Tag Archives: data

Review: Capital in the 21st Century by Thomas Piketty

Now I have done my share to make Capital in the Twenty-First Century by Thomas Piketty a little less unread. Good timing, since he is visiting Oslo this week.

What I will write here will be the notes I took while reading the book, edited for clarity. It is well known now that the book is thick and contains very much information, Piketty explicitly wants to consider both big and small phenomena, both long and short term changes, hence disclaimer: There will inevitably be much that I will not have caught or write about, and there may also be things that I have misunderstood. But maybe these notes will be helpful for those who have themselves read parts of the book or know a little bit and are trying to understand more. Many others have written better proper reviews.

Conclusion: Much great empirics to learn from. Inequality is an important topic. Good figures. Good at using qualifications and caveats, necessary when writing such a broad book. Good that also tries to explain. Objections by others that the recent rise in inequality is caused by human capital, etc. are misguided. Fundamentals of and cause and effect relationships in theoretical framework should have been clearer, and objections often relevant. Recommended.

Part One: Income and capital
-Focus on US and France, to some extent Germany, Britain, and to a lesser extent Sweden.
-Convinced that inherited wealth will make a comeback.
-The capital/income ratio is defined as β = K/I. It is fine to call and think of capital as “wealth”. This ratio was quite high in the time around 1900 (about 7), meaning that there was a lot more wealth than income. This was a period with very high inequality, particularly in Europe. But with depressions and world wars, much capital was destroyed, thus β declined. In the post-war period, β has been rising again.
-The first fundamental law of capitalism: α=r x β. This is an accounting identity saying that capital’s share of income α equals the return on capital r times the capital/income ratio β. W
-The second law: β = s/g. s is saving. Also like an accounting identity. Supposed to hold asymptotically.
-Capital ownership is concentrated, so a high capital/income ratio is typically related to inequalty.
-Suggests competition between nations-> conservative revolution, Thatcher etc.
-Growth pessimist. Refers to his figures 2.2-2.5 as this largely determined but immediately afterwards when discussing inflation mentions that quality improvements and new products may mean that the growth rate really higher. In any case his growth pessimism is key since it means “wealth accumulated in the past will inevitably acquire disproportionate importance (p. 166).”
-Piketty emphasizes both that the capital/income ratio has increased and that inequality has increased since the world wars, and that r>g is a force for greater inequality, but he stresses repeatedly and clearly that r>g is not what is driving this recent rise in inequality. So people saying things like “Piketty is wrong because he claims that the rise in inequality is due to capital income while in truth it is due to a rise in wage inequality” (a group including notables such as Debraj Ray) simply cannot have read the book seriously. Indeed the point of a rise in wage inequality is treated at considerable length and referred to repeatedly. I have mentioned this before.
Part Two: The Dynamics of the Capital/Income Ratio
-On the other hand, his use of the “second law” β = s/g is a legitimate target for criticism, since he writes as if s and g are exogenous and independent variables determining the capital/income ratio, which is clearly not the case.
-The fundamental force for divergence: r > g. This has more bite than the “laws”. Piketty takes it as an empirical fact that the rate of return on capital, r, historically has been greater than economic growth g. In his view, the second half of the 20th century, in which the opposite was the case is an exception. Debraj Ray has forcefully attacked the emphasis on this inequality, pointing out that it is the saving propensities of the rich versus the poor that matters, not the form of their saving (here and here).
-Piketty refers to 19th century literature throughout, but he does so to show certain features of the economic environment: in the novels of that time there was no inflation; and the rate of return on capital was constant – capital and rent of ca 5% considered the same, and type of capital seen as unimportant (p 207).
-Most wealth privately held, tab 3.1.
-p 151 strange about “inevitably” about volume and price effects in housing.
-p 163 takes swipe at authors of wb reports regarding monetary value of hum cap.
-Volume effects, like savings, together with slow growth explaining most (though not all) of increase in private cap from 1970s. Ex of second law. Privatization also a cause, and increase in asset prices. Might one make the arg that privatization increased value? Higher value placed on the privately held capital.
-Rate of return depending on size of capital. I come back to this under Part three.
-p 220: “the novelty of this study…capital-labor split and the recent increase of capital’s share of national income in a broader historical context by focusing on the evolution of the capital/income ratio from the eighteenth century until now”.
-Acknowledges role of recent prod growth and diffusion of knowledge – have made it possible to avoid the apocalypse predicted by Marx (p. 234). But has not altered the deep structures of capital. Also sees “progressive taxes on income and inheritance (p. 278)” in the postwar period as reasons why rentiers have not come back.
Part three-the structure of inequality
-Destruction of war, bankruptcies in depression an policies like rent control reduced wealth inequality, allowed new start.
-France: large decrease in inequality since the Belle Époque, driven by income from capital (figs 8.1-2). Increase after WWII, due to both capital and labor, decrease from 1968, when minimum wage was increased faster than the average wage, then increase from 1982-83, when this policy was stopped. Overall, though, wage inequality stayed about the same, but increased at top end of the distribution since 1990’s.
-US: started the 20th century as more egalitarian than France and Europe, but has now overtaken them and become as unequal as France was at that time (fig 8.7). Decreased in prewar and WWII times, but has shot up sine the 70’s due to the rise of “supersalaries” going the top 1 percent income receivers (fig 8.8).
-The very rich still get most of their income from capital in both France and US. Figs 8.3-4 and 8.9-10.
9. Inequality of labor income
-Also consider other countries. Many of the same developments.
-Thinks we are not going away from society of rentiers, only going to another form. Agrees with Goldin and Katz that increased spending on education would be good, but considers the wage-according-to-skill-only-story, or the theory of marginal productivity, as incomplete, since “it fails to explain the diversity of the wage distributions we observe in different countries at different times (p. 308)”, sees it rather as holding in the long run. Also mentions the intrinsic value of health and education. In the US the minimum wage has gone in the opposite way as in France (fig 9.1). Minimum wage can be efficient if firms have monopoly power, this is its justification. Hard to see differences in objective measures if skill between top and almost top income earners; supermanagers emerged mostly in Anglo-Saxon countries, but not other developed countries. Social norms and acceptability. Difficult to explain observed pay differences by non-external variances-“pay for luck” present, and more so in countries with lower top marginal tax rate. Bargaining model-with lower marginal taxes, it paid for executives to bargain harder.
10. Inequality of capital ownership
-Is becoming increasingly concentrated.
-Wealth more concentrated. Data go longer back. Wealth inequality decreased substantially in the first half of the 20th century, some wealth going to “the patrimonial middle class”, the middle 40% of the wealth hierarchy. Decrease largely due to the shocks of 1914-45 and introduction of taxes on capital and capital income.
-Predicts that r will again rise above g, but says at the same time that this prediction rests on the assumption “that no significant political reaction will alter the course of capitalism and financial globalization over the course of the next two centuries (p. 358),” which is dubious “precisely because it’s inegalitarian consequences would be considerable and would probably not be tolerated indefinitely (p. 358).” This feels somewhat like overhedgning.
-Demographics also relevant. And international competition may drive capital taxes down.
11. Merit and inheritance in the long run
-Dynamics of inheritance
If r>g “the past tends to devour the future (p. 378)”
“Nearly all of the capital stock came from inheritance (p. 379)”. Drop after the war, but steady increasing (fig 11.1) in France. Can be seen as result of three u-shaped forces: capital/income ratio, mortality rate, wealth at death/average wealth. Shocks of 20th century relevant also here-destruction if what was to be inherited, and many died young. Even if people are paid according to merit, income inequality will turn into inheritance inequality. (Andrew Gelman has an old post related to this). Difference from before: wealth less concentrated, and income and wealth more correlated.
-The global dynamics of wealth
Similar patterns in some other countries as in France, but data difficulties in many countries, so uncertain. Also uncertainties about savings and consumption behaviors.
12. Global inequality of wealth in the twenty-first century
-Share of billionaires in the work is increasing (figs 12.1-2).
-Rate of return depending positively on wealth, ref Forbes rankings, table 12.1.
-University endowments, for which we have good data, also show high return and positive association between wealth and return (fig 12.2). Piketty believes this is due to “economies of scale in portfolio management (p. 450)”. Here the evidence does not seem so strong, and it is difficult to believe that capital owners would not run into diminishing returns at some point.

-Much entrepreneurial wealth on Forbes list, but probably more inherited. And entrepreneurs and their children in any case turn into rentiers. Comments on Gates and Slim on pp. 444-445 with footnote, (individual) merit not so straightforward. (Gates wrote a positive review nevertheless.)

Part four. Regulating capital in the 21st century
-I do not have so much to say about this one.
-The state more involved today than in the past. Taxes increased between 1910 and 1980, after that constant (fig 13.1). Social spending-construction of a “social state”. Mobility relatively low in the US, and “parents’ income has become an almost perfect predictor of university access (p. 485).” Argues for more public financing of universities.
Pay-as-you-go retirement systems connect generations and “makes for a virtuous and harmonious society (p. 488),” but face challenges when r becomes > g, which is the rate of return for these systems. Claims that the volatility of the return in capital was the reason for the introduction of paygo systems after WWII, when many people’s savings had been wiped out (p. 633n45), something I never learned about, and that this justification is still valid today. So how to deal with the challenges of r>g? Raising retirement age is one possibility, but we should be better at distinguishing between types of workers (intellectual occupations and others). Mentions possibility that share of intellectual workers will increase; and “One of the most important reforms the twenty-first century social state needs to make is to establish a unified retirement scheme based on individual accounts with equal rights for everyone, no matter how complex one’s career path.”-too many different rules and too much complexity. Sees the Swedish reforms of the 90’s as largely accomplishing this (p. 490n46).
13. Rethinking the progressive income tax
-Progressive taxes on income and inheritances new in the 20th century. Dismisses a direct tax on consumption quickly (p. 494), promising to come back to it (634n1), but hardly does (only briefly in 638n34). Rise of tax competition among countries, the tax system becoming regressive at the high end of the distribution. Sees this as threatening the common supper for taxation. US and Britain led the introduction of progressive taxes on income and inheritances, and had higher rates than France and Germany (figs 14.1-2). So high (marginal) rates that the purpose was not to raise revenue, but rather to limit such wealth. Rates substantially reduced since 80’s. Do not see effect on productivity, but believes the reduction increased the bargaining power of executives.
14. A global tax on capital
-The ideal tool, would “avoid an endless inegalitarian spiral (p. 515).” A useful utopia. Many challenges, but has some ideas about how Europe could implement it. The alternative is worse-capital control and protectionism, we already see some of this, in governments’ support of national champions. Hails redistribution in the US: Immigration is the mortar that holds the United States together, the stabilizing force that prevents accumulated capital the importance it has in Europe; it is also the force that makes the increasingly large inequalities of labor income in the United States politically and socially bearable. For a fair proportion of Americans in the bottom 50 percent of the income distribution, these inequalities are if secondary importance for the very simple reason that they were born in a less wealthy country and see themselves as being on an upward trajectory (p. 538).”
Other comments
-Wants economists to become more broad social scientists. Disses mathematical economists, then says that we cannot spend time on internal squabbles. I happen to agree w much of this, but it could maybe have been put differently.
-Footnotes contain large amounts of information, but are sometimes only dryly humorous or polemical:
621n57 taxi problem in the EU.
624n20 A&R on Gates and Slim. Also in text, but continues in the footnote.



Aging Ironmen

When following the Norseman Xtreme triathlon online last month, I googled Graeme Stewart, who lead the race for quite some time, and came across his blog. A post of his pointed me to an interesting article about the age and performance of the top ten finishers in the yearly Ironman World Championships in Kona, Hawaii. The authors (Gallmann, Knechtle, Rüst, Rosemann, Lepers ) first show that as in many other sports, the finishing times (of the top 10) have improved considerably the last 30 years (1982-2012): Total race performance-Gallmann Knechtle Rüst Rosemann Lepers 2013 Elite triathletes in Ironman Hawaii

This can be due to many factors. The surprising finding is that concurrently with going faster, the top finishers have become older. The average age of the top 10 men increased by 7 years, from 27 in 1983 to 34 in 2012!

Age of top finishers-Gallmann Knechtle Rüst Rosemann Lepers 2013 Elite triathletes in Ironman Hawaii

The authors speculate that experience may have become more important with time or that the race may have become more attractive to former professional athletes, but do not know the answer. The point Stewart makes is that the results suggest that it takes time to build endurance and that for most people there is room for improvement regardless of age.

Is Lionel Messi impossible?


Shooting Efficiency vs. Shooting Volume, by Benjamin Morris, http://fivethirtyeight.com/features/lionel-messi-is-impossible/

“Lionel Messi is impossible” by Benjamin Morris on Lionel Messi’s goal and assist statistics the last four years is the best player-focused piece of data-based analytical football writing I have ever read, and a perfect warm-up for tonight’s match. Short summary: Messi is an outlier in nearly every desirable offensive way. Above is one example.


“[…] by pushing his jam always forward into the future, [the “purposive” man] strives to secure for his act of boiling it an immortality”

John M. Keynes thus criticized an excessive preoccupation with the future in his essay Economic possibilities for our grandchildren (1930). I was a bit puzzled by this, and the full quote does not really help:

The “purposive” man is always trying to secure a spurious and delusive immortality for his acts by pushing his interest in them forward into time. He does not love his cat, but his cat’s kittens; nor, in truth, the kittens, but only the kittens’ kittens, and so on forward forever to the end of cat-dom. For him jam is not jam unless it is a case of jam to-morrow and never jam to-day. Thus by pushing his jam always forward into the future, he strives to secure for his act of boiling it an immortality.

Helpfully, there is a Wikipedia page on the “jam tomorrow“.  It turns out that the jam reference comes from Lewis Carroll’s Through the Looking Glass (1871), in which the White Queen offers Alice to work in exchange for jam that she (Alice) will always receive tomorrow, i.e. never. Back to Keynes: Such forward-looking behavior is helping to solve the economic problem, but as soon as that is done (it will take at least 100 years), we can stop pushing the jam into the future.

(And presumably start eating it, though Alice says she does not care for jam, but perhaps that is another story.)

Enhanced by Zemanta

Did Manchester United’s decline start under Ferguson?

Carl Bialik (previously known as The Numbers Guy) at 538 argues that Manchester United was exceptionally lucky in Alex Ferguson’s last season, and that David Moyes just kept steady a downward trend. Bialik’s point is that looking at finer measures of performance than simply points, such as the share of shots taken and the goal differential, United should not have done as well as ghey did last year, so Moyes should not be blamed for the (apparent) decline this year. There is a nice graph of the development of United’s share of shots taken the last 12 years in the post showing this clearly.

How the decentralization of technology work against a “surveillance state dystopia”

The coming of a sureveillance state dystopia has been predicted for some time. Ramez Naam writes a guest post at Charles Stross’ blog, and claims that the decentralization of technology has been responsible for the postponement. E.g., getting away with photoshopping images is a lot harder today than in Stalin’s time.

Naam spells out three technological trends that will help the little man even further: 1. Cheap cameras for self-protection. “[Camera] technology, when expensive benefits the big players. The technology getting cheaper becomes distributed, benefiting the citizenry.” “2. Crypto and Anonymity Blunt Surveillance Tools.” If someone is not looking for you in particular, anonomity tools are quite effective. 3. Information is becoming easier to spread. Naam ends by emphasizing that these trends will be no panacea, we will still need the law and proper oversight.

The effect of football on work motivation and well-being

“Is soccer good for you? The motivational impact of big sporting events on the unemployed” is an article in Economic Letters (ungated) by Philipp Doerrenberg and Sebastian Siegloch at IZA that I believe a lot of people wished they had written. The authors analyze the effect of the Euro Cup and the World Cup on the unemployed in Germany:

We examine the effect of salient international soccer tournaments on the motivation of unemployed individuals to search for employment using the German Socio Economic Panel 1984–2010. Exploiting the random scheduling of survey interviews […] We show that respondents who are interviewed after a tournament have an increased motivation to work but, at the same time, request higher reservation wages. Furthermore, the sporting events increase the perceived health status as well as worries about the general economic situation. We also find effects on the subjective well-being of men.

The unemployed are made more motivated to work and more worried, and to perceive themselves as being healthier, but men’s well-being is decreased. Ht: Kevin Lewis.