Recent Posts

Sunday, March 30, 2014

Book Review: Capital In The Twenty-First Century

The book “Capital In The Twenty-First Century” by the Parisian academic Thomas Piketty (published in English March 10, 2014) has already made considerable waves, reigniting the debate around income and wealth inequality. I believe that the book is interesting, and is aimed at reaching a wide audience. However, the process of eliminating technical details makes the analysis less satisfying. The book offers considerable data about the evolution of inequality. However the explanation of why inequality has risen, and his proposed solutions for inequality are less convincing as a result of the simplifications made.


The book is lengthy, weighing in at 696 pages, and was translated from French by Arthur Goldhammer. The book is divided into 4 parts:
  1. Income and Capita
  2. The Dynamics of the Capital/Income Ratio
  3. The Structure of Inequality
  4. Regulating Capital In The Twenty-First Century
The scope of the book is large: he looks at the trends in wealth over centuries for some countries, and also includes broad international comparisons. There is a focus on France and the United States, since there is a long history of data as the result of estate and income taxes. The first three parts create an analytical framework and analyse the data on inequality, the final part presents Piketty’s preferred policy responses.

The book is aimed at non-economists, and he explains the various concepts used. He uses considerable literary references to illustrate earlier era’s attitudes towards wealth. For example, he notes how price stability in the 19th century allowed writers to summarise a character’s social status just by giving their annual or monthly wages. A certain income level would be enough to immediately place a character within a particular social class.

The book is non-mathematical, other than some simple equations that characterise the inter-relationship between growth rates, the rate of return on capital, and wealth, or capital as he characterises it.
His key argument is that the rich do not need to spend much of their existing wealth in order to lead an extravagant lifestyle. If you spend 1% of your wealth each year, as long as your after-tax rate of return is 1% greater than the rate of growth of worker’s wages, your stock of capital will grow relative to wages. He argues that since the very rich can only spend a very small percentage of their wealth, the possibility exists of increasing dynastic concentration of wealth. He summarises this condition as r>g; the rate of return on capital (wealth) is greater than the growth rate.

Questions Of Theory

There is no doubt that inequality has been increasing since the end of World War II, and this tendency towards inequality accelerated starting in the 1980’s. Brad DeLong has a list of what he refers to as worthwhile reviews of the book here; those reviewers generally discuss the observation that inequality is increasing, and the implications. As was observed by Ian Welsh, this is obvious; all you needed to do is pay slight attention to the trends in the news.

Where the problems start with the book is the exact mechanisms described by Piketty. The review “Kapital for the Twenty-First Century?” by James K. Galbraith discusses some of the issues that arise from the simplifications made. Piketty defines “capital” as being the same thing as “wealth”; someone with $1 billion in assets has $1 billion in “capital” under his definition. Galbraith gives the history of the historical debates over the definition of capital within economics ("The Cambridge Capital Debates"), and an extremely brief summary is that equating the two concepts is internally inconsistent. On a less theoretical note, any experience with real-world financial markets tells one that the relationship between physical capital and financial wealth is tenuous at best.
Chart: U.S. 10-Year Treasury Bond Yield

In simplifying the math, he is forced to look at steady state relationships. However, there have been obvious changes in financial markets since the early 1980’s, which tell us that we were not in a steady state over that period. Interest rates fell dramatically (as shown above). Stock markets had a major bull run. Some of that reflects a rising profit share of national income, but it also reflects expansion of valuation multiples. Since the rich have the highest proportion of equities, a rising stock market means that inequality automatically rises. A very significant portion of inequality has been created via the concentrated positions of the original owners of companies that have gone public. As long as equity investors are willing to pay large multiples for growth stocks, this is not easily redressed.

If he had used Stock-Flow Consistent modelling methodology, he could have more clearly distinguished these dynamic effects from steady-state conditions. However, it would have made the mathematics difficult, and so he would have lost a lot of potential readers.

As I discussed in a previous article, the rise in the profit share of the economy has followed the trajectory of inflation. Wage inflation is being suppressed by the inflation-targeting, and it is not an amazing coincidence that wages as a share of national income is falling. With the trend of disinflation largely finished (it is very difficult for deflation to occur in practice), is this trend of falling wage share/rising profit share largely behind us?

Galbraith also raises an important issue with respect to the income tax data Piketty uses. The high inequality detected in the United States relative to international comparisons may just reflect the relative competence of the “feared” Internal Revenue Service. Additionally, if we just look at the trends over time within one country, the data are influenced by changes of behaviour. The high statutory income tax rates of earlier eras meant that compensation to executives was often in the form of non-taxable perks, and wealth was just been left to accumulate in corporate holdings, outside the reach of the personal tax system.

I have some concerns about how he approaches wealth data. He has chosen a methodology that ignores the stabilising impact of the welfare state. He looks at estate data – which excludes the implicit wealth of defined benefit and state-provided pensions.

The amounts involved are non-trivial. Using the Canadian system as an example, the Canadian Pension Plan plus Old Age Security payments add up to about a maximum of $1,500 per month for a Canadian citizen that spent their adult life working. To purchase that income stream using an annuity would currently cost at least $300,000, and this is not including the considerable value of inflation indexation and the preferred tax treatment those payments receive. The age cohorts who are in retirement and/or near retiring worked in an era when defined benefit pensions were widespread, and those payments are often much larger than this state-provided safety net. However, these benefits disappear at the death of a surviving spouse. They are thus invisible to Piketty’s methodology. But the cash flows they provide are very real, and important.

If you believe that the welfare state is doing very little to help inequality, the implications are unsettling. The preferred conclusion by progressives is that this means that it is necessary to do more. But this misses an alternative interpretation– if the welfare state is doing little, there is little cost to dismantling it. As a result, it is necessary to measure accurately the value of what the welfare state delivers.

Analysis Of Bonds

His analysis of the bond market is extremely conventional, or “out of paradigm” from the point of view of Modern Monetary Theory (MMT). To be fair, he is writing in France, which has placed itself in yet another fixed exchange rate regime. As a result, he has an excuse for operating within a Gold Standard analysis framework.

For example, at one point, he discusses the idea of extinguishing government debt levels via a one-time wealth levy. Even a few minutes of thinking of how that would work within a stock-flow consistent model tells us that such a policy would be a catastrophe. There would be no way of the owners of capital to raise the cash to meet such a tax obligation.

He largely treats government bond payments as a subsidy to the rich. However, government liabilities pretty much have to have a non-zero interest rate if you want to keep the private sector term structure away from zero, allowing the central bank to regulate economic activity, as per mainstream economic analysis.* Low interest rates appear to act as a subsidy to the corporate sector at the expense of households who hold fixed income assets. Since the very rich have portfolios weighted towards equities, low interest rates presumably accentuate inequality.


Piketty argues that “Taxation is not a technical matter. It is preeminantly a political and philosophical issue, perhaps the most important of all political issues. [Chapter 14]”

I disagree with that assessment. It may have been a philosophical issue a few hundred years ago, but I would argue that those issues have been mainly settled, with a few free parameters policy makers can modify. For moral and technical reasons, income taxation is progressive – even cold-hearted reactionaries realise that you cannot get blood from a stone. Even “flat tax” proposals are progressive – there is an initial threshold where income is not taxed, which means that the total tax rate rises along with income. They are less progressive than systems with rising marginal rates, but the point remains that there is no serious debate about progressivity of total tax rates per se.

Given an emphasis on political statements, and not the messier study of real world results, Piketty is not surprisingly in favour of bringing back very high marginal rates on high income owners. But as James K. Galbraith notes in his review, and I discussed previously, those high rates were loophole-ridden. The tax system was widely viewed as unfair in consequence. Very high marginal tax rates were effective during World War II, when countries were locked in a total war. The return to what passes for peace in the post-1945 era made those extremely high marginal rates politically unsustainable, and so the statutory tax rates lost their economic significance.

There are other examples of tax reforms gone awry. The ability to deduct executive pay above $1 million was curtailed in a 1992 reform in the United States. This helped push the movement towards stock options, which made gargantuan executive compensation possible. Very simply, no company would have paid in cash the windfalls that stock option based compensation ultimately produced.

And even if executive pay is clamped down on, who benefits? Curbing the compensation of people who are nominally employees will probably just flow into the pockets of the owners. To my mind, one of the big risks for stock returns in the long-term is the tendency for top management to shower stock-based compensation over themselves. As Galbraith notes, it is probably better to raise wages at the bottom of the distribution - raising the minimum wage as one possible policy - rather attempt to set the wages at the top.

He also skips over problems associated with income taxation. He seems to imply in Chapter 15 that income in holding companies and trusts avoid income tax (which is why advocates a tax on capital).** In countries with sensible tax systems, such income is taxable, but it is done within the corporate income tax system.

He largely dismisses the technical difficulties of taxing corporate income as just being the result of tax havens. He follows the French official line that Ireland is acting in bad faith with its low statutory corporate tax rates. However, the Irish point out that French effective corporate tax rates are almost identical to Irish rates. Apparently, French corporate taxes are relatively porous, so as to not disadvantage their national champions. Raising Irish statutory corporate tax rates in Ireland to match French levels, without the French correspondingly plugging whatever holes exist in their tax laws, would create a huge tax “advantage” for France.

It is easy to be in favour of shifting the tax burden back towards multinational corporations. How this can be achieved  in practice, particularly for smaller countries, is not clear to me.

The Global Capital Tax

In order to break up accumulating pools of capital, Piketty proposes a progressive global tax on capital. As James K. Galbraith observes,
In any case, as Piketty admits, this proposal is “utopian.” To begin with, in a world where only a few countries accurately measure high incomes, it would require an entirely new tax base, a worldwide Domesday Book recording an annual measure of everyone’s personal net worth. That is beyond the abilities of even the NSA. And if the proposal is utopian, which is a synonym for futile, then why make it? Why spend an entire chapter on it—unless perhaps to incite the naive?

To give a simple example of the technical problems such a tax would raise, take the following example. Imagine that a consultant creates a corporation to operate under, and the corporation generates $100,000 in annual income. That private corporation is legally indistinguishable from any other private corporation, and so in principle its value would have to be taxed. If the valuation metric is to use a price/earnings multiple of 12, the corporation is worth $1.2 million. Additionally, every extra dollar of income raises the value imputed to the corporation by $12. If the owner faces a 1% wealth tax, it means that extra dollar of income will add $0.12 to the wealth tax paid. The wealth tax is equivalent to a 12% increase to the marginal income tax rate.

Meanwhile, the consultant’s next door neighbour is also a consultant who operates as a sole proprietorship. There is no asset to tax, hence there is no wealth tax to pay. So in other words, people in economically equivalent situations will face effective marginal income tax rates that are 12% different. This is obviously unfair. But if the tax is not applied in this fashion, the lack of taxation of a type of private corporation creates a hole in the law that accountants would drive a truck through.

Tax laws have been in a continuous state of reform for decades to remove such unfair situations. Since “capital” or “market value” is largely unobserved, a wealth tax is largely based on fictitious values. Incomes are defined on market transactions, and so there is at least a hope of assigning a value to them. (Estate taxes are based on asset valuations, but since the person involved is presumably dead, there is no question of the valuation of interactions between the person and the business. It is a final liquidation, and so there is better chance of valuing assets. The question of the equity of the division amongst heirs also raises the incentive for accurate valuations.)

He also discusses a narrower wealth tax – taxing bank deposits. Such a move flies in the face of the experience of the last financial crisis – regulators would like cash to be parked in the regular banking system, not in the unregulated shadow banking system. A recurring tax on deposits is an incitement to flee the regulated financial system, making surveillance and tax compliance even more difficult. Such bank taxes have been resorted to by various countries historically. However, those levies are typically an indication of a weak tax system, and they are thus not an example to emulate.


Inequality has become again the rallying point for progressive politics, and this book has ridden this wave. It provides an interesting an interesting introduction to economics. However, the reader will need to keep in mind that some simplifications have been made. Reading about corporate tax reform, such as thousand page tomes about the riddles of transfer pricing, is much less exciting than reviewing Balzac’s observations on the lives of the wealthy. Unfortunately, it is those technical details that have to be grappled with if the objective is to effectively shift tax burdens.


* It would is possible to finance the central government solely with bank reserves, as has been advocated by Modern Monetary Theory economists. And we are halfway there, courtesy of Quantitative Easing. This has the effect that the government balance sheet is now intermediated by the banking system. If interest is not paid on reserves, the banking system would have a huge competitive disadvantage to the unregulated shadow banking system. But once the government is paying interest on reserves, there is only a limited difference from issuing bonds.
** In Chapter 15, he states: “There are several ways to deal with this problem. One would be to tax all of a person’s income, including the part that accumulates in trusts, holding companies and partnerships.” My interpretation of this text within its context is that this is not currently being done. It may be that the French original text does not match my interpretation.

See Also:
(c) Brian Romanchuk 2014

1 comment:

  1. This comment has been removed by a blog administrator.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.