 percentile but income losses at the top 90 percentile , and a fall in inequality from 0.41 to 0.31.
Tamil Nadu also experienced an increase in the bottom 20 percentile of rural incomes (from Rs 9,704 to Rs 13,449 ) and a fall in the top 90 percentile (from Rs 27,248 to Rs 26,540), with a decline in the Gini coefficient from 0.27 to 0.20. Urban Tamil Nadu saw an improvement in the incomes of the bottom 20 percentile from Rs 10,500 per month to Rs 12,560 per month. At the top, income at the top 90 percentile declined from Rs 33,500 to Rs.27,700 per month.
Gujarat accounts for 8.2% of India’s GDP. Rural Gujarat saw a decline in inequality as the incomes of the bottom 20 percentile increased from Rs 7,500 to Rs 8,300 while those at the top fell from Rs 30,000 to R 27,480, leading to a fall in the Gini from 0.57 to 0.44. The Gujarat urban story reflects that for urban India as a whole, with incomes from the bottom improving but those at the top going down, with inequality declining from 0.33 to 0.29.
Rural West Bengal too saw an improvement across the board, with rural incomes at the bottom going up from Rs 7,545 per month to Rs 10,700 per month, with inequality going down from 0.28 to 0.24. Urban West Bengal reflects the experience of the rest of urban India with incomes improving at the bottom, but dropping at the top, with Gini declining from 0.40 to 0.30.
So, the K-shaped story of increased inequality does not seem to hold, with the exception of rural Maharashtra.
The postulate of income increases across the board also does not hold as we see that while there have been income improvements in the bottom but losses at the top for both rural and urban regions. Hence we can say that it is neither a K-shaped recovery nor an across-the board income increase.
Then what explains the persistent weak demand in the economy?
For this we need to understand where the demand gets generated. This is explained by precisely the kind of movements we have observed. Incomes at the lower end have improved, but not enough to generate even moderate quantities of disposable income.
Take the report of the Expert Committee on Determining the Methodology for Fixing the National Minimum Wage, appointed by the Ministry of Labour and Employment, from 2019. For the states Gujarat, Maharashtra and Tamil Nadu, the committee recommended a daily minimum wage for a family of two adults and two children under 14 as Rs 502, translating into a monthly income of Rs 15,060.
This would take into account the expenditure necessary to meet a balanced diet, and reasonable spending for essential non-food items like clothing, fuel, light, house rent, education, medical expenses, footwear and transport.
The median household was barely above the threshold. These households are not likely to have much disposable income, notwithstanding the improvements at the bottom of the pyramid.
Decline in incomes at the top?
It is important to understand that, with the exception of a handful of really wealthy people, being in the top 10% of the income bracket does not necessarily make a person rich in the usual sense of the term of having vast amounts of disposable incomes.
For rural India, a monthly household income of Rs 38,800 in January 2023 was enough to put a household in the top 10% income earners. For urban India, this figure is Rs 48,400. These households do have a limited amount of disposable income.
But post-Covid-19, especially urban India, the incomes of these households have been hit. This is what has led to a loss of purchasing power in the economy, given that households account for significant consumption expenditure in the country. This is what probably explains the slowdown in the sale of consumer durables, fast moving consumer goods, low-cost housing and more.
The point to note here is that while the demand for basic needs of the poorest have been taken care of, to a large extent, by various welfare schemes, the decrease in disposable income of the not-so-poor, coupled with no social security support, has severely hampered the overall demand situation in the country.
The authors are at the School of Development, Azim Premji University, Bengaluru. The views expressed in the article are their individual views and the University does not necessarily share the same. Research assistance by Prasanna Surathkal and Sanket Gharat is gratefully acknowledged.
This article is part of the works at the Development Dialogues with Data initiative of the School of Development. Communication can be sent to this email ID: neeraj.hatekar@gmail.com