The Economist revisa “Capital in Twenty-First Century”

Por más de 5 semanas (27 de Febrero – 25 de Abril), The Economist publicó un análisis  por partes del bestseller de Thomas Piketty, “Capital in Twenty-First Century“.  Su lectura es un buen acercamiento a la obra del economista francés, además  que permite conocer la posición del semanario inglés.

Aquí puede revisarse la serie completa.


LAST year Thomas Piketty, an economist at the Paris School of Economics and a renowned expert on global inequality, published a book titled “Capital in the Twenty-first Century”—in French. It will be released in English on March 10th. We reviewed the book earlier this year, but it is detailed and important enough, in our opinion, to deserve additional discussion. We will therefore be publishing a series of posts over the next few weeks—live-blogging the book, as it were—to draw out its arguments at slightly greater length. Starting today, with the book’s introduction.

Capital, as I will refer to Mr Piketty’s book from here on out, is an incredibly ambitious book. The author has self-consciously put the book forward as a companion to, and perhaps the intellectual equal of, Karl Marx’s Capital. Like Marx, Mr Piketty aims to provide a political economy theory of everything. More specifically, he attempts to re-establish distribution as the central issue in economics, and in doing so to reorient our perceptions of the trajectory of growth in the modern economic era. Mr Piketty’s great advantage in attempting all this, relative to past peers, is a wealth of data and analysis, compiled by himself and others over the last 15 or so years.

Mr Piketty begins in an introduction that proceeds in two parts. He first describes the intellectual tradition into which the book falls. The second, which is the basic outline of his theory, I will tackle in the next post.

The study of political economy emerged in the first decades of the Industrial Revolution, in the late 18th century, in Britain and France. The great thinkers of the era were attempting to understand the dramatic societal and economic changes of the day and to describe their mechanics in a way that would allow them to anticipate future developments. To a great extent they focused on distributional issues—and worried that distribution spelled serious trouble for the capitalist system. The Reverend Thomas Malthus, for instance, famously worried that overpopulation would drive down wages to subsistence level, leading to dangerous political upheaval. To short-circuit this possibility the compassionate reverend recommended that governments cut off assistance to the poor and limit their reproduction.

David Ricardo’s 19th century analysis was more measured but nonetheless similar in its concern about the sustainability of the contemporary economic system. He focused his attention on the relative scarcity of factors of production, and the effect of scarcity on shares of national income. Output and population were rising fast, he noted, while land supplies remained fixed, suggesting that land prices might rise without bound. As a result, he speculated, land rents would come to eat up a steadily rising share of national income, threatening the capitalist system.

Ricardo was wrong in the long run—soaring agricultural productivity (which both he and Malthus failed to anticipate) meant that agricultural land was not the scarce factor for very long. But he was right in the short run, and the short run matters. A period of a few decades in which the price of a scarce resource soars can lead to enormous accumulation of wealth in the hands of relatively few owners of capital. That concentration can persist even after technological change eases the initial scarcity: a point Mr Piketty notes is relevant in thinking about soaring prices for urban property or natural resources.

And then there was Marx. He (along with Friedrich Engels) was the first of the great political economists to wrestle directly with the effects of industrial capitalism. Marx was reacting to the reality of industrial growth at the time: through the first century or so of industrialisation output grew steadily, but there was virtually no meaningful increase in real wages. In the “hungry 1840s”, when the Communist Manifesto was published, capitalism seemed like an incredibly raw deal for workers. That had begun to change by the time Marx published the first volume of Capital, in 1867. But the emergence of steady wage growth did little to diminish concentrated wealth.

Marx saw capitalism as fundamentally flawed, containing the roots of its own destruction. As owners of capital gobbled up the gains from growth, they would accumulate still greater piles of capital—”infinite accumulation”. This would either drive the return on capital down to nothing, leading the capitalists to destroy the system by battling it out with each other, or it would allow the capitalists to capture a rising share of national income (like Ricardo’s landowners), leading the workers to revolt. But Marx also turned out to be mistaken. He did so in large part, says Mr Piketty, because of a lack of data, and because he and others failed to anticipate that rapid technological growth could reduce the relevance of past wealth accumulation.

This latter factor helps shape one of the main elements of Mr Piketty’s theory of everything: that the rate of growth is hugely important in determining how long a shadow old wealth casts. If not exactly an equaliser, fast growth nonetheless puts a finger on the scale on the side of those without great wealth.

Now, this entire line of theoretical work was thrown into upheaval by the events of the period from 1914-1945. The chaos and policy shifts of the period wiped out much of the world’s previously accumulated wealth and set the stage for a burst of rapid, broad-based growth. Meanwhile, economists were for the first time gathering detailed data on personal incomes. And so when Simon Kuznets began looking at inequality trends in the 1950s, the data suggested to him that in “advanced phases” of capitalist development inequality tended to fall. The idea that inequality rose and then fell as an economy developed became known as the Kuznets curve. For the first time hard data had been brought to bear on distributional questions, and the news seemed pretty good. Kuznets’ view became the foundation from which modern economics approach distributional issues, despite the fact that it was based on a very limited period during which declining inequality could not remotely be considered the result of natural economic processes.

That, as Mr Piketty sees it, is where he comes in. For most of the 20th century the distribution of incomes was a minor issue within economics. Growth and management of the business cycle were the sexy economic issues.

That is now changing, based in part on the academic work Mr Piketty has done (much of which is accessible at the World Top Incomes Database). While emerging-market growth has narrowed global inequality, income inequality within countries, including many large emerging markets, has been rising. The return of the importance of scarce land, resources, and intellectual property has contributed to a resurgence in wealth accumulation. Distributional worries are back, and Mr Piketty argues that that is the natural state of affairs rather than an aberration.

“In a way”, Mr Piketty writes, “we are in the same position at the beginning of the twenty-first century as our forebears were in the early nineteenth century: we are witnessing impressive changes in economies around the world, and it is very difficult to know how extensive they will turn out to be…”. It can be frightening and disorienting to find oneself at such an economic juncture. We look back on Malthus’ worrying with the smugness of hindsight. But those who read Malthus at the turn of the 19th century and concluded that industrialisation would result in political upheaval, misery, and war turned out to be right. It also led to quite a few good things. But the process of getting there has been anything but smooth, and distributional issues inevitably play a role when the march toward prosperity slows or beats a temporary retreat.

This is an important book arriving at an appropriate moment. It’s also an entertaining and informative read, and so I hope you’ll join me in reading through and discussing it.

Introduction (continued)

In the first half of Capital’s introduction, Thomas Piketty lays out the intellectual background for the work. In the second he shares the main results of the book. The first is that levels of inequality represent the result of political choices, rather than deterministic technological or economic outcomes (a point reinforced by recent IMF research). Whether or not structural economic shifts generate rising or falling, low or high inequality is down to the structure of the political system and the way it chooses to accommodate those changes. “Skill-biased technological change” or “superstar economics” are incomplete explanations of rising inequality. They may actually leave out the more interesting half of the story.

The second result is that economies do not naturally evolve toward more equal distributions of resources as they mature. There are some forces pushing toward greater equality, like the spread of new technologies from rich areas to poor—what he calls “the principal force for convergence”. And there are some forces pushing toward less equality, one of the most important of which is the ability of the rich to secure further economic and political benefits for themselves. Importantly, he notes, the equalising power of the diffusion of knowledge is closely linked to state policy: to investment in infrastructure, education, research, and a regulatory environment conducive to entrepreneurship and competition. It isn’t a natural force for convergence at all, but must be actively cultivated (and may be resisted by those with wealth and power).

At the moment, Mr Piketty observes, the world is looking at two key dynamics pushing the world toward greater income divergence. One is soaring inequality in labour income. This trend is especially noticeable in America, where the top 10% of earners now captures close to 50% of national income, up from about 35% for the first three postwar decades. The second dynamic is the return of wealth.

Mr Piketty introduces a statistic that features prominently throughout the book: the ratio of private wealth in an economy to GDP. Across most rich countries this ratio was consistently high in the 19th century, tumbled during the interwar period, and has since rebounded back within shouting distance of 19th century highs. To explain this, Mr Piketty unveils a “fundamental force for divergence”: r>g.

The r in r>g is the return on capital. It’s worth mentioning now that capital in Capital is equivalent to wealth, and wealth means anything other than labour which generates income: land, financial assets, physical capital, and so on. (Human capital doesn’t count, as that augments labour income.) R, then, is the income generated by that wealth (meaning rents, dividends, profits, and so on) as a share of total wealth. The g, on the other hand, is simply the growth rate of the economy.

If and when r>g, then, wealth grows faster than output, the ratio of wealth to output increases, and the share of capital income in total income rises. And there is an important corollary: a slowdown in overall economic growth is itself a force for greater concentrations of wealth. This relationship forms a critical part of the book’s argument, though it is not necessarily intuitive: shouldn’t lower g imply lower r? But we’ll get to that. We should be clear, though, about what Mr Piketty is after: breaking much of modern economic history down into a few constituent forces that can be easily captured in statements like r>g. The value of his book will depend, in part, on how much we need to abstract away from the world to accept that r>g matters across times and places. The more reality you strip away from a model the easier it is to divine universal truths, and the less interesting the truths become.

It’s worth mentioning that Mr Piketty very much wants to explain the real world. He closes the Introduction with an interesting discussion of his motivations, not easily summarised here, which includes the factoid that he left America to return to his native France because he found American economists too interested in theory and generally unconvincing. That’s not particularly charitable—there are a lot of American economists working on a lot of different things! It is of a piece with the francophilia that suffuses the book, and which the author (rather unsuccessfully) argues has nothing to do with home-side bias. Nothing wrong with that, of course, but it is a recurrent and noticeable enough part of the book that it seems worth pointing out.

Chapter 1

FIRST on the agenda, a bit of housekeeping: Capital is now available for sale. Second on the agenda, this week’s discussion, which is a look at Part 1, “Income and Capital”, beginning with the first Chapter. I don’t anticipate devoting a post to each chapter of the book, since that would keep us here a while, but in the early stages it’s worth doing, to keep the posts at a manageable length.

Mr Piketty begins Part 1, Chapter 1 by defining terms. It’s a useful exercise, but for our purposes the only thing worth calling out is the definition of capital itself, which he treats as equivalent to wealth, and which he considers to be sources of value that can be traded (so “human capital” doesn’t count). He recognises that there is a difference between appropriatable wealth (like land or natural resources) and accumulatable wealth (like financial or industrial capital). There are times when it’s useful to distinguish between the two, he notes, and times when it isn’t necessary. Mr Piketty writes that in most of the rich world wealth is split nearly evenly between residential capital (housing mostly) and productive capital. A useful rule of thumb.

He then defines an important statistic for the book, β, or the ratio of capital in an economy to national income. It is a measure of the importance of capital in a society, and in Part 2 he describes why and how it varies across times and countries. And this all leads up to what he calls, rather grandly, the “first fundamental law of capitalism”: α = r * β. This is basically an accounting identity that defines capital’s share of income (α) as the rate of return on capital multiplied by the capital stock. Which makes it a useful way to calculate a quick and dirty rate of return, assuming you know the capital share of income and wealth as a share of output.

At this point I should mention another interesting habit of Mr Piketty’s which some readers may find annoying, but I which I found quite endearing: his penchant for illustrating 19th century trends with examples from contemporary fiction. It turns out to be quite a useful device, as it happens, because many of the writers at the time (like Jane Austen, for instance) lingered over the details of characters’ estates and took for granted (and understood that readers took for granted) that so much wealth was associated with x annual income, implying y rate of return. It’s especially fun later when he discusses changes in the macroeconomic framework (toward systematic inflation, for instance) and identifies corresponding changes in the obsessions of contemporary writers. But I digress.

So does Mr Piketty. He wanders into a history of national accounts in order to make the point that figures like GDP are social constructs—which is true but somewhat tangential. He then discusses global inequality; real income per person ranges from about 150 euros per month in the poorest regions (like sub-Saharan Africa) to 3,000 euros per month in the richest. The geographical distribution of global output has been changing, of course. Europe’s share peaked on the eve of the first world war, America’s in the 1950s.

It does matter that he mentions this, I should say, since one classic method for deflecting expressions of concern about inequality is to point out that at a global level inequality has, at least recently, been falling. And Mr Piketty does have an interesting story to tell about global convergence. Those countries that have successfully completed much of the economic catch-up process have typically been economies that self-financed industrialisation on the back of high domestic savings rates. But in other regions, like Africa, a large share of industrial capital is foreign-owned. That may be due to the fact that domestic institutions, including the financial sector, are weak. But foreign ownership can alsoperpetuate institutional weakness, he writes, since it creates a strong incentive for governments to break contracts and expropriate foreign capital.

Mr Piketty uses this to argue that while there are gains from openness, those gains are almost entirely down to the transfer of knowledge, rather than the efficiency benefits of free trade and capital flows. The latter are mostly real, he suggests, but are generally modest. Or to put it somewhat differently: access to global goods and capital markets is useful to the extent that it facilitates an improvement in an economy’s technological capabilities.

That’s excessively pessimistic on openness for my taste. Maybe openness mostly entails foreign ownership that undermines institutional strength and delays convergence (counterpoint: maybe it doesn’t!), but cutting the poor off from goods markets has historically been a good way to keep them poorer than they need to be and to reinforce cronyist regimes.

That said, Mr Piketty’s emphasis on the importance of the diffusion of knowledge in influencing income distributions (across time, and both across and within countries) is dead on. As we shall see in later chapters.

Chapter 2

One of Capital’s primary themes is that economic states we conventionally view as the norm are in fact historical abberations. Mr Piketty launches his book by saying that the natural tendancy of economies to become more equal as they mature is a myth, built on the unusually compressed distributions of incomes and wealth that prevailed in the middle of the last century. That period was actually an oddity that resulted from the unique historical circumstances of the tumultuous early 20th century; most of the time inequality is the norm rather than the exception.

In Chapter 2 Mr Piketty extends this revisionism to ideas about growth. The middle of the last century was unusual in its growth rates as well as in the distribution of income; the good times most of us see as our due as residents of rich economies were in fact a fleeting anomaly. Most readers will not be surprised to hear that growth prior to the Industrial Revolution was extremely slow. Mr Piketty argues that even within the industrial era growth has typically been slower than was generally the case in the postwar boom decades.

The growth analysis in Capital is built on a division of growth into two components: population growth and per capita growth. While that’s as defensible a method of growth accounting as any other, it’s worth remembering that the strict division is artificial; in practice, population growth rates and productivity growth rates influence each other. We will discuss later whether Mr Piketty relies too heavily on an inappropriately rigid distinction between the two.

But here’s the broad point: over the last 300 years, economic growth has been roughly half attributable to growth in population and half attributable to growth in productivity. That is important, because the world is on the downslope of the great demographic convulsion of the past few centuries. Population growth rates soared from 1700 to the middle of last century, when global population growth peaked at an annual rate of 1.9%. But population growth rates are now falling and are expected to return to very low, pre-industrial rates by the end of this century.

Similarly, the rate of growth of per capital income also appears to be near what is likely to be a peak. In the 18th century output per person grew imperceptibly faster than in the long centuries of almost no growth before. In the century to the first world war growth sped up to about 0.9% per year on average (across the world as a whole), and in the century to 2012 growth averaged 1.6%. In the very recent past rapid emerging-market catch-up pushed the global rate of per capita growth above 2%, but that seems unlikely to be sustained. Mr Piketty sees forecasts from economists like Robert Gordon, who thinks a return to pre-industrial rates of per capita growth may be ahead, as too dire. He nonetheless thinks that global per capita growth will converge toward 1% by the end of the century.

Taking the two trends together an interesting picture emerges. In the long centuries leading up to Industrial Revolution total economic growth averaged no more than 0.2% per year. But global growth rates soared to an average of as much as 4% per year over the past 60 years. Yet a subtle deceleration has begun, which will ultimately bring global growth back to something like 1.2% by the end of the century.

Why does this matter? Because, Mr Piketty says, of the power of cumulative growth. At growth rates of 0.2% per year the economy expands by just 6% per generation, and by only 22% per century. In effect, society recreates itself almost unchanged, generation after generation. Culture, society, and class structures are stagnant over long periods of time. At 3.5% annual growth, by contrast, each generation has an economy 2.8 times larger than the last, and a century means a 31-fold increase in economic output. That means dramatic social change and the constant replacement of the old with the new. That was the world of the middle of last century.

And in between? At a growth rate of 1.2% each generation enjoys economic output about 50% larger than the previous, and a century leads to a three-fold increase in output. That is not nothing. Over the course of a millenium the resulting change is unimaginably significant. But at human timescales the permanence of society—its rigidity—is in many ways more similar to that of the pre-industrial era than the relatively recent past.

Growth is important for lots of reasons, but it is important for Mr Piketty’s purposes because it governs the length of the shadow cast by the past on the present. As growth rates fall, that shadow will lengthen, strengthening the economic and social importance of past wealth and status.

But is that right? Is growth actually about to fall dramatically? And can we be sure that slowing growth in this modern era will have anything like the same economic and social effects of low growth rates in the pre-industrial era? We’ll move on to Mr Piketty’s evidence next week.

Chapter 3 and 4

WE ARE picking up the pace a bit now, tackling two chapters at a stroke. In Chapters 3 and 4 Mr Piketty describes the evolution of capital over time and across the large economies of North America and Europe. There is a lot of interesting detail, but the broad picture is relatively straightfoward. As of the early 19th century wealth consisted mostly of agricultural land holdings. Over time the importance of such land fell to almost nothing while housing and other domestic capital (including commercial real estate and industrial capital) became dominant. As we have already discussed, stocks of capital relative to national income were high during industrialisation, fell dramatically in Europe from 1914 to 1950, rebounded over the past 60 years and are now approaching prewar levels. What’s more, capital has never been as important in North America as in Europe, thanks to the low cost of abundant land and the face that North American economies basically started afresh (roughly) two centuries ago.

In this post, however, I want to focus on one particular thread of analysis that runs through these two chapters: the division between public and private capital.

National capital, as Mr Piketty has it, is the sum total of wealth in the economy, which he presents as a share of national income. One can divide that up into public and private wealth, and the distribution between the two can move around without necessarily changing the total. So for instance, in the early 19th century the British government racks up enormous debts fighting wars, which ultimately rise to about 200% of national income. But during this period national capital relative to national income doesn’t much change. What does change is the distribution between private and public; public capital falls as the government borrows from wealthy Britons, and there is a corresponding rise in private capital.

The trend in the decades after the second world war is quite different. National capital is rising during this period, from the postwar nadir. But private capital is not; indeed, it actually falls a bit relative to national income from the 1950s to the 1970s. Public capital, by contrast, rises sharply. There are two things going on here. First, the government’s net asset position is improving as it inflates away its debt, and secondly the government is creating a rather large public sector. The point is that government policy can influence the level of national capital, but also its distribution and its salience. And the distribution and salience matter tremendously for how policy is made and how the benefits of growth are shared.

The British case is again instructive. Britain has had two dramatic public-debt peaks in its modern history, in the early 19th century and the mid-20th. But between these two eras there is a major shift in the nature of public debt—from a tool of private wealth to benefactor of the poor—which is rooted in the return to lending to the government. Changes in that return are driven mostly by a shift in the behaviour of inflation—and one begins to get a sense of the interrelationship between the interests of rich and poor on matters of public finance and inflation.

Britain addressed its debt very differently in these two periods. The debt of the 19th century was managed down over the course of a century, by running surpluses larger than the state’s education budget in an essentially inflation-free environment. This is the world of the rentier, of a rich elite collecting a reliable and substantial real income from their holdings of government debt. This is an era, Mr Piketty notes, in which Marxists looked with suspicion upon government borrowing, which they saw as a means to funnel resources to the rich.

By contrast, the debts of the interwar period were wiped away remarkably quickly, thanks largely to financial repression and inflation—that is, through sharply negative returns to bondholders (“the euthanasia of the rentier”). Public borrowing, meanwhile, became an important mechanism for macroeconomic management, designed to limit economic hardship among working people.

Inflation, Mr Piketty notes, is a crude method for getting rid of debt. Over the course of the 20th century, the ranks of creditors came to include plenty of non-rich individuals, both through direct saving and via large pension funds. It is not surprising that support for disinflation in the 1980s ran well beyond the elite. But one might also note that the end of inflation coincided with an acceleration in the rise of national capital, with growing inequality and with rising indebtedness. It is interesting to consider how class power structures influence inflation, and how inflation influences class power structures, and how both influence the distribution of economic pain and gain.

The rentier has not exactly returned to modern Britain; George Osborne is working to return the government to surpluses, but he has not pressed the Bank of England to ring inflation out of the system. Power structures and political choices have been somewhat different in Europe and America. But each case illustrates one of Mr Piketty’s key themes: that the demotion of distribution as a critical economic issue was a mistake. Debt has a way of making that clear; or in Mr Piketty’s phrase: “Debt is the vehicle of important internal redistributions when it is repaid as well as when it is not.”

Chapters 5 and 6

NOW we arrive at the Second Fundamental Law of Capitalism. Ready? Here it is: β=s/g. Or, the ratio of capital to income is equal to the savings rate divided by the growth rate of the economy. So if you have an economy that saves 10% of its income and grows at 2%, then in the long run it will have a ratio of wealth to income of about 5 to 1. And we get a look at some of the dynamics Mr Piketty has in mind. Rising savings mean a larger stock of wealth to income, as does a slower growth rate. Indeed, the growth rate is hugely important here. If growth halves, from 2% per year to 1%, then according to Mr Piketty the stock of wealth will double to ten times national income (over the long run). The faster an economy grows the less large past accumulations of wealth loom.

This law is more like a rule of thumb; over long horizons wealth approaches the ratio of s to g, but all sorts of thing can disturb its path. And if there is some sort of fundamental shift in the economy, its impact on the dominance of capital may not be felt in full for decades. Mr Piketty also offers a pretty consequential caveat: if economies are reliant on scarce resources, or if asset prices tend to grow much faster than consumer prices, wealth can grow much larger much faster than this law implies.

This seems like a good place for a chart:

So, there you get a sense of the recent trend as well as the effect on variations around the trend of asset-price booms. This trend provides the context for Mr Piketty’s view of the relationship between growth and capital. Over the last 40 years the rich world has experienced a slowdown in both contributors to growth—population and technological progress—and that accounts for most of the broad, shared upward trend in wealth accumulation. It’s not the whole of it. As Mr Piketty notes, growth in private wealth has been boosted by declines in public wealth, via privatisation of public assets and a shift to deficit financing. But it’s most of it.

Surveying long-run trends Mr Piketty reckons that wealth as a share of national income is trending back toward the levels last seen prior to the first world war, and will, across most of the world, come to rest between 600% and 700% of national income. But, he notes, levels of capital in high saving, low growth economies may end up rising much higher. Investors in those economies are likely to begin accumulating large net foreign asset positions (another throwback to the world before the first world war).

Having discussed trends in the stock of capital, Mr Piketty turns to a discussion of the return on those stocks. This return is critical, because it governs the share of income in an economy flowing to capital rather than labour. And the rate of return is principally determined by two things: the stock of capital (declining marginal utility applies, so returns fall as the stock of capital rises), and technology, which determines how useful capital is.

Mr Piketty writes a long and useful description of the historical battle within economics over whether the capital and labour shares in national income are constant. In fact they appear not to be (something that will come as no surprise to our regular readers). Indeed, over the very long run, he writes, the elasticity of substitution between labour and capital appears to be greater than one. What does that mean?

Well, additions to the stock of capital would tend to reduce the return to capital, as mentioned above. But based on that alone we can’t conclude anything about whether a larger capital stock leads to a larger or smaller income share for capital. The critical question is whether the drop in the return to capital is big enough to offset the effect of the rise in the stock. If a small rise in the stock of capital leads to a big drop in the rate of return (perhaps because technological progress is slow and there is nothing at all useful for new capital to do) then accumulation of new wealth reduces the labour share, and we get an elasticity of less than one. This was the way economies behaved in pre-industrial, primarily agricultural society.

But if a big rise in the stock of capital leads to a relatively small decline in the return to capital (perhaps because technological developments mean there are always new ways to deploy capital), then more wealth means more a larger income share for capital. This, Mr Piketty reckons, describes the world in which we live now. From the 1970s on, the stock of capital has been rising steadily. And so, too, has capital’s share of income.

There are two broad questions raised by these chapters that I think loom particularly large. One is what the path for capital accumulation and capital income looks like in a world of slow demographic growth and steady (or perhaps reasonably rapid) technological growth. Mr Piketty focuses on the demographic side of things, which argues in favour of rising wealth. He mentions the possibility that technological progress may increase the demand for human capital, which will augment the labour income share and push against the demographic trend. Indeed, this is one interpretation of the rise in wages and stabilisation of the capital share in the late 19th century, after long decades in which the benefits of industrialisation overwhelmingly flowed to capital.

Regular readers will anticipate my concern: that technological progress will in fact push in the same direction as demographic change, by magnificently expanding the array of tasks capital is capable of doing. In that world, labour could only retain its seat at the table by accepting tumbling wages. Whether labour hangs on in such fashion or is brushed aside through automation, the implications for labour earnings are grim.

The second big question is: in this world of rising capital stocks and income shares, what is the scarce factor, which is able to capture the benefits of economic growth? Not financial capital, that’s for sure. Intellectual property is a better candidate; think of the entrepreneurs behind Instagram, capable of realising a vast fortune on the back of a fairly straightforward idea. And then there’s this:

Where do the rich, powerful, and productive work and play? For the most part, they do it in a handful of very expensive, very well-educated, very cosmopolitan global cities. People in those cities have access to social networks and consumption amenities that cannot be had at any price in other locations. Such cities may be the ultimate scarce factor and the owners of land in those cities the ultimate beneficiaries of these broad trends.

Chapter 7,8, and 9

IN PART 3 Mr Piketty turns to the heart of the matter: inequality and the concentration of wealth and income. The first three chapters of the section are generally focused on trends in the concentration of income. Here is a summary for you. For a time in the mid-20th century, hard work was the surest way to obtain a good living. For most of the modern era before that, it wasn’t; only rarely did the fruits of one’s labour elevate him to the ranks of the elite, which was instead populated mostly by the idle rich. (America outside the South was an exception, but became less of one over the course of the 19th century.) Then the dramatic changes of the interwar era reset the clock.

This reset has much less to do with rising worker productivity or bargaining power than to the blows dealt to the very rich in the interwar period. Extremely unequal societies used to be built upon extreme concentrations of wealth, which allowed a small class of people to live on capital income alone. In the interwar period, the rentiers were “euthanised”, in Mr Piketty’s phrase; they were not merely overtaken by the working rich.

Mr Piketty devotes much of the real estate in these chapters to a detailed walk through the data, explaining which factors contributed to shifts in income distributions over the course of the last century. The French income distribution was highly compressed by the interwar years, but inequality began to grow rapidly in the postwar era, thanks to rapid recovery and a greater focus on rebuilding than on distributional issues. That trend toward rising inequality was snuffed out by the political changes of the late 1960s, but inequality began growing again from the early 1990s.

Indeed, the broad point is that almost everywhere in the world that data on top incomes is available—including emerging markets—inequality hit something of a nadir in the 1960s or 1970s but has since begun rising again. But the rise is most pronounced in anglophone economies, and it is basically unprecedented in America, which is blazing new territory. Basically unique among rich economies, America has returned to and actually surpassed the levels of income concentration experienced at the beginning of the 20th century. Strikingly, it has done this through a remarkable increase in labour-income inequality.

In discussing the American experience, Mr Piketty makes two points that have sparked some argument in recent weeks. The first concerns secular stagnation. Here’s your secular stagnation and crisis, according to Mr Piketty: in 1980, roughly one-third of national labour income went to the top 10%. Over the next three decades that share rose by 15 full percentage points. That represents an enormous shift in purchasing power to those with much less propensity to spend, and that, in turn, means that adequate demand became ever harder to generate in the absence of borrowing by those more likely to consume the marginal dollar. In terms of magnitudes, this shift in incomes is significantly larger than the contemporaneous rise in America’s current-account deficit. As Mr Piketty sees things, inequality is a bigger factor in stagnation than global imbalances.

The source of controversy is that one manifestation of inadequate demand as a result of excess saving is a falling—or even negative—real interest rate. And if the real interest rate is falling, critics argue, then how can Mr Piketty’s r be greater than g, the growth rate, thereby raising the value of wealth to income in the economy?

But this glosses over the rate of return we are interested in. The short-term real interest rate on risk-free debt has been near or below zero for much of the past decade. The long-term real interest rate on risk-free debt has been falling for a couple of decades, and it has been low enough to dip below the economy’s actual or potential real growth rate since about 2005. But Mr Piketty is not interested in the risk-free rate. His rate of return on capital is the sum of all income derived from wealth each year (which includes profits, rents, dividends, royalties, and so on) as a share of national wealth. As he notes in earlier chapters, the historical r was an amalgam of different returns on different sorts of investment carrying different levels of risk; the return on agricultural land in Europe was different from that on government bonds, and both were lower than the return on riskier investments in industry or ventures abroad. While it is true that putting one’s money in American government debt has rarely paid less, it is not at all true that the wealthy are unable to wring real capital incomes larger than the rate of growth out of their fortunes.

The second contentious point relates to the discussion of the sources of American inequality. Over the long run, Mr Piketty says, the supply and demand for skills is critical in determining the distribution of labour income. Over shorter horizons and smaller margins policies like the minimum wage matter. But to explain the extraordinary performance of the incomes of America’s top 1% requires a different story. Recourse to “superstar” explanations gets you only so far, since other similar economies, including Britain, have experienced a rising top income share but nothing remotely as dramatic as that in America. Mr Piketty reckons that one needs to turn to norms at the very top. Productivities are hard to assess among top executives, and salaries are often determined by sympathetic boards or supervisors or peers, who at any rate share similar ideas about what top executives are worth. In America, this peer group votes itself massive raises that would be considered obscene in other advanced economies.

That may not be right. But Mr Piketty notes that if America’s current income gaps reflect actual productivity differences, then the dispersion in productivity at the top and bottom of the spectrum is greater in modern America than in apartheid South Africa. And it is hard to believe that could be the case.

Chapter 10, 11 and 12

THIS week the Financial Times columnist Martin Wolf gave us his review of Mr Piketty’s book. It was glowing but had one main criticism: that “Capital” does not tell us why inequality matters. I don’t believe I agree with that. It is in these chapters that Mr Piketty explains why we should be concerned about rising inequality.

As we have discussed several times, a key part of Mr Piketty’s story is that the rate of return on capital, r, is greater than the rate of economic growth, g. Critics have suggested that this is at odds with economic theory, but Mr Piketty is not making a theoretical point; he is observing that, empirically, r has been greater than g throughout most of history. If the theory is inconsistent with that, then that is the theory’s problem.

Mr Piketty does offer an intuition for why this should be so. If r is less than g, he writes, then incomes rise faster than the burden of debt, and people have an incentive to borrow and consume without bound, knowing that it will be trivial to repay the loans. But people can’t borrow endlessly without pushing up the rate of return on capital to at least the level of the rate of growth.

In any event, the regularity r>g is a strong force for concentration of wealth. Because the distribution of wealth is always more unequal than the distribution of income, a dynamic in which wealth grows faster than income leads to ever greater concentration of resources in the hands of a few. National wealth grows relative to national income, capital income grows relative to labour income, and the rich get richer.

But so what? Mr Piketty’s answer is that this state of affairs is simply not sustainable, and however it ends, it ends badly. It could be that the process will peter out on its own as massive accumulation of capital eventually pushes down the rate of return. But this process can take a very, very long time and result in intolerably high concentrations of wealth. Or it could be the case that the process stops when a relatively small group of individuals owns everything. Either way, the danger is that society rejects the injustice of these concentrations and reacts—perhaps violently.

The main problem is a meta-problem, in other words. Inequality matters because, like it or not, inequality matters. In most states of the world, inequality will tend to rise unless countered, by economic shocks or deliberate policy choices. Active concern over and management of inequality may help reduce the odds that society rejects as unjust the institutions underlying an economy, potentially in chaotic and violent fashion.

That doesn’t seem like a particularly outlandish view of the world. Humans intuit that—whether one is born into a fortune or one earns it in the market—there is only so much credit one can take for great wealth. Birth is the ultimate lottery, and it is wrong for the stakes of that lottery to be excessively high. Given the economic realities that Mr Piketty describes, it is therefore inevitable that distributional concerns will arise and motivate policy. The strange thing—as Mr Piketty points out—is that we consider the return of distribution as a worry to be an unusual or unnatural thing.

Part 4, Conclusión and recap 

AND now we come to Part 4, which focuses on the critical question of how to structure policy in the world Mr Piketty describes. As this section contains some of the book’s weaker arguments, this seems like an appropriate place to end our discussion with a grand assessment of the work’s biggest contributions and shortcomings. I’ll start with the former.

So what are the book’s main contributions? First and perhaps most important are the data. As impressive as “Capital” is, it may ultimately prove less influential and significant than the World Top Incomes Database, on which much of the book is based. That, and the data on the distribution and evolution of wealth, represent an enormous achievement and the basis for the narrative. It’s worth pointing out again that Mr Piketty has been just one of many economists working to pull these figures together. Were it not for this effort, the ongoing discussion on inequality would not be as serious and relevant as it is.

Second, the book challenges conventional wisdom concerning the economic history of the rich world in several important ways. Once again, Mr Piketty is not alone in demonstrating, for instance, Simon Kuznets’ view that inequality would rise and then fall as industrialisation proceeded, or that the shares of national income flowing to capital and labour are not constant. But Mr Piketty has given these ideas new prominence, and with them the view that the 20th century’s dramatic compression of wealth and incomes was largely down to the one-off shock of the interwar era. Putting this all together, Mr Piketty’s book goes a long way toward challenging the 20th-century orthodoxy that distribution is not particularly important.

Third, “Capital” provides a framework for thinking about how inequality might evolve in future. Mr Piketty’s data give us a view of the past. He also gives us his thoughts on how things might unfold in future (albeit with plenty of caveats). But even if readers doubt his forecasts for the rate of return on capital or for economic growth, they will have a way to think about how key distributions will change, thanks to this book. Among pundits, policy discussions have already begun to reflect this: the distributional effects of possible policy changes are beginning to be discussed in terms of how the policy might shift r or g (or s, the savings rate, or other key variables).

And fourth, “Capital”, by dint of its extraordinary success, has created a focal point for an important discussion. It is the right book at the right time, you might say. It has given the debate about inequality a boost, and it has provided that debate with a mechanical framework to help shape and deepen the discussion. It is undoubtedly an important book, which is why this newspaper has devoted so much space to its discussion.

Now, the shortcomings.

First, there have been criticisms of the book that Mr Piketty probably ought to have seen coming and addressed pre-emptively. Writing a book is hard, and the temptation is often to try to address every criticism a reader might possibly have. That can make for an unwieldy and less persuasive book. Given the success “Capital” has had, I’m reluctant to second-guess Mr Piketty, but it does seem as though it would have been worth dwelling a bit more on how his r relates to other interest rates. Although the secular stagnation discussion had not really got going when Mr Piketty was writing this book, interest rates on rich-world government bonds had fallen to historically low levels and looked like staying there for years to come. Given the thrust of his argument, that r is typically larger than g, the prospect of low rates for years to come ought to have merited some discussion.

Second, given the nature and potential severity of the problem he describes, the proposed solutions look pretty uninspiring. In Chapter 13, Mr Piketty briefly looks at the social welfare state and its prospects. Perhaps a much larger state would be a good thing at some point, he suggests, but not until bureaucracies can be reformed and shown capable of managing two-thirds or more of national output. In the meantime, he reckons, governments need to examine how they provide education for their citizens, and they should reform pensions. How? He writes, for instance, that, “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.” Well, all right.

In Chapter 14, he turns to income taxes. He states, interestingly, that progressive taxation is important in maintaining political support for globalisation. He observes that it was famously tax-antagonistic America that first toyed with top rates above 70%, first on income and then on estates. He argues that higher top tax rates were not particularly detrimental to growth. The main effect of their reduction was to encourage executives to bargain harder for big rises in pay packages, leading to soaring top incomes but not necessarily to faster growth. As a result, he says, higher top rates would be welcome, but he is not optimistic on this score; the egalitarian spirit and pressures of war that drove initial rises in top tax rates have faded.

And then in Chapter 15 he lays out his primary recommendation: a progressive global tax on capital. This, he freely admits, is a “utopian idea” which is unlikely to be adopted anytime soon. It is nonetheless a useful thing to talk about, he reckons, if only as a reference point for other policy proposals. He describes what a capital tax might look like, and he argues that alternatives (like communism, protectionism or capital controls) would be much more costly.

And that’s the gist of it. But surely there should be more. If the threat to society is that r is growing by more than g, then why not outline proposals to dramatically expand capital ownership? Why not propose a universal basic income: an inheritance, effectively, for everyone? The space allocated to Part 4 feels wasted.

And that brings us to the third shortcoming, which strikes me as the biggest. The economics gets a serious treatment in this book, but the politics does not. That’s somewhat ironic; Mr Piketty winds down his conclusion by saying that economics should focus less on its aspirations to be a science and return to its roots, to political economy. But theories of political economy should be theories of politics. And there is no r>g for politics in this book.

There are nods towards the importance of the interdependence between the political and economic. He notes that epoch-ending political shifts, like the French and American revolutions, were motivated in large part by fiscal questions. Similarly, he observes that progressive income taxation tended to emerge alongside the development of democracy and the expansion of the franchise. (Though, he also admits, the fiscal demands of the first world war deserve most credit for adoption of meaningful income taxation across the rich world.) And he discusses how concern about rising inequality (often among elites) helped motivate rising tax rates in America in the early 20th century.

But the ending the book deserved was another look back at the data, to see whether patterns in the interaction between wealth concentration and political shifts could be detected and described. That’s not Mr Piketty’s area of expertise, necessarily, but neither is most of the stuff in Part 4. And this really is the critical question. If the most likely outcome of the trends Mr Piketty describes is that somewhere down the line a left-of-centre government is elected and passes higher top income-tax rates, higher estate-tax rates and pension reforms, and that defuses the crisis, well, that puts the rest of the book in perspective. If the most likely outcome is revolution, well, that does too. And while it would be absurd to expect Mr Piketty to say definitely whether one possibility or another is bound to occur, I don’t think it’s asking too much, given the ambition of the rest of the book, to think we ought to be given some sense of his view on how social and political movements generally evolve in response to widening inequality, and how that evolution tends to be reflected in policy. What good is it to suggest utopian ideas about how to fix these problems without at least gesturing toward the political mechanisms needed to bring them about?

Maybe that is grist for “Capital 2: Capitalised!“. But it belonged here.

For my part, I would say I generally think Mr Piketty’s analysis is on the mark, with a few exceptions related to my particular view of how labour markets have been evolving. But “Capital” was critical to me in forming that view. I think the really fascinating dynamic in this book is the substitutability between labour and capital, something Mr Piketty mentions but does not devote an especially long amount of time to. But the stories we might tell in which capital becomes a dominant, potentially malign force are those in which many of the world’s workers can only hold off the automation of their work by accepting ever lower levels of compensation. But that’s my hobby horse and not his, and I don’t fault him for not devoting masses of pages to it.

At any rate, I am grateful for the opportunity to think through these issues and debate them, and to have views of mine upended. It’s not often that a book manages that, and this one did. I hope all of you, readers, have got something out of the book, and this series, as well.