Capital is so 19^th century, so aged and hobbling. Why do we still bother? Put another way, perhaps, why the spectacular return performance of a concept from two centuries ago, with a sort of Karl Marx revival as its advertising billboard? Former French President Nicholas Sarkozy allowed himself to be photographed leafing through the pages of Das Kapital; a German filmmaker, Alexander Kluge, made Das Kapital into a ten-hour movie; and a prominent Roman Catholic cardinal, Reinhard Marx, used his name to market a book addressed as an epistle to his 19^th-century namesake, predictably if somewhat unimaginatively entitled Das Kapital. And of course we have the Thomas Piketty phenomenon of this past year: a doorstop that seems to have achieved bestseller status because most of its readers were aging supporters of his conclusions.

In light of yet another effort to eclipse it, it is worth thinking about the origins and circumstances of capitalism as a concept. Capitalism started in the mid-19^th century as a term applied by revolutionaries and social critics to describe the transformation of work practices and the inhuman subjection of mankind to violence and abuse. By the end of the century it had become a simple description of a new, rationalized form of conduct in which taking a wage was the most normal thing in the world. In the 20^th century, capitalism again became a term of opprobrium, until a few businessmen thought it would be iconoclastic to present themselves as “capitalist tools.” Then, after the 2008 financial crisis, critics diagnosed (yet again) a “crisis of capitalism.” As a concept, capitalism hovers uneasily between the normative (in practice condemnatory) and the descriptive.

Describing capitalism is not so easy, however, and actually never has been. By now, after all the shifting around between opprobrium and celebration, no definition that can win broad acceptance seems possible. The mere attempt to define the term induces nervous handwringing. For instance, Louis Hyman, a young self-proclaimed champion of an allegedly new wave of scholars studying the history of capitalism, recently stated:

The essential problem is not to primly define capitalism like a schoolmarm, but to think about why capitalism, which appears to be so simple, evades easy definitions. And in the last decade, there has been a renewed interest among historians in not only challenging existing definitions, but in historicizing that very untidiness.¹

It should not be that hard to define capitalism. The Cambridge History of Capitalism, a recent volume edited by economists Larry Neal and Jeffrey G. Williamson, does precisely that in quite a satisfying way, as constituted by four elements: private property rights, contracts enforceable by third parties, markets with responsive prices, and supportive governments. The problem with the definition is that the first three aspects are found in many human societies—I’m tempted to say they are pretty near universal—and the fourth is very much open to debate. Some libertarians would claim that governments are by nature hostile to markets. Most commentators, however, see governments as crucial in providing a security framework, but that again is a characteristic feature of government, and Aristotle and his Indian and Chinese contemporaries (for instance in the famous first century BCE Chinese text Yán Tiě Lùn or The Discourses of Salt and Iron) were quite happy with that.

Fortunately, the charge of historicizing concepts can be stood on its head. Sven Beckert’s important new book, Empire of Cotton, stresses the constantly changing character of economic reality as the feature that makes capitalism so difficult to define:

The geographic rearrangement of economic relations is not just a noteworthy element of capitalism or an interesting aspect of its history; rather the constant shifting recombination of various systems of labor, and various compositions of capital and politics is the very essence of capitalism.

The evasiveness of the present generation of historical analysts of capitalism is in stark contrast to the certainties of previous interpreters. The British Marxist economist Maurice Dobb, whose oeuvre has just been resurrected in a splendid biography by Timothy Shenk, saw the formulation of a precise definition of capitalism as the core of his political mission:

If Capitalism does not exist as an historical entity, critics of the present economic order who call for a change of system are tilting at windmills; and Marx in particular, who was originally responsible for the talk about a capitalist system, was following a will o’ the wisp.²

Dobb and his generation tried to bring history and economics together—but because that same exercise was once done in the name of Stalinism, the effort soon looked rather discredited.

The study of capitalism, however, is necessarily an intertwining of history and economics. History is needed because the accumulation of capital—the essence of the character of capitalism—occurs over time, and over time things change. Accumulation is what gives the past power over the present. Economics is part of the story because it describes what happens to the historical legacy of accumulation in terms of production, exchange, and the distribution of wealth.

But the relationship of the two disciplines has been strained as the two ends of the definitional link have pulled away from each other.Economics has moved away from an interest in history in large part because the sophisticated techniques of the modern discipline need a corresponding sophistication and depth of data that is simply not available for past epochs. So because it is hard to investigate empirically, economists either give up altogether, or over-extrapolate from insignificant information, rather as if a future age were trying to reconstruct the present on the basis of random pages surviving from the Peoria telephone directory. Responding to this dismal failure of economists to provide historical analysis, French economist Thomas Piketty made many enemies in the economics profession by starting off his book on capital with a denunciation of economists’ “childish passion for mathematics and for purely theoretical and often highly ideological speculation.”

History also largely lost sight of economics for a mixture of reasons, all of them bad.³ Its practitioners were in part just not mathematically endowed enough to handle modern economics. Some historians had the notion that economics was a discipline that served to uphold rather than criticize (or “deconstruct”) the status quo. And many disliked the principle of reasoned thought that necessarily underlies any notion of discipline and preferred a self-consciously unstructured and decentered approach to what they often termed “unpacking” history.

The 2007–08 financial crisis, however, produced a new sensitivity to the significance of historical experience for economic interaction, as well as the broader issue of the long-term development of society. Some analysts even diagnosed a new disease, “financial amnesia.” Forgetting or ignoring past events had led, in the words of an anonymous editorialist, “risk to be mispriced, bubbles to develop and crises to break.” The remedy lay in formal requirements for investment professionals to study financial history.⁴

So there are some bold and novel attempts to bring the disciplines needed to comprehend capitalism closer together once more. My burden here is to examine two of the more prominent and successful of recent attempts, one from each side of the disciplinary divide.

Sven Beckert has written an engaging, highly readable, and comprehensive history of capitalism based around the commodity that he sees as essential to the early phase of modern capitalism: cotton. By the end of the 19^th century, it was the world’s most highly traded good. The power of his story depends on the way he consistently links the local with global developments.

There is an optimistic story of cotton that can be told, and it does make an appearance in Beckert’s pages. The availability of cheap and washable cotton textiles contributed significantly to improved health and thus greatly lengthened life expectancy. People no longer sewed themselves into woolen undergarments for a whole winter. It is an innovation, like the toothbrush, that we easily take for granted, but that brings immense gains. The revolution in textile production also meant that a much lower share of average incomes was spent on clothing. (There was a similar effect in food production around the same time.)

But there was a cost, and it was high. Beckert concentrates on the enormous dislocations that the modernization of cotton growing and production entailed: child labor and unendurably long hours in the textile mills, but, above all, the reliance on cotton farmed by slaves in the American South. After the Civil War, which is the epochal turning moment in this book, cotton production became globalized, with imperial states implementing carefully planned strategies to replace American slave-produced cotton with cotton grown in a Global South by peasants whose former ways of subsisting were being destroyed by a modern, brutal imperialism. The title of the book, empire of cotton, is thus peculiarly apposite.

Beckert distinguishes two types of modern capitalism: “war capitalism” and “industrial capitalism.” What he terms war capitalism was often discussed in earlier historical accounts as merchant or mercantile capitalism. But it elided into modern industrial capitalism: In the first half of the 19^th century, as he puts it, it “reinvigorated” war capitalism. Everywhere, he thinks, the role of violence, directed by the state, is visible in the brutal reshaping of social relations. Beckert is reacting against a tradition that sees the key role of the state and politics in securing abstract rules that made property secure, and hence made for transparency in economic transactions. But the mechanics of the transmission process in which capitalism used the state is never made completely clear: How were governments and ministers taken in by the cotton masters? And how does the idea of war capitalism, in which the state brutally directs the exploitation process, fit in with Britain’s energetic 19^th-century championing of abolition of the slave trade?

Beckert rather argues for the centrality of “war capitalism” by looking at the sources of capital for the introduction of capitalism. He points out that the pioneers of the British Industrial Revolution relied on extraordinary profits extracted from elsewhere. (That demonstration depends in large part on vignettes, not on aggregate data. That is the way historians often operate, and a human story makes the process intelligible and accessible.) Beckert focuses on Samuel Greg, the Manchester cotton pioneer who had “deep roots in war capitalism, its violent appropriation of slave labor, as well as its reliance on the imperial state to secure new technologies and markets.” He builds on this case to present a general aphorism: “In other words, the power they [the pioneers] harnessed in water was only possible because of the power harnessed by war capitalism.” Greg’s wife, Beckert writes, was “born into a family involved in the slave trade.”

Much of this is rhetorical assertion. Beckert does not pay attention to the precise calculations of economic historians such as Patrick O’Brien, who pretty conclusively demonstrated thirty years ago that the claim that European development rested on the exploitation of the periphery in 18^th century and in the first half of the 19^th century “foundered on the numbers.” As O’Brien put it, “exaggeration has occurred because the new history expands micro evidence into macro generalizations.”⁵

Beckert’s representative specimen, Samuel Greg, had the largest spinning and weaving business of the Industrial Revolution, but he was something of an outlier. The pioneers of the business were rather humble individuals with little wealth of their own. Richard Arkwright, who produced the spinning frame, was the youngest of 13 children, and his father was a tailor; Samuel Crompton, the maker of the “spinning mule”, was the son of a caretaker. In general, the cotton textile business required relatively little in the way of start-up capital. It did not require the giant resources that were later needed to build up steel mills—let alone the railroad networks or the large chemical or electrical plants of the later Industrial Revolution. Beckert gives more detailed figures for producers in Mulhouse, Alsace, where a cotton factory required a capital of 16,216 francs. A modest professional person’s annual income (for instance a lawyer’s) was 5,000 francs, so a bit more than three times that does not require a gargantuan partnership or the necessary partnership of the state.

Beckert’s analysis is a resuscitation of the vision of Marx: that capitalism necessarily began with a violent act of expropriation. At the same time that Marx was writing, Richard Wagner set the idea to music in the Ring of the Nibelung cycle, which begins with a creative theft as the dwarf Alberich steals a golden hoard from nature—the River Rhine. For Marx, the English enclosures of common land in the 16^th century, when a new gentry planted hedges to keep sheep as private property, or the even more violent process of clearing the Scottish highlands, marked the beginning of a process. Marx called this “primitive accumulation”, and he drew explicit theological parallels of the process that “plays in Political Economy about the same part as original sin in theology.” Marx asserted a clear identity emerging “between national wealth and the poverty of the people.” And the conclusion was presented with great emotional drama: “A good deal of capital, which appears today in the United States without any certificate of birth, was yesterday, in England, the capitalised blood of children.”⁶ Beckert uses the same kinds of suggestive images of the “blood-stained produce” to depict the immorality at the foundation of modern capitalism.

Industrial capitalism, in which there is a voluntary exchange or contract by which workers received wages for their labor, provided the next stage of capitalism. That was the new concept in the core countries, but in Beckert’s presentation the new factories still relied on raw materials provided elsewhere by violence and compulsion. The American Civil War and the end of slavery thus brought an epochal caesura: It seemed to threaten the end of cotton’s supremacy. But the large cotton merchants simply drove the model of coerced work elsewhere, to South America, Egypt, India, and other locales in Asia.

Illustrations work the same way as the human interest vignettes to underline the dismal transformation. The book begins with a drawing of a happy Aztec woman spinning cotton with what looks like a contented smile, and ends with a photo of grim and scowling children weighed down by their bags in the cotton harvest in modern Azerbaijan. Beckert’s account also ends perhaps with something of an exaggeration, speaking of 350 million people (or 3–4 percent of the world’s population) currently producing cotton, as a number “never before reached in one industry.”

Thomas Piketty has a simpler and apparently rigorous approach to the question of the definition of capitalism. For him, capitalism is the product of a long accumulation of wealth that acts as a controlling force in society. He starts with a clear definition: “Capital is defined as the sum total of nonhuman assets that can be owned or exchanged on some market.” The story of capitalism is thus obviously the story of the spread of markets, of exchangeability. It is a story driven by a long-run tendency—that the rate of return on capital is higher than the rate of economic growth, and thus that capital always accumulates.

This is not a new idea, but the way Piketty goes about demonstrating it is. He relies on a prodigious database. At first, he built it up in the case of his native France, where the egalitarian principles of the French Revolution required the assessment of wealth in order to levy a tax. There is also a substantial amount of historical data for the United States and the United Kingdom. One of the effects of the stunning success of his book has been that many emerging market countries are offering to open up their historical tax databases. His work is a triumph of the application of big data, a model for a new history of a type that really only became possible in the 21^st century.

There have been some gripes about this, particularly accusations of sloppiness or omission.⁷ One accusation is that the rise of inequality at the end of the 20^th century looks much worse in Piketty’s data because of the omission of the income derived from social security in old age and, indeed, all the other transfer payments of the modern welfare or semi-welfare state. After all, the modern state organizes what is in practice a substantial annuity, and one that obviously has a present value. Another powerful criticism, by MIT graduate student Matthew Rognlie, reworks the figures to show diminishing net returns on capital (with the important exception of recent housing booms).⁸ But the striking feature of Piketty’s book is actually the degree of openness and accessibility of the data. All you need to do in order to replicate the graphs is to have a computer connected to the internet and a spreadsheet program. By contrast, in order to find out which cotton businessmen of the British Industrial Revolution were not taken into account by Beckert, you would need to be for many months plunked down in the library of a major research university.

The basic story can be told by a simple U shape. In both income and wealth distribution, there was massive inequality before World War I; that fell abruptly after the War and after the Great Depression. From the 1970s, however, at first in the United States and the United Kingdom, inequality rose once more. For the United States, the U is almost symmetrical: The top 1 percent took 18 percent of all income in 1910, as little as 9 percent in the 1970s, and then 20 percent in 2010. In the 19^th-century case, inequality arose mostly out of the accumulation of wealth; in the 21^st century, this is less so. In 1929, income from capital exceeded income from earnings for the top 1 percent of the income distribution; by 2007 this still was the case, but only for the top 0.1 percent.

The omission of transfer payments aside, Piketty makes no effort to place a sociological filter on the data set—odd, in a way, for someone claiming to reunite history and economics. There is little consideration of reasons why some significant cohorts in a population might choose not to earn more money, thus skewing the data—reasons like living alone because of divorce or extended age, extended educational periods for odyssey-years youth, and so forth.

Besides, at the center of Piketty’s book is a puzzle that doesn’t require a great deal of mathematical sophistication. He lays it out most clearly by retelling Honoré de Balzac’s short novel Le Père Goriot. Master criminal Vautrin tells the ambitious but impoverished young nobleman Eugène de Rastignac that it is pointless to try to achieve riches by study and hard work; that would only produce a modest marginal income. Better to marry an heiress, and then connive in a crime that would increase her inheritance.

Capital accumulates, and the main message of Capital in the Twenty-First Century has often been distilled into a simple equation: The rate of return on capital is persistently greater than growth, and thus that capital accumulates. Above all it did so at first in the 18^th and 19^th centuries in the form of government securities that produced a high yield in a no-inflation, specie-currency-based world. That is Piketty’s version of the story of state capitalism on the cusp of the Industrial Revolution. But, however attractive, this model obviously cannot be projected backwards and forwards indefinitely, for a simple arithmetical reason: An accumulation growing at a geometric or exponential rate would swallow up all production. Indeed, there is a substantial literature devoted to the demonstration that, in the long term, the interest rate moves to set the growth rate of capital as equal to the growth rate of incomes.⁹

This is a study of returns on all sorts of capital—and above all, even during the Industrial Revolution, of government securities or land. There is as a consequence little attention paid to the factors that most powerfully drive growth—innovation, entrepreneurship—and that in the end also affect the returns on state debt or landed property as incomes from the sources of wealth accumulate, or as capital itself becomes more abundant. In consequence, as countries move along the path of development, returns on capital fall; or, put another way, more capital input is needed to get increased output.

Managing the sustained accumulation of capital is thus not simple. It requires an ever more sophisticated and sustained institutional and legal infrastructure. As Pikettty puts it, “The growing sophistication of capital markets and financial intermediation tends to separate owners from managers more and more and thus to sharpen the distinction between pure capital income and labor income.”

The story of the problem of management can also be told anecdotally, as with Balzac’s Vautrin and his story’s illustration of large accumulations of unearned wealth. Literature is indeed full of such illustration. In The Count of Monte Cristo, by Alexandre Dumas, Edmond Dantès learns the location of a hidden fortune from his fellow convict at the island fortress of the Château d’If; Jean Valjean, from Victor Hugo’s Les Misérables, is a former convict who starts off with almost nothing and through a process of making and selling fake jewelry accumulates 1.7 million francs within eight years.¹⁰ All this makes for wonderful stories, but none of them is plausible as an exemplar of actual economy history.

Far more characteristic is the autobiography of the greatest British historian of all time, Edward Gibbon. Gibbon describes how his family repeatedly lost large sums of money. His grandfather was a director of the South Sea Company that collapsed so catastrophically in 1702: “The labours of thirty years were blasted in a single day”, and “the relics of his fortune afforded a bare annuity for his maintenance.” For the next generation, “the last indispensable condition, the freedom from debt, was wanting to my father’s felicity.” Gibbon himself could only remove the heavy mortgage by the sale of the estate. He could only comfort himself by insisting that “the temper of a mind exempt from avarice suggests some reasonable consolation.”¹¹ Before a system of wealth management existed, there was much more opportunity to reduce than to augment patrimonies.

That new system involved a diversification of investments. In particular, societies with a high rate of accumulation tend to export capital, such as 19^th-century Britain or France in “the first globalization” epoch, or Japan and Germany in the more recent wave of globalization. That is, they progressively build up claims on other countries. An inability to build up such claims would slow down the process of accumulation. The U in income distribution also thus corresponds to the U that scholars of globalization have indentified. Thus, while the Gini coefficient may have been getting worse in the United States and other labor-expensive advanced countries, it has been declining for the world as a whole. This is because the beginning and end of the 20^th century were marked by an extensive globalization of goods and capital markets, and by a high degree of labor mobility. By contrast, the aftermath of World War I and the Great Depression curtailed this movement and introduced a phase of deglobalization.

Piketty, however, does not see things quite this way. He can neatly link the history of capitalism to the story of globalization by observing that “global income distribution is more unequal than the output distribution, because the countries with the highest per capita output are also more likely to own part of the capital of other countries and therefore to produce a positive flow of capital from capital originating in countries with a lower level of per capita output.” That may be so, but it has not prevented the post-Cold War evolution of global capitalism from raising more people out of poverty than at any time in history.

Max Weber, probably the most insightful analyst of capitalism, made the argument, in his last series of lectures on economic history, that capitalism rested on the elimination of differences—on a recognition that it was now possible to engage in transactions and exchanges with very different cultures and societies. The local and the familiar no longer bore special privilege; there was no special treatment for the neighbor.¹² That is the essence of what we now think of globalization, and to that extent modern capitalism and globalization are two aspects of the same basic process. But states exist to protect communities, and representative government and democratization work more and more to insist on the ability of the state to serve as a carapace or protector that can limit the globalization of trade flows, capital flows, and labor flows.

Capitalism, or the market economy, cannot simply go on forever driven by an internal momentum or dynamic. Any of the basic proclivities that drive capitalism are, on their own, destructive of long-term success. Capitalism depends on investment, but if there is too much investment there are production gluts, and the market is oversatiated. That was the dynamic that Marx highlighted a long time ago. Capitalism depends on consumption, but we cannot all just go on consuming more and more forever. Capitalism depends on competition, but competition can be brutal and destructive. In consequence, very elaborate systems of laws are needed to ensure that competition is open and fair, and that monopolies and trusts do not destroy the character of the economy. But even this may not be enough, since each legal reform and amplification is answered by entrepreneurial ingenuity on the part of those who want to circumvent the new restraints.

In consequence, some thinkers, most notably Weber, floated the idea that capitalism needed to be sustained by a value system that could not initially be created from within. The key issue that both Beckert and Piketty raise at least by implication, but cannot ultimately answer, is how capitalism will collide with a system of ethics that is suitable for a technologically interconnected world, one in which the concept of neighborhood is extended such that all humans are neighbors. Data sets have nothing much to do with this.