Unpicking Piketty

Thomas Piketty Capital in the 21st century Harvard 2014, pp677, £29.95

Publishing 'sensation'

Once upon a time, there was an obscure French mainstream economist who collaborated with others (Emmanuel Saez, Anthony Atkinson and Gabriel Zucman) on various studies of inequality of income and wealth in modern economies like the US. After some years of research, this Frenchman developed a theory that the inequality of wealth in capitalist economies tends to increase to the point where it could cause major social instability. This tendency is the “central contradiction of capitalism”.

That economist is Thomas Piketty. When he published his magnum opus of 677 pages in France last October, it was greeted pretty much with silence and even cynicism, apart from a few French economists. But then it was translated into English and published in America. Everything changed. It became not just the best-selling economics book, but the top non-fiction book of the year, ahead of cook books by famous chefs and travel books by celebrities.

There has been a profusion of reviews, debates and interviews with the man of the moment. The book has been greeted rapturously by such as Branko Milanović, the expert on the inequality of wealth in the world, who called it “one of the watershed books in economic thinking”,1 and by the guru of liberal Keynesian economics, Paul Krugman, who, writing in the New York Review of Books, said it was “truly superb”.2 Martin Wolf of the Financial Times called it “extraordinarily important” and “awesome”.3 John Cassidy, in the New Yorker, said: “Piketty has written a book that nobody interested in a defining issue of our era can afford to ignore.”4

The title is a clear allusion to Karl Marx’s Capital, published in 1867. Piketty seems to suggest that he is updating (and indeed correcting) Marx’s analysis of 19th century capitalism for the 21st century. But Piketty is no Marxist.

He was brought up in Clichy, a mainly working class district of Paris. His parents were both militant members of Lutte Ouvrière (Workers’ Struggle) - a Trotskyist party, which still has a significant following in France. On a trip with a close friend to Romania in early 1990, after the collapse of the Soviet empire, he had a revelation: “This sort of vaccinated me for life against lazy, anti-capitalist rhetoric, because when you see these empty shops, you see these people queuing for nothing in the street,” he said, “it became clear to me that we need private property and market institutions, not just for economic efficiency, but for personal freedom.”5 Piketty rejected what he saw as Marxism and opted for social reform. Indeed, he was an adviser to the Blairite, Ségolène Royal, when she was the Socialist Party candidate in the 2007 presidential elections.

Why ‘Marx is wrong’

According to Piketty, Marx needs correcting because, despite his clever intuition that “private capital accumulation could lead to the concentration of wealth in ever fewer hands” (p1), he got the whole mechanism for this development totally wrong. Marx thought that capitalism would have an “apocalyptic” end, but, thanks to “modern economic growth and the diffusion of knowledge”, that has been avoided. But there is still the problem of the “deep structures of capital inequality”.

Piketty goes on to inform us that the basis of Marx’s prediction of an apocalyptic end to capitalism was “either the rate of return on capital would steadily diminish (thereby killing the engine of accumulation and leading to violent conflict among capitalists) or capital’s share of national income would increase indefinitely until the workers went into revolt” (p9).

Marx reckoned that wages would be stagnant or falling. This was wrong, because “like his predecessors, Marx totally neglected the possibility of durable technological progress and steadily increasing productivity, which is a force that can to some extent serve as a counterweight to the process of accumulation and concentration of capital” (p10). Unfortunately, you see, Marx failed to use the stats available in the 19th century and “devoted little thought” to how a non-capitalist society might work. If he had done so, he might have sorted out his mistakes.

Already, it will be clear to a student of Marx’s analysis of a capitalist economy that Piketty is unaware that Marx saw the drive to raise the productivity of labour through technological advance as the flipside of the accumulation of capital. Instead, Piketty accepts the distortion by mainstream economics that Marx’s theory is based on an ‘iron law of wages’ and a zero rise in productivity: “Marx’s theory implicitly relies on a strict assumption of zero productivity growth over the long run” (p27).

It is not surprising that Piketty can write in such a way when we learn that he admits he has never read the very book that carries the same title as his own: “I never managed really to read it. I mean, I don’t know if you’ve tried to read it. Have you tried?… The Communist manifesto of 1848 is a short and strong piece. Das Kapital, I think, is very difficult to read and for me it was not very influential ... The big difference is that my book is a book about the history of capital. In the books of Marx there’s no data.”6

Again, the view that Marx’s Capital contains no data to back up his theory of the law of value and exploitation and the laws of motion of capitalism shows Piketty’s ignorance of the work whose name he has adopted for his own book.

The data attack

Piketty’s book is bursting with data - and, in my view, this is all to the good. Its merit is that it compiles evidence and tries to develop a theory and laws from there. For example, he says: “All social scientists and all citizens must take a serious interest in money, its measurement, the facts surrounding it and its history. Those who have a lot of it never fail to defend their interest. Refusing to deal with numbers rarely serves the interest of the least well-off” (p577).

However, compiling lots of data can lead to errors of measurement, difficulties in interpretation and bias in analysis. And this is exactly where recent criticism of Piketty’s book has concentrated. The FT’s economics editor, Chris Giles, has gone through the wealth of data used. He found that Piketty had made simple mistakes in transcribing some of it. He also claimed that the author had made “arbitrary” changes in some of his estimated data without explanation. Piketty “cherry-picked” his sources, using different measures in different countries at different times. Giles made new calculations with other data sources and found that there is no “obvious upward trend” in inequality of wealth in Europe.

Piketty has vigorously defended his work from Giles’ critique and I have sympathy with him. Data are always inadequate and often inconsistent and it is also easy to make simple mistakes. But it is better to try and provide evidence and, above all, release sources and your workings for all your data, so that others can check and - even better - try and replicate your results. That is the scientific method. As Piketty says in his reply to Giles, at least he has put all his data and workings online for people to consider.7 And that is more than we can say about the bulk of mainstream economics, which either offer no evidence to back up theoretical claims or fail to provide any workings, or both. He has been more transparent than most with his evidence. Piketty also argues that more recent work on inequality of wealth by his colleagues, Saez and Zucman, using different measurement methods, “confirm and reinforce my findings”.8 So he reckons that any mistakes or biases in his own data “will not have much of an impact on the general findings”.9

Capital and wealth

In my view, there are more important deficiencies in Piketty’s work than inconsistencies in the data. For one, there is the key difference between wealth and capital that he ignores. For Piketty, “Capital is defined as the sum total of non-human assets that can be owned and exchanged on some market. Capital includes all forms of real property (including residential real estate) as well as financial and professional capital (plants, infrastructure, machinery, patents and so on) used by firms and government agencies” (p46). In effect, for Piketty, capital and wealth (mainly personal wealth) are the same: “To simplify the test, I use the words ‘capital’ and ‘wealth’ interchangeably, as if they were perfectly synonymous” (p47).

This is clearly different from capital as defined by Marx. For him, capital is a social relation specific to the capitalist mode of production. Under the capitalist mode of production, things and services that people need are produced simply as a money-making exercise, but this money comes from value created by the exertion of labour-power, with the surplus over and above the living needs of labour appropriated by the owners of capital. Thus the circuit of capital, for Marx, is M-C…P…C1 to M1: that is, capitalists have money capital (M), which is invested in commodities (C), means of production and raw materials, which are used by labour in production (P) to produce commodities (C1) for sale on the market for more money (M1). Capital (M) expands value to accumulate more capital (M1). But only labour creates that new value.

For Piketty, this process of exploitation of labour and its social relations are ignored. Capital is wealth and wealth is capital. But wealth existed before the capitalist mode of production became dominant in the world and is not specific to capitalism. Indeed, wealth is really a measure of accumulated assets, tangible and financial. So for Piketty the capital process is M…M1. Money accumulates more money (or wealth). It does not matter how and so there is no need to define capital as different from wealth.

This is what Marx called “vulgar economics”: ie, failing to see the underlying process of accumulation and just observing the appearance - indeed seeing things from the view of the holder of wealth alone. In the book, Piketty refers us to the novels of Jane Austen and Honoré de Balzac, where all the characters who are holders of wealth live off the income from it (p53). All they were interested in was the return on that wealth, not how it was generated (whether by slaves, wage labour, land rents or interest on government debt).

Piketty specifically rules out the approach of the classical economists and Marx: “Some definitions of capital hold that the term should apply only to those components of wealth directly employed in the production process … this limitation strikes me as neither desirable nor practical” (p48). So “I ruled out the idea of excluding residential real estate from capital on the grounds that it is ‘unproductive’, unlike productive capital used by firms and governments … the truth is that all these forms of wealth are useful and productive and reflect capital’s two major economic functions.”

Well, residential property is obviously useful to the user - it has use-value, as Marx would say. But this form of wealth is not productive of new value (or profit), unless it is owned by a real estate company which rents it out as a business. Nevertheless, Piketty concocts a way for this wealth to deliver income: “residential real estate can be seen as a capital asset that yields ‘housing services’, whose value is measured by their rental equivalent”. Does this matter? Oh yes.

By including residential property, net financial assets and land in his definition of capital, Piketty reaches opposite conclusions from Marx on the return on capital, or what Marx called the rate of profit. And that matters. For a start, it means that Piketty is interested in the distribution of wealth and not on how it is produced. For him, the former provides the key contradiction of capitalism, while for Marx that contradiction lies in the latter process. For Marx, private ownership of the means of production for profit is the major fault line in modern society; for Piketty, private wealth is accepted forever; it is just too unequal.

R greater than g

This brings us to what Piketty designates grandiosely as his “first fundamental law of capitalism” (p52). Capital’s share of national income (α) is equal to the capital income ratio (β) in an economy, multiplied by the net rate of return on capital (r). So inequality of wealth, as expressed by capital’s share of income, will rise if the rate of return on the existing wealth ratio (the capital income ratio) rises. Alternatively, the wealth ratio will rise if capital’s share of national income rises.

According to Piketty, his law is better than Marx’s law of the tendency of the rate of profit to fall. As he says, “the rate of return on capital is a central concept in many economic theories. In particular, Marxist analysis emphasises the falling rate of profit - a historical prediction that has turned out to be quite wrong, although it does contain an interesting intuition” (p52). Marx was wrong because he reckoned that r would fall over time and this caused recurrent crises. Instead, Piketty tells us that actually r does not fall over time, but rises or at least stays pretty steady. So the issue for 21st century capitalism is: if r (the rate of return on capital) is greater than g (net real national income growth rate), then capital’s share of income will grow and the global capital/income ratio will eventually reach socially unacceptable levels.

The central crisis for capitalism is thus a distributional one. When the net rate of return on capital outstrips the growth of net national income - ie, when r is greater than g - “the inequality r>g in one sense implies that the past tends to devour the future: wealth originating in the past automatically grows more rapidly even without labour than wealth stemming from work which can be saved”. So even an “apparently small gap between the return on capital and the rate of growth can in the long run have powerful and destabilising effects on the structure and dynamics of social inequality” (p77).

There is little or nothing in Piketty’s book about booms and slumps, or about the great depression, the great recession or other recessions, except the comment that the great recession was a “financial panic” (as claimed by Ben Bernanke) and was not as bad as the great depression because of the intervention of the central banks and the state. There is nothing about the waste of production, jobs and incomes caused by recurrent crises in the capitalist mode of production.

Instead, Piketty adopts the usual neoclassical explanation that these events, like wars, were exogenous “shocks” to the long-term expansion of productivity and economic growth under capitalism (p170). Crises are just short-term shocks and we can revert to his fundamental law instead, “as it allows us to understand the potential equilibrium level toward which the capital income ratio tends in the long run when the effects of shocks and crises have dissipated”. Keynes might retort: ‘We are all dead in the long run.’

For Piketty, r>g is a tendency that is sometimes overcome by counter-tendencies, or a divergence sometimes countered by convergence. For example, between 1913 and 1950, r fell sharply and so in the period after the war g was higher than r and inequality fell (see fig 1). The other side of the coin is Piketty’s forecast that r will exceed g for the rest of this century and thus increase capital’s share of income and inequality.

This is because global growth will slow. Output per head has increased on average by 1.6% a year since 1700 - half due to population growth and half to productivity growth. Growth rates of 3%-4% only existed for brief periods. Also “population growth is slowing from 1.3% a year to 0.4% by the 2030s and “there is no historical example of a country at the world technological frontier whose growth in per capita output exceeded 1.5% over a lengthy period of time” (p93). The 20th century saw emerging economies like Japan, Korea, China and India ‘catch up’ with slowing advanced economies and so keep the global rate high by historic standards. But in the 21st century there are no catch-up economies of any size left (p97). Economies have reached the end of the technology frontier. In contrast, Piketty claims that his r “is pretty much steady around 4%-5%” (p55).

But is Piketty’s r steady and likely to stay so? Part of it is made up of returns on financial capital (stocks and bonds). The long-term return on interest-bearing and dividend-bearing financial capital has been falling, not rising, since the 1930s.10 On current trends, it is heading for zero by 2050 - not over 4%, as Piketty projects.11 But then, Piketty’s r also incorporates a return from property, synthetically generated as equivalent rents from ‘housing services’. This assumes that if you own your house you are making an income from it, even though you just live in it!

Without that, Piketty’s r would be falling, not rising. That is because the share of housing in ‘capital’ in Piketty’s data was more than half by 2010 compared to much less than half in 1940s. This is what affects r. The overall value of r has not changed because land has been replaced in Piketty’s ‘capital’ mostly by housing (p118). Farmland was two-thirds of capital in the 18th century, but hardly more than 2% in France and UK now: “once it was mainly land, but has become primarily housing plus industrial and financial assets (half in half)” (p122).

This has concerned other reviewers.12 If capital includes property and net financial assets, as well as tangible assets like industrial plant, offices, machinery and technology, then capital values can be volatile and deliver a net rate of return that is not steady. Piketty’s data show that the biggest reversal of the inexorable rise in the capital income ratio in the 20th century took place during the great depression and the ensuing world war. This delivered a U-shape to the movement of the global capital-income ratio (see fig 2). But from the 1950s the capital income ratio began to rise inexorably.

Piketty admits that a financial asset price bubble accounted for one-third of that increase in national capital to national income in this period (p91). From the 1980s onwards, where there is a big jump in inequality, is precisely when financial asset prices boomed. Piketty dismisses the argument that financial speculation will distort his “steady” rate of return, because, over the long run, he expects financial asset prices to be in line with the value of tangible assets. But it would have to be a very long run, because in the last 60 years that has not been the case.

r not a marginal return

This brings us to what Piketty, again rather self-importantly, calls “the second fundamental law of capitalism”. This is β=s/g. In words: the capital/income ratio (β) is equal to the savings rate (s) divided by the growth rate (g) over the long run. Piketty reckons that his ‘second law’ provides the explanation of why the global capital income ratio will rise: net income growth (g) will slow, while the net rate of return (r) will stabilise at a significant level above the growth rate and the net savings rate will reach an equilibrium level over time that is much higher than now.

Here, Piketty turns to the traditional neoclassical aggregate production function model developed by Robert Solow.13 In this model, all ‘factors of production’ make a contribution to growth. If there is an increase in one factor relative to another in contributing to output, then its ‘marginal productivity’ will fall. An abundance of a factor, capital, will lead to diminishing returns on that factor: “Too much capital kills the return on capital … it is natural to expect that marginal productivity of capital decreases, as the stock of capital increases” (p215). This would suggest that r should fall. However, Piketty reckons that r will not drop fast enough to stop the share of capital income from rising: “This implies that the capital share in income is rising faster than the net rate of return is falling” (p173). In the neoclassical model, this assumes an ‘elasticity of substitution’ between capital and labour greater than one - namely that labour will be replaced by capital quicker than the accumulation of capital will lead to a fall in its ‘marginal product’. Actually, as other reviewers have pointed out, there is not one empirical study that shows such a high elasticity except Piketty’s. They all show that r starts to fall after a while. Piketty’s answer is that over the very long run it will not.

Anyway, this neoclassical model of growth was debunked a long time ago. Piketty refers to the great debate between the Cambridge economists of Massachusetts (Robert Solow, Paul Samuelson) and those of Cambridge, England (Joan Robinson, etc), which ended in defeat for the former. The latter group showed that, if capital is a physical entity in machines, plant, etc, it cannot be valued in money and it cannot be infinitely substituted for labour.14 So the theory bears no relation to reality. Piketty’s answer is to turn to the facts. The Cambridge debate could not be resolved because of a “lack of data”. It does not matter who was right because the capital-income ratio has been rising in recent decades and that is all we need to know.

In effect, Piketty dispenses with the aggregate production model that he started to use to explain his second law: “The main problem ... is quite simply that it fails to explain the diversity of the wage distribution we observe in different countries at different times” (p308). Instead he adopts an institutional explanation - that the wealthy control government and companies and so ensure that they collect more than their ‘just’ marginal return on capital: “There is every reason to believe that r will be much greater than g in the decades ahead because of ‘oligarchic divergence’” (p463). This divergence is even greater because the rich hide their wealth in tax havens (p466).

Piketty concludes that “Top managers by and large have the power to set their own remuneration - in some case without limit and in many cases without any clear relation to their individual productivity” (p24). These super-salaries are very often just “pay for luck” (p335). So the marginal productivity of capital has nothing to do with it. In effect, Piketty is agreeing with Marx that these obscene “wages of superintendence of labour” (Marx’s term) are a concealed portion of surplus value extracted from wage labour.

Marx or Piketty: whose r?

Piketty argues that Marx’s law of the tendency of the rate of profit to fall was based on an assumption that there was “an infinite accumulation of capital” and “ever more increasing quantities of capital lead inexorably to a falling rate of profit (ie, return on capital) and eventually to their own downfall, while growth in net income (g) falls to zero” (p228). Here Piketty imposes the marginal productivity theory of capital accumulation on Marx - the very one that he rejects for himself: namely, that an abundance of capital leads to diminishing returns.

Actually, for Marx, the movement in r is to be found not in “infinite accumulation”, but in the rise in value of the means of production relative to the value of labour (-power). Piketty says that after World War II, capital was scarce and so the return on capital was high. Marx would have said capital values had been destroyed (both physically and in value), so the rate of profit was high. It was not ‘scarcity of capital’, but the destruction of its value (by war or slump).

We can check if Marx’s law of the tendency of the rate of profit to fall bears out in reality over the long run against Piketty’s “steady state” r. Fig 3 shows Esteban Maito’s world rate of profit going back to 1869, using a Marxian definition of capital.15

Unlike Piketty, Maito leaves out residential property and financial assets, and correctly categorises capital as the value of the means of production owned and accumulated in the capitalist sector. The result is not some steady r, but a falling rate of profit à la Marx. There is a long-term decline, but there are various periods when the rate of profit rises or consolidates.16

I used Piketty’s own voluminous data for Germany to compare his rate of return with Maito’s Marxian rate of profit for that country since the 1950s. Piketty’s data produce a similar result to Maito’s. The rate of return for Germany falls from 1950 and then stabilises from the 1980s. This is because Germans have a much lower ownership of residential property. Only 44% of German households own their own homes, compared with 70%-80% in Greece, Italy and Spain. When residential property is not a large share of ‘capital’, Piketty’s and Marx’s r move in much the same way.

The good and the bad

Piketty shows compellingly that inequality of wealth and income is getting higher in most capitalist economies. The reason is a rise of income going to capital in the form of profits, rent and interest and not due to the more skilled labour getting higher income than the less skilled. And the rising capital-income ratio is driven mainly by inherited wealth. ‘From rags to riches’ is not the story of capitalist wealth: it is more ‘From father to son’ or ‘From husband to widow’.

But then Piketty tries to develop some “fundamental laws of capitalism” and comes a cropper. He conflates capital into wealth by including non-productive assets like housing, stocks and bonds in his measure. In doing so, he loses sight of how wealth is created and appropriated, as Marx shows with his law of value. And his net rate of return on capital becomes separated from the capitalist process of production. Indeed, if you strip out housing and financial assets from his measure of the rate of return, you get Marx’s rate of profit and it falls (and moves up and down), unlike Piketty’s “steady” r.

As a result, Piketty has no theory of crises in capitalism and assumes they are passing phenomena. So his policy prescriptions for a better world are confined to progressive taxation and a global wealth tax to ‘correct’ capitalist inequality. Yet Piketty recognises that it is utopian to expect the wealthy (who control governments) to agree to a reduction in their own wealth in order to save capitalism from future social upheaval. He never thinks of suggesting another way to achieve a reduction in inequality: namely, to raise wage income share through labour struggles and to free trade unions from the shackles of labour legislation.

And he does not raise more radical policies to take over the banks and large companies, stop the payment of grotesque salaries to top executives and end the risk-taking scams that have brought economies to their knees. For Piketty - in true social democratic fashion - the replacement of the capitalist mode of production is not necessary.

Michael Roberts

Michael Roberts is author of The great recession: a Marxist view, published by Lulu (2009). His next book, The long depression, is forthcoming. He regularly blogs at thenextrecession.wordpress.com.


1. B Milanović, ‘The return of patrimonial capitalism’ - review of Capital in the 21st century, World Bank, October 9 2013, draft for Journal of Economic Literature, June 2014.

2. P Krugman, ‘Why we are in a new gilded age’ The New York Review of Books May 8 2014.

3. M Wolf, ‘Capital in the 21st century’ Financial Times April 15 2014.

4. J Cassidy, ‘Forces of divergence’ The New Yorker March 31 2014.

5. www.nytimes.com/2014/04/20/business/international/taking-on-adam-smith-and-karl-marx.html?_r=0.

6. Interview with New Republic: www.newrepublic.com/article/117655/thomas-piketty-interview-economist-discusses-his-distaste-marx.

7. http://blogs.ft.com/money-supply/2014/05/23/piketty-response-to-ft-data-concerns.

8. http://thenextrecession.files.wordpress.com/2014/05/pikettyzucman2014hid.pdf.

9. Now in an open 10-page letter posted online, professor Piketty defends his use of data and his overarching conclusion: http://piketty.pse.ens.fr/files/capital21c/en/Piketty2014TechnicalAppendixResponsetoFT.pdf. He argues that apparent transcription errors were deliberate adjustments and he defends his use of certain data sources over others.

10. See R Ibbotson and R Sinquefield, ‘Stocks, bonds, bills and inflation: year by year historical returns’ Journal of Business January 1976.

11. See B Eichengreen Project Syndicate April 27 2014. He refers to International Monetary Fund data showing that the real interest rate on bonds has been falling for three decades and, at 2%-3%, is hardly above the potential growth rate of the major Organisation for Economic Cooperation and Development economies.

12. James Galbraith argues that Piketty “conflates physical capital equipment with all forms of money-valued wealth, including land and housing, whether that wealth is in productive use or not. He excludes only what neoclassical economists call ‘human capital’, presumably because it can’t be bought and sold. Then he estimates the market value of that wealth. His measure of capital is not physical but financial” (‘Capital for the 21st century?’ Dissent spring 2014). This leads to problems of measurement as asset prices are volatile, although Piketty claims that they are not over the long run.

13. See R Solow, ‘A contribution to the theory of economic growth’ Quarterly Journal of Economics February 1956.

14. “At least since Wicksell it is well known that capital goods cannot be measured and aggregated in physical units because of their heterogeneity: how does one add up an airplane and a printing machine? Therefore valuation measures must be used. The value of a capital good can be the cost of its production or the value of the output that it will produce in the future. Both approaches require an interest rate (discount rate), but that interest rate is usually determined by using the amount of capital in relation to output” (S Bergheim Long-run growth forecasting 2008).

15. E Maito, ‘The historical transience of capital, the downward trend in the rate of profit since the 19th century’: http://thenextrecession.wordpress.com/2014/04/23/a-world-rate-of-profit-revisited-with-maito-and-piketty.

16. See Piketty’s tables for Germany at http://piketty.pse.ens.fr/fr/capital21c.