Resurgence of Capital or rise of the Working Rich? On Piketty’s Capital in the 21st century

by Olivier Godechot on December 16, 2015

*Capital in the Twenty-First Century* was a classic as soon
as it was published.[^21] It deserves a place on bookshelves beside its
illustrious namesake of the 19th century. Capital, in
*Capital*, is the wealth of nations. It extends beyond
firms’ traditional productive capital to encompass the entire public and
private patrimony that can be sold on a market (thus excluding
non-transferable forms of capital such as human, cultural and social
capital). The book is the culmination of fifteen years of individual and
collective research on the evolution of income and wealth inequalities.
Thanks to data based on the collection of income tax, Thomas Piketty and
his colleagues had already widely explored income inequality in France,
the United-States, India, China, and more generally in the world by the
early 2000s, fuelling a unique and remarkable dataset: the *World
Top Incomes Database.* However, this work focused on income
rather than wealth, and hence provided an incomplete account of economic
inequality. This new book fills in this gap in a very timely fashion.

The thesis of the book could be summarized as follows: the simple fact
that the return on capital is durably higher than the economic growth
rate feeds implacable dynamics of inequality.

Under these conditions, the relative share of capital (as measured by
the national wealth to GDP ratio) increases inexorably. Far from being
exceptional, this configuration is currently the norm. Since the 1970s,
the decline in the growth rate which is a combination of the decline in
the population’s growth rate and that in individual productivity, have
helped capital come back. In France, the amount of total wealth owned by
all public and private actors has gone from a little less than four
years of income in the 1970s to more than six in the 2010s.

This resurgence of capital promotes increased inequality in several
ways. First, wealth is very unevenly distributed, with the poorer half
of the population holding less than 5% and the top 1% holding 25%. The
global increase in asset prices thus tends to help the latter. Second,
ability to save, and hence to accumulate wealth, increases with income.
Third, the higher the wealth, the greater the return on capital, thanks
to better financial advice, a weaker preference for liquidity and a
greater ability to bet on more remunerative (albeit risky) long term
investments. Meanwhile, the increase in wage inequality, due to the
emergence of “super-wages” for an elite of CEOs and finance
professionals, enables newcomers to find a position among the wealth
elite without weakening this elite’s domination.

After reaching a low point in 1970, wealth inequality is on the rise
again. The share of national wealth owned by the top 1% increased from
28% in 1979 to 34% in 2010 in the United States, from 23% to 28% in the
United-Kingdom and from 22% to 24% in France. It is still a long way
away from the concentration of wealth in the early 20th century where
the top 1% had 60% of national wealth in France and 70% in the UK.
However, *Capital* warns us that we may be on our way back
to that world if the dynamic of inequality is not halted. According to
Piketty, progressive taxation of income and capital (on the level of
global regions to limit the problem of tax competition) could set the
net return on capital below the economic growth rate and halt the
inequality dynamics.

But the book goes far beyond a simple demonstration of the unequal logic
of capitalism and the merits of redistributive taxation. It is a
remarkable multidisciplinary work on the economic and social history of
Western capitalism and inequalities, thanks to a combination of serial
statistics and specific examples from literature, film, political
history and history of economic thought that illuminates both past
events and perceptions. Wealth is the gateway to a new understanding of
two centuries of economic, social and political history. The opposition
in the nineteenth century between the Old World, dominated by past
wealth (with a capital equivalent to seven years of income) and the New
World, less subject to such a legacy (with a capital of less than five
years of income) is particularly striking. So is in the opposition in
the New World between the North of the United States and the South,
where slave ownership gave capital a preponderance similar to that
measured in Europe at the same period. The shocks of economic crises and
inflation peaks, the impact of fiscal policies, sometimes offensive,
sometimes accommodating, and moreover the destruction of wars greatly
shape the importance and the role of capital.

The book sometimes adopts a Braudelian tone, highlighting deep
structures of capitalism, such as the growth rate, that are unbudging
and impossible to guide. The decades of growth in the postwar era is an
exceptional stage that is futile to feel nostalgic for and try to return
to, since it was largely a phase of reconstruction and catch-up. After
reaching the technological frontier, growth can only continue in the
21st century at its long-term rate of 1 to 1.5% per year (of which half
is due to population growth). It seems slow when measured year by year,
but it is very fast and almost unbearable at the scale of centuries. In
addition to the growth rate, other factors such as the savings rate,
share of value added, and return on capital are also immobile or slowly
changing variables that seem to defeat any kind of political action. If
these quantities, which are the product of a kind of extended general
equilibrium, are not controllable, states still retain their two
original powers to shape capital accumulation and its unequal
consequences: wars (which one cannot want) and taxes.

Although the book contains many impressive and valuable statistical
series, it takes some risks in its treatment of the largely unexplored
topic of wealth. These first estimates will benefit from being
evaluated, tested, and corroborated by later work. For example, the
return on capital, which the author believes to be stable at around 5%,
is calculated from aggregate data from national accounts and defined as
the ratio between the share of capital in the value added from the
income accounts and the estimated value of the national capital in the
wealth accounts. Given the numerous conventions that govern the
construction of these data, the reconciliation of the macro approach
with a micro approach might add a lot, all the more so as the
performance of listed shares varies considerably even in the
medium-term. The ten-year yield (including capital gains) of the S&P500
(the index of the New York Stock Exchange) is greater than 10% between
1940 and 1960 and between 1980 and 1992, while it is negative or zero
between 1965 and 1975 or since 2000 (Figure 1). These significant
variations in performance at the micro level leads one to ask whether
the resurgence of capital is due to the inexorable mechanics of
capitalism or rather a product of contingent financial and real estate
bubbles, and more generally due to the phenomenon of financialization.
Answering this question could tell us whether, in addition to fiscal
policy, a sector-based economic policy of “definancialization” could
also contribute to fight inequality.

Screen Shot 2015-12-14 at 10.18.44 PM

Figure 1. Comparison of the return on capital over ten years calculated
by Thomas Piketty and that on some US famous assets. Note: A portfolio
replicating the S&P500 purchased in 1949 and sold in 1958 provides a
real annual return of 18% per year (capital gains included).We discount
changes due to inflation. Sources: Thomas Piketty’s data is available
those on yields of major US national assets

While the early work of Thomas Piketty and his colleagues focused mostly
on income inequality due to the emergence of a new elite, the Working
Rich, this new work focuses on wealth and offers a very different
diagnostic by showing the resurgence of capital in its most traditional
form: inheritance. The various forms of inequality are morally and
politically ranked by the frequent use of the concept of merit, a
concept difficult to define – and not specified here – sometimes used to
describe the collective perception (an emic conception), sometimes that of the
analyst (an etic conception). According to the author, wealth inequalities
(particularly those that are inherited) violate merit more than income
inequality, which contravene merit more than wage inequality.

Although the resurgence of capital is a proper cause of concern,
especially for political and moral reasons, should we consider it as the
most prominent transformation of contemporary inequalities? Certainly,
wealth inequalities are more pronounced than income inequality. But in
light of the data presented, their recent growth in the US is much more
limited than that of wages. In 2010, the share of the top 1% in wealth
is 1.3 times larger (in terms of odds ratios) than it was in 1970. Over
the same period, the share of the top 1% in wages increased by a factor
of 2.3. Measured by this yardstick, the emergence of the Working Rich is
a more radical transformation than the resurgence of capital. However,
it seems that the author, rigorously reasoning his way through
incomplete data, has had the remarkable intuition of a phenomenon that
the available data only showed imperfectly.

Hence, in a recent work, Gabriel Zucman and Emmanuel Saez (2014)
reestimate wealth composition from income tax. This method requires one
to make many assumptions to estimate stocks from flows and we must
remain careful. Nevertheless, they establish remarkable results that
fill in the missing piece of Thomas Piketty’s book. According to these
new estimates, the top 1%’s share of wealth has increased 1.5 times
between 1970 and 2010, moving from 28% to 37% of total US wealth and
that of the top 0.1% increased by 2.3 between 1970 and 2010, moving from
10% to 20%. If Emmanuel Saez and Gabriel Zucman’s work holds true, then
there is indeed a real resurgence of wealth inequality, moving back to
its 1920 level. Nevertheless this is smaller in relative amplitude than
the rise of the Working Rich: the share of the top 0.1% of wages
increased during the same period by 3.9, moving from 1.1% of payroll to
4.1% (Figure 2). It is true that the interpretation of the comparison of
two evolutions of percentage depends dramatically on the choice of the
metric: additive (+10 percentage points for wealth versus +3 percentage
points for wages) or multiplicative (as I do here: \*2.3 for wealth
versus \*3.9 for wages). However, whatever the metrics, the complexity
of recent changes in inequality is due to the necessity to think them as
both the product of the return of capital and the rise of the Working

Screen Shot 2015-12-14 at 10.19.38 PM
Figure 2. Comparison of the evolution of wealth inequality and wage
inequality in the United States

Note: The top 0.1 wage share amounted to 1.1% of payroll in 1970.
Sources: for
[salaries](; for

[^21]: An earlier version of this was published in French as Godechot
Olivier, 2015, “Le capital au XXIe siècle, T. Piketty.” Le Seuil,
Paris (2013). 970 pp., *Sociologie du travail*, vol. 57, 2, p.



Rakesh Bhandari 12.16.15 at 3:41 pm

On supermanager income, it is however, in the nature of things, for it to soon become the compounding wealth of a rentier and for those rents to become greater than income.

On the S&P Chart, I could not figure out what the olive line is representing. I am not sure what the green line is either.

Don’t know how the portfolios of the wealthy have changed. We would have to chart corporate bonds too. Weren’t those a bigger part of portfolios in the 60s? Did the corporate bonds that the wealthy tended to own compensate for weak returns in the S&P 500? And even in terms of stock indices were the wealthy able to beat the averages, especially when low?

Also even if r =g, the wealthy could get wealthier due to higher-than-average returns. Again we could look at the University endowment data.

In terms of stock performance in the decades after WWII, Piketty also does say that it took some time up after the depression of capital values from the War for the catch up effect to really take hold.


Peter T 12.16.15 at 8:43 pm

CEO compensation may be formally a wage. In practice, it’s just the currently-preferred form of income extraction from the common pool. Earlier centuries (or other places) would not have found it remarkable – elite incomes in, eg, the C16/C17 were often a mix of rents, state salaries and state-granted monopolies.


Cleric 12.16.15 at 10:03 pm

I’m curious:

“This method requires one to make many assumptions to estimate stocks from flows and we must remain careful.”

What might those assumptions be, and what must we remain careful about before jumping into the conclusion that…

“they establish remarkable results that fill in the missing piece of Thomas Piketty’s book”?


T 12.17.15 at 12:19 am

@2 @3
Peter T — Also, individuals would attempt to shift compensation from higher taxed categories to lower taxed categories, i.e. wages to cap gains or from short-term gains to long-term gains. And wealth measurements by individuals are tricky because cap gains are reported on sales of assets, not on held assets (the flows vs stock issue). Several recent studies have shown a significant share of income and wealth is effectively hidden. But this is the type of work that must be done to get the best estimates.

Comments on this entry are closed.