[This is the third of five posts on Capital in the 21st Century. The first is here.]

The second part of Piketty’s Capital in the 21st Century looks at inequality within developed countries. Two main types of inequality are considered: income inequality (inequality in how much people earn each year) and wealth inequality (inequality in how much people own).

For income inequality two main trends are evident, the rise of the top 10 percent (top decile), and the rise of the top one percent (top centile) or even 0.1 percent. The increasing share of the top decile is likely a reflection of changes in education, technology (computers and the internet), and globalisation; it is ‘bigger’ in the sense of accounting for a larger share of national income. The increasing share of the top centile is often described as the rise of the supermanagers (high-aid CEOs), but this is not enough, as the top 0.1 percent have risen even further and their income comes mostly from capital income. It is this rise of the top centile (and 0.1 percent) that Piketty considers most alarming.

Wealth Inequality fell substantially during the first half of the 20th century, stayed largely flat for a few decades, and has been increasing for the past few decades. The fall is explained by the two world wars, together with the period of high inflation that many countries experienced inbetween. Piketty argues that the flatlining of the next few decades occoured because the (after-tax) return on capital was lower than the economic growth rate, thus capital did not simply tend to accumulate when ‘left to itself’. This period of the (after-tax) return on capital being lower than the economic growth rate appears to be a historical anomaly, and has ceased to be the case in recent decades. As a result wealth inequality has begun to increase back towards its historical levels. There is one other major difference between post-war wealth inequality and historical wealth inequality, the part of the population not in the top decile actually holds a non-trivial fraction of the wealth (ie. many households own their houses).

 Income and Wealth Inequality:

• The income rich and the wealth rich are not the same people. There is some overlap, but generally the people who make the top incomes and the people with the top wealth are not the same people. Retired people, with lots of wealth but little income, are one example of this.
• Wealth is always and everywhere much more unequally distributed than income.
• For most people income consists of wages, capital income is the most important form of income only for the very rich. Around 1930 capital income was the main source of income for the top 1% incomes, nowadays capital income is the main source of income for the top 0.1% incomes (US & France: F8.3, F8.4, F8.9, F8.10).
• What does ‘low’ or ‘high’ inequality mean when talking about countries? Piketty gives some examples of countries that show us what such concepts might mean, for example Scandinavia in the 1970s provides a benchmark of low inequality, the US in 2010 a benchmark for high inequality. The following tables show what the shares of various groups are for labour income, capital ownership, and total income are for these cases, respectively: T7.1, T7.2, T7.3.

Income Inequality

• There are two main trends currently occurring. One is that the top quintile (20%) or decile (10%) are leaving the average and poor behind. The second is that the top 1% or even 0.1% are leaving everyone else even further behind.
• Income share of top decile in US and Europe: F9.7, F9.8.
• Income share of the top 1% in various countries: F9.2, F9.3, F9.4.
• Income share of the top 0.1% in various countries: F9.5, F9.6.
• That the top quintile or decile are leaving behind the median and poor seems mostly due to a combination of factors such as an increase demand for high-skill workers (due to technology such as telecommunications and computers that allow them to be more productive and reach larger audiences), a slowing supply of high-skill workers (since education levels appear to be levelling out; almost everyone already has high-school education, many people now have university education), and globalisation (which massively increased the supply of workers, and hence the competition in the non-university educated jobs).
• The reasons why the top 1% of incomes have seen such a large increase in their share of earnings is not so clear. Is this a return to a more normal state of affairs?, as suggested by the observation that pre-WW2 they were much higher than now. Or is it due to their being more productive (thanks to technological improvements) like for the top decile, only even more so. Or is it due to their capturing the political process and manipulating the rules in their favour? Or is it because CEO pay is set by friends of CEOs and so they simply scratch each others backs? While there is some evidence for each of these arguments we do not yet know their relative importance. What we do know is just that the top 1% and 0.1% have been making an increasingly large share of income over the last decades, that this is true in all developed counties, and that it seems likely to continue. Interestingly their share is much higher in the US than in the other Anglo-Saxon countries (F9.2, F9.5).
• Who are the top 1% of incomes in US? By vast majority CEOs, Managers, Finance, plus some Doctors and Lawyers. Who are not the top 1% of incomes in US? Film stars, Sports stars, and Small-Business Owners. Piketty mentions this point. The numbers can be found in Bakija, Cole, and Heim (2012WP).
• Historically income inequality fell substantially around the period of the two world wars, and has been increasing since. A substantial part of this is due to the simultaneous fall and then rise in the importance of capital (F8.1, F8.2, F8.7, F8.8), and fall and rise is much greater within the top 1% than in the rest of the top decile (F8.6). One interpretation of this is that the post-war period was the exception, and presently rising inequality represent a return to the norm, this is the argument made by Piketty.

Wealth Inequality

• The top decile and top 1% wealthiest hold very large shares of total wealth (F10.1, F10.3, F10.4, F10.5).
• The bottom third (approximately) of the population holds around zero wealth.
• The shares of wealthiest used to be much higher before the world wars, and have been increasing during the last few decades. Piketty argues this present increase likely represents a return to normal.
• Part of his reason for arguing that increasing wealth inequality is likely to continue is that capital income is much larger than savings (F10.8) and so most wealth accumulation likely occurs from the reinvestment of capital income, tending to lead to higher concentrations of capital. This is made in combination with the observation that the return on capital (\(r\)) is higher than the growth rate of the economy (\(g\)), a so simply by saving a fraction of the ‘extra \(r-g\)’ part of capital income there will be a tendency to increased concentration of capital ownership (F10.10, Piketty provides a few other graphs arguing that \(r>g\) is very much the historical norm, and that the only exception has been the couple of decades post-WW2) .
• This argument of Piketty’s, that \(r>g\) will lead to increasing wealth inequality, has been one of the most strongly criticised parts of the book. Disagreement is not so much that \(r>g\) is the normal state of affairs — which seems to have been widely accepted — but his conclusion that this naturally leads wealth inequality to increase involves some strong assumptions about how people decide how much to save. However Piketty’s assertion does not appear strongly at odds with the evidence, that generally suggests that the rich have higher savings rates. That Piketty does not present any evidence on what determines savings is though a major weak point in his arguments.
• The shares of the wealthiest are higher in Europe than in US, this is likely explained by the higher population growth in the US which makes wealth accumulated in the past become more dispersed (this logic is reasonable and related to the case of capital/income ratios, but not watertight). (F10.6)
• Superentrepreneurs, eg. Bill Gates, are very few in number and as such play no quantitative role in the increases in inequality of recent decades.

For more numbers on income and wealth inequality for a wider selection of countries, see the Chartbook of Economic Inequality.

Part 3: Inheritance of Capital Wealth