A Brief History of Wealth Inequality

So it’s time for this blog to jump on the Thomas Piketty bandwagon, like everyone else. At the heart of his improbably bestselling book is the relationship between r, the return on assets (gold, shares, land etc), and g, the rate of economic growth. The idea is that if returns on assets accumulated in the past (and wealthier families would have more such assets) are higher than the rate at which current incomes are growing, than the inequality of wealth will increase. Those with more assets accumulate even more of it at a rate faster than that at which current incomes are growing. It’s a simple idea though not without its critics.

Piketty’s book tracks the changing inequalities in wealth and incomes across a couple of hundred years for a number of countries.  He makes the point that r has almost always exceeded g except for a period in the mid-20th century when it was the other way round – the result being a time when inequalities narrowed. But that phase has ended and we are once again living in a new gilded age.

What about India? As opposed to wealth, Piketty analyses income inequality using income tax returns for India. Till the 1980s or so, the share of the top income earners in society, in total income national fell substantially, before rising again in the era of reform. But what about wealth inequality?

Data on wealth in India is difficult to compile, and its pattern and distribution across time even more so. So I thought I would start from the other end by looking at returns on different classes of assets over time and compare that with GDP growth. The shifts in those returns over time is an indirect indicator of how wealth inequality has shifted over the last sixty years, assuming the Piketty relationship between r and g.

Even this way of doing it is hardly problem-free. The big issue is cataloguing the list of relevant assets. I went with the most obvious ones – savings deposits, equity shares and gold. The giant gap here is of course, land. There is simply no good indicator of land prices or rentals over time that can be used. To the extent that this data excludes land prices, it is flawed, so make of it what you will.  All returns and growth rates in this chart are adjusted for inflation and are five-year moving averages.

(Hover your mouse anywhere in the chart to see actual values for that year. For those points at which the data shows zero and the line is not visible, assume that the data is not available. Data for gold and fixed deposits only available from the 70s)

Now of course, the returns on these different assets move in different ways over time, so it’s difficult to make a generalization that holds right across. But to the extent that we can do so, look at the period beginning in the 1980s.

First, the returns on equity became far more volatile, but at no time did they fall below the growth of GDP. Second, the 80s and 90s was the period when the great ‘bear’ phase in the gold market took place, driving down ‘returns’ on the staple investment of many Indian households, well below zero. It was also at this time, the GDP growth started to inch up, as did real returns on fixed deposits, after a prolonged phase in the 70s, when returns on it were negative. And while we don’t have data on land, it’s not a bad idea to assume that the 80s was a time when land prices really started to rise steeply, especially in large cities.

The net effect depends on how households across the distribution held their wealth. If very rich households held a larger chunk of their assets in the form of land or equity than households a bit lower down, than wealth inequalities would have widened even more than they did in earlier decades. The 80s were a period of flux in the wealth distribution but I suspect that flux was limited to the higher end. Remember, even at this time, the vast majority of households in India would have had net wealth that was either zero or negative.

And after the 2000s, wealth inequalities would have widened further, even though GDP growth expanded at a faster rate. This was simply because returns to gold grew much faster, and returns to equity rose much more sharply. India has not (yet) gone through the phase that many developed economies went through, when r fell below g. Going by historical experience, that is not going to happen till we are a much more ‘developed’ economy.* Or there is a massive ‘shock’ to the economy (e.g. war) which wipes out the value of capital.


All data is from the Reserve Bank of India. Returns on equity are compiled from an RBI dataset of the financials of public limited companies. Gold prices and fixed deposit rates (3-5 year term) are from the RBIs statistical handbook. All prices and yields were adjusted for inflation using the GDP deflator calculated from RBI data. All data are five-year centred moving averages.

Chart made using d3js. Chart tooltip code adapted from this example.

*By ‘developed’, I dont just mean everyone is richer. It’s also a qualitative shift in, among other things, the ability of governments to extract resources, and revenue in the form of taxes. In that sense, the Indian state is actually not that ‘developed’.

December 21, 2014

Comments are closed.