Lecture 10: The growth of capital income.


In this lecture we bring to an end our analysis of the capital/income ratio and turn our attention to the division of national income between labor and capital income.

Let’s remind ourselves of some important concepts:

  • Income can be broken down into two sources: income from labor (wages, salaries, bonuses) and income from capital (rent, dividend, interest, profit).
  • Income from capital is any income earned from the mere fact of owning capital, regardless of whether it is land, a government bond, a stock, a firm or real estate.

All capital assets will yield an income, although at very different rates. If you retire and own $1 million in assets (government bonds and real estate) and the rate of return on these assets is 3%, you will earn a capital income of $30,000.

Capital and income are very different things although they are related.

If you graduate and have a debt of $80,000 (US tuition fees) you will have negative wealth but you might get a job earning $85,000. So you will have a good flow of labour income. But from this you will have to use a significant percentage to pay off your student loan.

For Piketty the best way to analyze the importance of capital/wealth in society is to measure the amount of capital/wealth as it relates to income. This is what he calls the capital-income ratio (β). On Monday we concluded that the long-run capital/income ratio (β) depends on the savings rate (s) and the economic growth rate (g).

Rate of return on capital income

Piketty is arguing that when the rate of return on capital exceeds the rate of return of economic growth, wealth inequality grows (captured by increased β).

But what determines the rate of return on capital?

Different types of capital-assets yield different rates of return. The yield on the riskiest assets (for example, shares in an emergent high-tech firm) might yield 6-7+%. Real estate can be as low as 3-4%, whereas your savings accounts can yield less than 0.1%.

Figures 6.3 and 6.4 illustrate the pure annual rate of return on capital from 1770-2010 in France and Britain. You’ll see that the long-term average is between 4-5 percent.

Pre-tax returns!

There are three important observations to remember about this long-run trend on the pure rate of return to capital (which went from 6-7% in the 19th century to 4-5% today).

  1. They are pre-tax returns. The average tax rate on capital-income, in most rich countries, is around 30 percent. Companies adopt a whole variety of tax strategies to reduce this.
  2. They hide enormous disparities between capital-assets. The average pure rate of return is found by dividing the total capital stock by the annual flow of income from capital. But not all capital-assets are the same. Savings accounts are a form of ‘wealth’ but they yield very little. If the interest rate today is 0.05% it make more sense to spend.
  3. These are ‘real’ rates of return and do not deduct inflation. Nominal assets such as buying a government bond or your savings account (non-indexed) are subject to real inflation risk. This is not the case with real estate for example, given that house and rental prices generally rise with (or faster) than the consumer price index (CPI).

What factors lead to changes in the rate of return to capital?

In general we can say that there are two driving forces:

  1. Technology (what is capital used for?)
  2. Capital stock (too much capital kills the return on capital)

The capital stock

The stock of national capital in a nation-state usually fulfills two dominant functions: it provides housing services (the rate of return can be measured by the rental-value of dwellings) and it serves as a factor of production in producing goods and services (tools, machinery, equipment, land, computers etc).

If you have a farm and you want to produce food you need land, tools, machinery, workers. Combined they produce your product. The logic is the same if you are producing an iPhone.

In theory, the purpose of the financial and banking system is to find the best possible use of capital (national savings), such that each unit of capital is invested where it is most productive (or where it is most profitable).

This is the ideal of a perfectly efficient capital market.

In reality we know that financial capital markets are far from ideal (and it’s bizarre that they still teach it as such). They are ridden with waves of speculation, instability and bubbles. Most banks don’t lend for investment in business; rather than lend to buy and sell existing assets (such as housing).

In the Anglo-Saxon world most high risk capital investment is carried out by venture capital funds, not banks.

Technology and innovation 

The second driving force is that too much capital kills the return on capital. What does this mean?

It can be captured by thinking about the value of land and land rents in the USA during the 19th century. There was so much land that it did not yield much of a return. It was cheap and plentiful. There was simply too much landed capital to make a return. The same would apply to housing.

Why then does an increase in β (capital/income ratio) not lead to a decrease in R (rate of return on capital)?

Everything depends on technology and the extent to which society can substitute capital for labour, and labour for capital in the production process. There are always new ways to find new uses for capital (i.e. solar panels as a replacement for oil). This is called innovation.

But profits, rents and capital-income accumulate (and it is accumulating fast). Therefore, for Piketty, the volume effect outweighs the price effect.

The labour-capital split

This brings us to the division of national income between labour income and capital income

Figures 6.1 and 6.2 present the long-term trend in Britain and France.

Capital income absorbed around 35-40 percent of national income in the late 19th century before falling to 20-25 percent in the mid-20th century, and then increasing to 25-30 percent in the 21st century.

Note: this corresponds to an average rate of return to capital of 5-6 percent.

In basic terms, if you invest $1 million and earn $50,000 per annum, your rate of return is 5%. How is all this calculated?

Remember Pikettys first law of capitalism α = r X β. This is an accounting identity to determine the ratio of capital income in overall national income.

  • a = capital income,
  • r = rate of return on capital
  • β = capital/income ratio.

The percent of capital income in national income can be found if we multiply the rate of return on capital (r =5)  by the capital/income (β = 6). The outcome is 30. What remains is labour income, which equals 70 percent.

Figure 6.5 shows the increase in the share of capital income in national income from 1970. This has increased even faster since 2010. Why?

Because wages/salaries have been declining whilst the rate of return on capital-assets has been increasing. As the pie of national income grows, the gains are absorbed by capital-asset holders.

The stability of the capital-labour split

The stability of the 70-30 split, however, doesn’t really tell us anything about how capital or labour income is distributed. Very few people earn an income from owning capital assets because capital-wealth is significantly more concentrated than labour.

As we will see next week, the top 10 percent of the US population own 70 percent of wealth (capital), whereas the top 1 percent own 52 percent of wealth (capital). They control a large part of economic resources in society.

It is this 1 percent of the population that tends to gain when the share of capital income in national income grows. Hence, if the shape of capital income in national output increases, it’s likely to only benefit a very small percentage of the population (those who own financial assets).

Hence capital income in overall national income is concentrated in a much smaller segment of the population.

The stability of the labour-capital income split is usually attributed to the increased importance of human skills, with the implication that many economists think that capital doesn’t really matter anymore. It is all about human capital (i.e. high-tech skills). It’s not about capital rents.

Explaining the rise in capital income

The increase in capital income (in national income ) that we observe since 1970 is related to shifts in economic bargaining power between societal interest groups and changes to capital-tax laws.

It is more about politics than markets. This is something we will discuss in later weeks.

Further, the stability of capital’s share in national income in no way implies a stability in the capital-income ratio. This is what we have been analyzing over the past two weeks.

The big contribution of Piketty is to shift our attention toward the evolution of the capital-income ratio rather than the capital-labour split. Why is this shift so important? It allows us to assess the overall structural influence of wealth in a society.


Looking to the future: a capital/income ratio equal to 7-8 years of national income and a rate of return to capital of 4-5 percent means that capital’s share of global income could grow to 30-40 percent by 2050.

Free-market optimists suggest that technological change will favor a shift to labour-income (human capital and work) rather than capital-income (to the owners of assets/rent/interest). This is certainly plausible. Just think about Silicon Valley.

Piketty suggests otherwise.

If the rate of return on capital (5 percent) grows faster than economic growth (2 percent) then wealth from the past will accumulate in importance. Inheritance will matter more than entrepreneurship and work.

Ultimately, what Piketty is saying is that progress toward economic and technological rationality (more efficient and more competitive markets) does not imply progress toward democratic and meritocratic rationality. There is a clash between capitalist-markets and social-democratic rights.

Think back to week 2, advances in productivity and technological growth, and the rise of a middle class, have meant that we have avoided the Marxian apocalypse.  But has technology changed the deep structural influence of capital over society? Economic resources continue to be unequally distributed. How does this concentration of power and wealth impact upon politics?

From next week we analyze the distribution of income and wealth at the individual and household level.

The slides can be found here. Lecture 10


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s