Recalculate inequality — it hasn’t worsened over time, it’s stayed the same

Piketty and Chancel analysed different periods of taxation in India, and concluded that inequality dropped sharply from 1950 to the early 1980s under Nehru-Indira’s socialism, but then rose sharply in the era of economic liberalisation. Was the Nehru-Indira era good for the poor? No. The poverty ratio remained unchanged from 1947 to 1980.

The world grew fast after 1991 when countries moved toward liberalisation and globalisation. But critics say inequality has soared. Thomas Piketty, using income-tax data rather than survey data, concluded that the share of the top 1% in all US income soared from 9% to 19% between 1962 and 2019.

This dismal estimate has been exposed as a statistical bloomer in a peer-reviewed September 2023 paper ( by Gerald Auten and David Splinter. They show that, calculated correctly, US income inequality has been unchanged since 1960. This contradicts the narrative of inequality widely taken to be gospel truth.

What about India? Lucas Chancel and Piketty wrote a paper in 2017, estimating Indian inequality from 1922 to 2015 based on tax data and other sources. I attacked Piketty in a series of newspaper columns, followed by a November 2017 Cato Institute paper, ‘Piketty’s Botched Analysis of Inequality in India‘ ( Many of my criticisms of Piketty’s methodology have been repeated by Auten and Splinter, to considerable acclaim.

Taxpayers respond to incentives. In the US, Ronald Reagan slashed taxes after coming to power in 1980. Earlier, rich individuals sought complex tax shelters to reduce their tax liability. But once I-T rates were slashed, many taxpayers abandoned the tax shelters and reported more income.

Before Reagan, tax rates were much higher on individual income than corporate income. So, companies retained most of their earnings in their reserves, and did not distribute them as dividends (on which the rich would pay heavy taxes). This reduced the apparent income of the richest people. But after Reagan slashed tax rates, it became attractive for companies to declare high dividends. Much of the post-Reagan surge in reported incomes of the rich merely represented reduced use of tax shelters and increased dividends.

Piketty and Chancel analysed different periods of taxation in India, and concluded that inequality dropped sharply from 1950 to the early 1980s under Nehru-Indira’s socialism, but then rose sharply in the era of economic liberalisation. Was the Nehru-Indira era good for the poor? No. The poverty ratio remained unchanged from 1947 to 1980.

Since the population almost doubled, the absolute number of poor almost doubled, a terrible outcome. By contrast, poverty plummeted after liberalisation. Around 415 million people were lifted out of poverty between 2005 and 2021 according to the Global Multidimensional Poverty Index, even as inequality supposedly increased.

But did it really increase? Tax rates rose after Independence. Indira Gandhi raised the top rate to 97.75%. Businesspeople kept their earnings in tax shelters (such as tax-exempt trusts), or as plain black money, to survive. Companies avoided declaring high dividends, as in the US. The apparent fall in inequality was largely a mirage – the rich mostly found ways to cloak their income.

Later, the top tax rate was cut to 50% in 1985, 40% by Manmohan Singh, and to 30% by P Chidambaram in his 1997 budget. Naturally, the rich came out of tax shelters, reduced resort to black money, and declared higher dividends in response to much lower tax rates, just as Auten and Splinter showed it to happen in the US. This appeared to create a surge in the incomes of the Indian rich and, hence, inequality. The mirage of greater equality in the 1970s was buttressed by the mirage of growing inequality thereafter.

A sociological issue raised by Auten and Splinter is that the US marriage rate after the 1960s slumped among poor people, especially Blacks. It fell much less among the rich. So, incomes of the poorer half were split among a much larger number of households than of the rich. This created the illusion of a greater income share for the rich. This notion has been exposed as false by Auten and Splinter after making appropriate adjustments.

In India, Piketty used I-T data for the rich, but for the poor, he used a mix of survey data from National Sample Survey Office (NSSO) and Indian Human Development Surveys (IHDSs). This exaggerated the income share of the rich. India levies I-T on individuals, so the top taxpayers are mainly the richest males. But NSSO surveys and IHDSs estimate the income/consumption of entire households and divide that by the number of persons per household, to produce a per-capita estimate. This reduces the income share of each poor person. Yet, Piketty makes no adjustment for this.

Piketty and his collaborators typically make inequality estimates without adjusting for taxes paid by the rich and welfare benefits going to the poor. This seriously exaggerates inequality. Auten and Splinter have adjusted for both these factors, plus other technical adjustments. These adjustments make the difference between black and white.

Their new calculations show that the pre-tax share of the rich increased by only 2.6 percentage points between 1960 and 2019, not remotely close to the 19% claimed by Piketty. Adjusting for tax payments and transfers to the poor, the income of the top 1% was the same in the early 1960s as it is today – inequality has not increased at all.

In India, too, Piketty’s estimates should not be taken seriously. Estimates of some Indian economists suggest that inequality has not worsened here either. Shed no tears.

This article was originally published by The Economic Times on January 4, 2023.

Leave a Comment

Your email address will not be published. Required fields are marked *