Share, , Google Plus, Pinterest,


Picking at Piketty.

By Constantin Gurdgiev.

Thomas Piketty’s ‘Capital in the Twenty First Century’ (Harvard University Press, 2014) has ignited both public and professional debates about the economics of income and wealth distribution not seen since the inter-war period a century ago, when applied Marxism collided with laissez-faire economics. To give the credit due to the author and his book, this attention is deserved.

Like Marx’s ‘Das Kapital’, Piketty’s volume is sizeable enough to induce unwavering submission from the reader to its meticulously factual and theoretically all-encompassing virtues. Like Marx’s opus, ‘Capital in the Twenty First Century’ is impenetrable to anyone unequipped with an advanced degree in political economy and understanding of economic theory. They both aim to herald a Revolution, indeed essentially the same revolution: the dis-endowed against the endowed. Like Marxist debates of the 1930s, Piketty’s thesis comes at a time of major upheaval and crisis.

And Piketty’s work is destined to stay with us for a long, long time. Its thesis of the coming age of chaos rising from the chain reactions of growing wealth inequality will be fuelling activists’ imaginations for decades. Yet, perhaps to the surprise of the majority of non-specialists, the book has within a month of its publication faded into the background in the world of economics.

The reason for this is that the comprehensiveness of the book’s ambition – of creating a unified theory of future economic development – renders it an easy target for criticism, challenge and, ultimately, negation amongst economists.
Before diving deeper into Piketty’s work, let me state three facts.

Firstly, I admire Piketty for his audacity to challenge the orthodoxy of macroeconomics and tackle a broad-ranging set of targets. Ninety-nine point nine percent of economic literature explores the minutiae of empirical or theoretical cul-de-sacs in specific sub-divisions of sub-fields of economics. Piketty falls into the 0.1 percent of economists who pursue the big picture.

Secondly, witnessing the vitriol with which Piketty’s book was greeted in economic policy circles, I have defended his work in the media and on my blog.

Lastly, having read Piketty’s academic publications and working papers in the past, I found his book to be inferior to his academic publications. ‘Capital in the Twenty First Century’ is too long and stylistically un-engaging to be worth returning to in the future.

The last fact means that you should read Piketty’s thesis and be aware of his core evidence, as well as the growing evidence of its shortcomings. The best means for acquiring this information is by reading Piketty’s articles and interviews, as well as taking in the debates surrounding his book.

But you should not buy ‘Capital in the Twenty First Century’, unless you are determined to impress your friends with your economic scrupulousness, in which case you had better avail of Flann O’Brien’s gentlemanly service that can get the tome thumbed, marked and annotated for you with scientific-sounding marginalia.

Piketty’s core thesis is based on what he defines as the “fundamental laws” of capitalism. Both of these laws stem directly from his view that economic inputs can be grouped into only two categories: capital (something that can be bought and sold, and thus accumulated without limit) and labour (something that cannot be sold, although it does collect wage returns, and cannot be accumulated without limit). Incidentally, outside undergraduate economics, this division remains valid only in the pre-1980s literature.

Piketty’s First Law states that capital’s share of income is the ratio of income from capital (or return to capital times the quantum or stock of capital) divided by national income (for example, GDP).

As anyone with a basic knowledge of economics would know, this is not a law, but an accounting identity. Furthermore, any student of economics would spot a glaring problem with the above definition: it applies to all forms of capital, including the ones that Piketty omits.

This brings us to the first major problem with Piketty’s core thesis: capital itself is neither homogeneous, nor does it yield a deterministic and singular rate of return. Instead, capital takes various forms.

There is financial capital – where the rate of return is measured in the form of equity returns, bond returns, financial portfolio returns and so on. There is intellectual capital that can be traded. This generates financial returns to holders/investors, but also yields productivity gains to its users, including workers.

There is human capital – which (along with other inputs into production) generates returns to labour (wages and performance-related bonuses), but also returns to entrepreneurship, creativity of employees and so on. There is managerial and technological know-how that can be invested in and transferred or sold, albeit imperfectly, in so far as it often attaches to labour and skills.

To measure the income share of all of these forms of capital, one simply needs to divide income from the specific form of capital by total income. It is the same for labour’s share – and for any other input share. This is neither Piketty’s own discovery, nor a law of Capitalism.

The problem is that in many cases we cannot easily measure returns to the more complex forms of capital. And a further problem is that returns to one form of capital are linked to returns to other forms of capital. A good example here is urban land. Return to this form of capital is strongly determined by the returns to human capital that can be deployed on this land, as well as by know-how and technology that attaches to the economic activity that can take place on it.

Piketty’s second fundamental law is a theoretical proposition derived from mainstream macroeconomic theory. The author claims that the ratio of the stock of capital to income will be equal to the ratio of the savings rate to the sum of the growth rates in technology and population. Together with the first law this implies that the income share of capital equals the ratio of the product of the return on capital and savings rate to the combined growth rate in technology and population.

Piketty’s main thesis is that over time, as growth rates in technology and population fall, capital’s share of income will rise, resulting is a sharp rise in inequality.

The core corollary of this for Piketty is his call for a global tax on capital (or wealth) coupled with a massive rise in income tax on super-earners. These measures, in his view, can ameliorate the increase in the income share of capital triggered by slower growth.

There are numerous and significant problems with Piketty’s analysis and even more problems with conjectures he draws out of data. Some of these have been discovered by other researchers, few are the result of my own assessment of Piketty’s work.

Although Piketty presents numerous factual arguments describing the rise and fall and the rise again in income and wealth inequalities, his factual arguments are tangential to his theoretical proposition. Per Krusell (Stockholm University) and Tony Smith (Yale University) have pointed out that “Piketty’s forecast does not rest primarily on an extrapolation of recent trends that he has uncovered in the data…”.

Krussell and Smith go on to show that Piketty’s second fundamental law relies not on data, but on an assumption that the ‘net’ saving rate is constant and positive over time. This means that capital stock rises by an amount that is a constant fraction of national income.

Now, suppose that Piketty is correct. And suppose that population growth and technological progress fall to near-zero. Piketty’s assumption then implies that an ever-greater share of economic output will have to be used to maintain capital stock. This will crowd out investments in education, health and new technologies. Eventually capital formation will consume the entire GDP. As a number of observers pointed out, this has never been observed in the past and cannot be true in the future.

Now, personally, I do believe we are staring into the prospect of diminished rates of growth in the advanced economies. But I also believe that savings follow growth over the long run, implying that gross investment – investment including replacement of capital depreciation and amortisation – is relatively constant as a proportion of national income. At times of structurally slow growth, therefore, savings are also low. And when it comes to demographics: older and shrinking populations imply dissaving.

This is supported by historical evidence and contradicts Piketty’s conjecture.

Furthermore, this macroeconomic evidence is supported by data from individual consumers’ behaviour. In cyclical recessions, households do engage in increased savings, known as precautionary savings. But this is short-lived and does not contribute to increased investment. Over time slower growth in income brings lower rates of savings.

Aside from the above, Piketty’s suggestion that a wealth tax can stem the rise of inequality is illogical.

Wealth taxes tend to decrease the quantity of capital, thus raising the scarcity and the quality of it. The result is higher returns to capital in the long run that will at least in part neuter the wealth tax effects on capital stocks. Scarcer goods tend to command higher prices.

The problem with wealth inequality is the distorting nature of taxation, not tax levels per se. To see this, take three forms of capital: financial assets, intellectual property and human capital.

Tax rates on financial assets normally run close to zero since those well-off enough to afford the financial engineering services required to attract such rates in the first place usually deploy off-shore schemes for tax optimisation.

Each percentage point in return to financial assets held by a wealthy Irish owner attracts a tax of under 10 percent (inclusive of costs of tax optimisation). Capital-gains rates run also well below income-tax rates. In Ireland today, the headline rate is 30%. Intellectual property attracts an effectively near-zero tax rate.

Whereas professional or institutional investors in traditional capital collect roughly 85-90 cents on each euro of gains, intellectual property investors collect closer to 90 cents and retail investors pocket around 70 cents. On the other hand, human-capital returns are taxed at high marginal rates. Thus a professional consultant will collect around 45 cents on each euro returned to her from added investment in her education and skills.

The result of this asymmetric treatment of returns from various forms of capital is that households simply have no surplus income left to invest and from which to accumulate wealth. Instead, wealth grows in the hands of those who live off rents and start their lives with inherited capital.

To make things worse, Piketty also calls for dramatic rises in upper marginal tax rate – to hit high earners. This too is directly contradictory to the objectives he claims to pursue.

Upper marginal income tax hits those who live off the wealth of the businesses they built and the skills they acquired. Capital gains tax hits those who either dispose of the businesses they built or sell capital they accumulated or inherited.

Two of these groups of earners are collecting on value they created. One is collecting on what others created for them. Tarring them all with one brush will simply reduce future rates of growth and/or reduce rates of return on non-capital income. In other words, Piketty’s income-tax policy proposal will lead to higher wealth and income inequality in the long run.


The solution to this dilemma is not to tax all capital more, but to equalise the rates of taxation on all capital: physical, financial, technological and human. And focus on what Jacob Hacker of Yale University calls “pre-distribution” of labour income. The latter requires simultaneously addressing three determinants of market wages: education and skills (for those with low incomes), enterprise policy (supporting demand for these skills) and mobility and efficiency of the labour market (increasing returns to these skills).

Another part of the solution is to expand the holdings of capital (financial, physical, intellectual and human) across populations by levelling the playing-field for household investments relative to professional and institutional investors.

Piketty’s work deserves huge credit for bringing to the fore of the economics debate legitimate concerns with inequality. However, here too the book is open to criticism for being based on occasionally thin evidence.

‘Capital in the Twenty First Century’ is premised on the assumption that wealth inequality is tearing societies apart, leading to violent conflicts and breakdowns of   civic and state institutions. There is very little evidence to support this assertion amongst the advanced economies.

Extreme poverty, measured in absolute terms, can be exceptionally dangerous. So much is true. But relative inequality to-date has not been a major flash-point for revolutions whenever such inequality is anchored in some meritocratic foundations for wealth distribution.

All of the recent disturbances in advanced economies have been about income and wealth inequality according to activists and the media. But looked at closely, all have been linked to either public policies relating to income and opportunities available to the less-well-off or to diminished growth rates in the local economies, or both.

More importantly, current research shows that individual perceptions of relative income and wealth inequality strongly depend on which reference group is selected for benchmarking against.

For example, a 2013 paper by Daniel Sacks, Betsey Stevenson and Justin Wolfer, ‘The New Stylized Facts About Income and Subjective Well-Being’ finds that there is little evidence to support theories that emphasise the importance of relative income. In simple terms, if you are concerned with inequality, you should focus on increasing the rates of growth in the economy, not depressing the rates of return on capital.

Another study, by Maria Dahlin, Arie Kapteyn and Caroline Tassot, titled ‘Who are the Joneses?’ (CESR, June 2014) shows that individuals are “much more likely to compare their income to the incomes of their family and friends, their co-workers and people their age than to people living in the same street, town … or in the world”. We reference our own wellbeing against the wellbeing of those close to us socially.

In this case, Piketty’s policy prescription should call for taxing rich people with greater familial networks at a higher rate than those with fewer familial ties. Which, of course, is absurd.

Perhaps the greatest error in Piketty’s logic is the failure to account for other forms of capital – an error exactly identical to that committed by Marx.

Ricardo Hausmann from Harvard  shows that Piketty’s argument completely falls apart at the national accounts level in the case of advanced and emerging economies. Furthermore, Haussman’s argument dovetails with my view that hiking upper marginal tax rates to combat income and wealth inequality is simply counterproductive.

Piketty’s assumption that the rate of return to capital is following a historically constant trend of 4-5 percent per annum is also questionable. Dani Rodrik of Princeton University reminds us that the return to capital is likely to decline if the economy becomes too rich in capital relative to labour and other resources and the rate of innovation slows down.

So if innovation were to fall, as Piketty assumes, the rate of return to capital is likely to decline in line with diminished economic growth. This decline is going to be further accelerated by the rise in the quantum of capital accumulated prior to the economic slowdown.

Lastly, since capital is non-homogeneous, even a constant average return can conceal wide variations in returns to various forms of capital. For example: agricultural land vs industrial property, private equity vs listed shares and so on – all command different, and over time varying, returns.

Imposing a uniform tax on all wealth will raise the cost of investing in more productive and less certain (thus ‘pricier’) capital associated with new technologies and new industries. In turn, this will only reduce the mobility of wealth in society, increasing, not lowering, long-run wealth inequality and supporting currently endowed elites at the expense of challengers.

The truth is that the Marxist world of the epic confrontation between labour and capital has been bypassed by reality. Today, we live in a highly complex, dynamic and less heterogeneous economy. This does not mean that the burdens of rising income and wealth inequality should be ignored. But it does mean that policy responses to these challenges must be based on more complex analysis, anchored in macroeconomic and behavioural theory.

Piketty’s ‘Capital in the Twenty First Century’ spectacularly succeeded in raising to prominence the debate about income and wealth distributions. But its analysis and recommendations are flawed. •