Has Indian inequality declined? Post-tax income data holds the answer

A recent World Bank (WB) report has claimed that between 2011-12 and 2022-23, India significantly reduced consumption inequality. India’s Gini coefficient (a measure of inequality) is ranked as the fourth-lowest in the world. The WB’s estimates of Indian inequality are based on the official Household Consumption Expenditure Survey (HCES) 2022-23 data, after accounting for some but not all government-provided free goods, and excluding consumer durables. 

Many who are accustomed to media reports claiming high inequality in the country have responded to the WB’s claims with scepticism or outright dismissal. The critics argue that, as HCES data does not capture consumption by the elite, the WB has underestimated consumption inequality. Furthermore, they cite World Inequality Lab (WIL) studies to argue that India has a very high level of income inequality. For an informed debate, we must encounter the devil in the details. 

All survey data across countries fail to capture the elite consumption and income. The problem is universal. India is not an exception. The HCES 2022-23 uses the Modified Mixed Recall Period (MMRP) method in line with international best practices, making the Indian data suitable for international comparisons. 

While one can quibble over the precise decline in India’s consumption inequality — whether the Gini decreased precisely from 28.8 in 2011-12 to 25.5 in 2022-23 or by less — the decrease is substantial and indisputable. Similarly, a significant improvement in India’s international ranking is factual. 

Regarding income inequality, the media and commentators are particularly fixated on the national income shares of the top 1 per cent, as estimated by the WIL, often unaware of the critical limitations of these estimates. 

Unlike consumption, there is no official data yet on income distribution in India. So, the WIL derives income distributions using tax data for the top income groups. It uses old consumption data and its estimates of the income-consumption relationship for low- and middle-income households. The latter assumes that for 70-80 per cent of households, the consumption expenditure exceeds their income. Simply put, except for the top 20-25 per cent, all families spend more than they earn, year after year! 

As an implication of this implausible assumption, the income of the bottom 80 per cent is underestimated. This pulls down their national income shares. Conversely, the shares of top income groups are overestimated. 

Let’s ignore these limitations and ask: Is income inequality increasing? No. Taking the WIL estimates as is, between 2017 and 2022, the income shares of the bottom 50 per cent increased from 13.9 per cent to 15 per cent, while those of the top 10 per cent decreased from 58.8 per cent to 57.7 per cent. 

A high share of the top 1 per cent is a concern. However, since 2016-17, the income shares of the top 1 per cent has increased by only 0.3 percentage points. Due to anti-tax evasion measures taken during this period, the high-income groups report more truthfully than before. My research indicates that most of the increases in top income shares since 2014 are attributable to better tax enforcement. Improved compliance should not be mistaken for increased inequality. 

To be meaningful, income inequality should be assessed based on post-tax, rather than pre-tax, incomes. For instance, in the assessment year 2023-24, the top 1 per cent of individual taxpayers paid 42 per cent of the total tax. Considering all taxpayers, the top 1 per cent paid 72.77 per cent of the total tax. This means that the post-tax income of top-income groups is only 65 to 75 per cent of their incomes taken in WIL estimates. In contrast, for low-income groups, the estimates do not account for the all-time high welfare transfers, which amount to approximately 8 per cent of gross domestic product (GDP). On a post-tax, post-subsidy income basis, one will find a decline in income inequality over the last decade. 

What has gone unnoticed is that, in recent years, inherited wealth is not the primary determinant of individual incomes. As many as 60 per cent of the top-income reporters are first-generation entrepreneurs, cricketers, and other leaders from the fintech and unicorn spaces. India has produced the largest number of first-generation billionaires, several in their 20s and early 30s. 

From a macroeconomic point of view, the major determinants of income inequality are the rate of return on capital vis-à-vis the GDP growth rate. When the growth rate is higher than the post-tax returns on capital, an increasing share of national income goes to labour. This reduces inequality, ceteris paribus. Data on the weighted average real lending rates and average returns on capital adjusted for risk, inflation, and taxes suggest that the rate of return is smaller than the growth rate. This, along with the Centre’s fiscal support, has made growth inclusive. 

Data speaks for itself. Between 2012 and 2025, India has pulled around 320 million out of extreme poverty, based on the International Poverty Line of $3 (at 2021 purchasing power parity). Per capita consumption of fruit, vegetables, milk, eggs, and other protein-rich products is at an all-time high. As the two most recent rounds of HCES show, the share of cereals in total calorie intake has decreased, while that of healthier products has increased, for all strata. Improvements are most striking for the bottom 20 per cent. If we factored in all the policies targeted at the poor, such as Ayushman Bharat, the aggregate welfare levels would look even better than the WB numbers suggest. 

Admittedly, we need to do much more. The inequality can be reduced further by plugging the loopholes in tax laws. But let’s celebrate India’s successes too. 

The author is director, Delhi School of Economics, and a member of the RBI’s Monetary Policy Committee and the Technical Expert Group for the first Household Income Survey by Mospi.  The views are personal