What does “capitalism” mean across the disciplines that use the word?

If the discipline barely uses the word, who within it does? A small set of economists kept “capitalism” central — and surfacing them at strength shows what the word does when an economist does pick it up.

Capitalism is not, and never can be, stationary. The fundamental impulse that sets and keeps the capitalist engine in motion comes from the new consumers’ goods, the new methods of production or transportation, the new markets, the new forms of industrial organization that the capitalist enterprise creates. The opening up of new markets and the organizational development from the craft shop to the modern corporation illustrate the same process of industrial mutation — if I may use that biological term — that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one. This process of creative destruction is the essential fact about capitalism. It is what capitalism consists in and what every capitalist concern has got to live in. The textbook picture of competition as a matter of firms shaving prices misses the point entirely: the competition that counts is the competition from the new commodity, the new technology, the new source of supply, the new type of organization — competition which strikes not at the margins of the profits of existing firms but at their foundations and their very lives. And because this is so, the conventional case for and against capitalism on grounds of static efficiency is largely beside the point. The relevant problem is not how capitalism administers existing structures but how it creates and destroys them. A system that is at every point of time fully exploiting its possibilities may yet be inferior, in the long run, to one that does so at no single point because the level of performance over time may be raised by the very gales of destruction the static analyst deplores.

— the argument of Joseph Schumpeter, Capitalism, Socialism and Democracy (1942), rendered in his own register. For the lineage of creative destruction as economics’s own dynamic-system reading of the word, see History of Thought Ch.7 (Schumpeter and creative destruction).

The second economist who kept the word is Milton Friedman, and he used it to do political-philosophical work the price diagrams cannot. The kind of economic organization that provides economic freedom directly, namely competitive capitalism, also promotes political freedom because it separates economic power from political power and in this way enables the one to offset the other. History suggests only that capitalism is a necessary condition for political freedom; it is clearly not a sufficient one. But a society that concentrates the control of economic activity in the same hands that hold political power has surrendered the chief check on that power. For Friedman the word “capitalism” carries a claim economics-as-price-theory never makes: that the market is not merely efficient but a precondition of a free society. The Chicago and Austrian re-use of the word in this register is traced in History of Thought Ch.10 (the counter-revolution), with the Mises and Hayek strand in Ch.6 (the Austrian tradition).

— the argument of Milton Friedman, Capitalism and Freedom (1962).

A modern mainstream version exists too. Acemoglu and Robinson’s Why Nations Fail (2012) slides between “inclusive institutions” and “capitalism” almost interchangeably — the word reappears as a synonym for secure property rights plus a market economy plus broad political access. The institutions lineage in economics runs through History of Thought Ch.15 (the institutionalist tradition, Veblen to Acemoglu).

So the small set who keep the word use it in a specific way: as a roughly-synonymous-with-market-economy noun with dynamic-system overtones (Schumpeter) or political-philosophical stakes (Friedman), but not as the load-bearing analytical term of their formal work. And notice what none of these uses engages. Whether there are different kinds of capitalism — Stage 2’s question. Whether capitalism has a beginning and an end — Stage 3’s question. Whether the wage relation is its defining structure — Stage 4’s question. Economics’s apparatus is precise about what it engages and silent about what it doesn’t.

Take the founding move at full strength. The firm, not the market and not the state, is the central actor in a capitalist economy — and firms face coordination problems they can solve only through the institutional environment that surrounds them: how they raise capital, how they secure a skilled workforce, how they bargain with labor, how they relate to other firms. In liberal market economies — the United States, Britain, Ireland, Australia — firms coordinate primarily through competitive markets and arm’s-length contracting; equity finance, fluid labor markets, and general skills form a complementary set. In coordinated market economies — Germany, Japan, the Nordic countries — firms coordinate through non-market relationships: bank-based finance with patient capital, industry-wide wage bargaining, vocational training that produces firm- and industry-specific skills, and dense business associations. These are not points on a continuum from less to more capitalist; they are distinct equilibria, each producing its own pattern of comparative institutional advantage. Liberal economies excel at radical innovation, where mobile labor and venture finance let firms reconfigure fast; coordinated economies excel at incremental innovation, where long employment horizons and patient capital let firms accumulate deep, specific competence. The plural “capitalisms” is not loose talk. It encodes the analytic claim that there are kinds of capitalism that are internally coherent, reproduce themselves, and generate falsifiable predictions about which economy does what well — predictions the single undifferentiated “market economy” of price theory cannot make.

— the argument of Peter Hall and David Soskice, Varieties of Capitalism (2001), in their own register.

Wolfgang Streeck started inside this program and then pushed past it — and the way he pushed is itself diagnostic. A comparative-statics typology, he came to think, freezes capitalism into a set of stable national equilibria precisely when the interesting thing about it is that it does not hold still. The postwar settlement was not one of several equally durable types; it was a historically specific regime — democratically embedded, growth-financed, conflict-pacifying — that has been coming apart in stages since the 1970s, through inflation, then public debt, then private debt, and now a fiscal exhaustion that no national variety has escaped. To say this, Streeck had to import a frame the typology could not supply: periodization. The varieties exist, but they are varieties within a moving sequence, and the sequence is the thing. That a leading Varieties-of-Capitalism scholar found the typology frame unable to discharge his own questions — and reached into historical sociology for the missing apparatus — tells you where CPE’s use of the word runs out.

— the argument of Wolfgang Streeck, Buying Time (2014) and How Will Capitalism End? (2016).

A parallel typology does the same work one layer down, at the welfare state. There are not welfare states arrayed on a single axis of more or less generous; there are three qualitatively different worlds of welfare capitalism, distinguished by how far they decommodify — how far they let people maintain a livelihood without selling their labor on the market. The liberal world (the United States, Britain) keeps benefits residual and means-tested, so the market still disciplines. The conservative-corporatist world (Germany, Austria, France) ties benefits to employment status and preserves existing class and status differences. The social-democratic world (the Nordic countries) makes generous benefits a right of citizenship, breaking the market’s grip on welfare. Here too “capitalism” is the typology-bearing noun: not whether you have a welfare state, but which kind, and what that kind does to the relationship between markets and lives.

— the argument of Gøsta Esping-Andersen, The Three Worlds of Welfare Capitalism (1990).

A note worth making explicit: none of these authors lives inside the History of Economic Thought textbook. Hall, Soskice, Streeck, and Esping-Andersen are political scientists and sociologists, not economists, and the C-book’s scope is the history of economic thought. The institutionalist adjacency — Veblen through to Acemoglu — is the closest the textbook comes, in History of Thought Ch.15; the comparative-political-economy primary sources are cited directly here because the field that uses the word this way sits across a disciplinary border from economics.

Stay with Braudel a moment longer, because the two-floor image is doing more than it first appears. The lower floor — what he calls material life and the market economy — is ancient, near-universal, and roughly competitive: peasants selling grain, artisans selling cloth, prices set by the visible tug of supply against demand. On that floor no one is large enough to dictate terms. The upper floor is where capitalism proper lives, and its defining feature is the opposite of competition. The great merchants of Genoa, Antwerp, Amsterdam, and London did not thrive by accepting the market price; they thrived by escaping it — through monopoly, privileged information, long-distance arbitrage between zones that could not see each other’s prices, and an intimacy with state power that let them write the rules of the trade they dominated. Capitalism, in this reading, is the zone where the market is bent to the advantage of those large enough to bend it. The category mistake economics makes, on Braudel’s terms, is to take the transparent competitive lower floor — the part that is genuinely a market — and call that capitalism, when what makes capitalism capitalism is precisely the upper-floor relationship between accumulation and power that competitive price theory cannot see.

— the argument of Fernand Braudel, Civilization and Capitalism, Vol. 2 (1979).

See Braudel’s “upper floor” on the trade map — the Atlantic long-distance system where monopoly and finance lived

Karl Polanyi makes the most analytically powerful single move in the whole stage. For nearly all of human history, he argues, the economy was embedded in society: production and exchange ran on custom, kinship, reciprocity, and redistribution, and the idea of a self-regulating market governing the whole of social life would have been unintelligible. What happened in nineteenth-century England was not the natural flowering of a tendency always present but a violent and deliberate institutional revolution — the attempt to organize an entire society around self-regulating markets for the three things that are not, in truth, produced for sale at all: land, labor, and money. He calls these fictitious commodities, because to treat human beings, nature, and purchasing power as ordinary goods produced for the market is to subordinate the substance of society to a mechanism, and no society could survive the full working-out of such a principle. So it does not. Against the disembedding of the market there arises, spontaneously and from every direction, a counter-movement — factory acts, trade unions, central banking, tariffs, and ultimately the welfare state — society protecting itself. This double movement is the real engine of the nineteenth and twentieth centuries, and the convulsions of the 1930s, fascism and the New Deal alike, are best read as rival attempts to re-embed a market that had been let loose. On this reading, capitalism is the historical anomaly — the brief, unstable episode in which society tried to serve the market rather than the market society — and to project it backward onto the substantive economies of the past is to misunderstand both.

— the argument of Karl Polanyi, The Great Transformation (1944). The nineteenth-century commodification he reads as the great transformation is the period in Economic History Ch.7 (the Industrial Revolution).

Immanuel Wallerstein refuses the question CPE asks. There are not, in the deep sense, different capitalisms — there is one capitalist world-system, in place since roughly the long sixteenth century, and the only question that matters is your position within it. The system is organized into a core, where high-wage, high-technology, capital-intensive production concentrates; a periphery, locked into low-wage, coerced or semi-coerced production of raw materials and food; and a semi-periphery that holds the structure together by being exploited by the core while exploiting the periphery in turn. This is a single division of labor spanning the globe, and it reproduces inequality not as an accident to be corrected but as the very mechanism by which it works: the periphery’s poverty is not a failure to develop but the condition of the core’s wealth. From this vantage, the liberal-versus-coordinated distinction that absorbs comparative political economy is real but second-order — intra-core variation among the already-rich. The load-bearing line on the planet is not between Germany and the United States; it is between the core as a whole and the periphery it was built on. The formation of that world-system in the Atlantic, through plantation slavery and colonial extraction, is its founding act, not a regrettable side-chapter.

— the argument of Immanuel Wallerstein, The Modern World-System (1974). The Atlantic world that is constitutive of his core-periphery structure is the history in Economic History Ch.9 (the Atlantic world: slavery, abolition, and after) and the wider imperial system in Ch.10 (imperialism and colonial economies).

See the core-periphery flows Wallerstein reads as constitutive — the Atlantic system on the trade map

Giovanni Arrighi extends Braudel’s upper floor into a rhythm. The history of capitalism is the history of systemic cycles of accumulation, each centered on a leading state-and-capital complex: a Genoese-Iberian cycle, a Dutch cycle, a British cycle, an American cycle. Each begins with a material expansion — capital flowing into trade and production — and each ends, as competition erodes profits, with a turn to finance, a belle époque of speculative wealth that is also the signal of the hegemon’s autumn before a new center takes over. The financialization of the late twentieth century, on this reading, is not a new normal but the closing phase of the American cycle — the long twentieth century ending, with whatever comes next still contested.

— the argument of Giovanni Arrighi, The Long Twentieth Century (1994).

As in Stage 2, the disciplinary border is real and worth naming. Braudel, Polanyi, Wallerstein, and Arrighi are Annales-school history and historical sociology, not the history of economic thought, so they have no chapter in the C-book — they are cited directly. The closest the textbook comes is Marx, who is the shared precursor much of this work builds on, in History of Thought Ch.4; and the post-2008 heterodox revival that has pulled Polanyi-style and Wallerstein-style framings back into live economic argument is in Ch.17 (modern pluralism).

Take the mode-of-production frame at full strength. What makes a society capitalist is not that it has markets — markets are older than capitalism by millennia — and not that it has private property, which feudal lords had in abundance. What makes it capitalist is a single, historically specific relation: that the great mass of producers have been separated from the means of production and can live only by selling their labor-power, while a class that owns those means buys that labor-power and sets it to work. And here is the hinge of the whole analysis. The worker is paid the value of his labor-power — what it costs to reproduce him, to keep him alive and able to work and to raise the next generation of workers. But labor-power, once bought, can be made to produce more value than it costs. The working day does not stop when the worker has produced the equivalent of his wage; it continues, and that continuation — surplus-value — is appropriated, without exchange, by the owner of capital. This is not cheating, not a market imperfection, not a deviation from competitive equilibrium that better policy could correct. It is the normal, lawful operation of the system working exactly as designed: exploitation is structural, built into the wage form itself, and concealed by it, because the wage looks like payment for the whole working day when it is payment for only part. The varieties of capitalism that comparative political economy so carefully catalogues are, from here, varieties of legal-institutional dressing on an unchanged relation of class. And the price-system frame of Stage 1 — the analysis of how commodities exchange at their values in the bright transparent sphere of circulation — is what Marx called vulgar political economy: an account that takes the surface of the system for its substance, describing the marketplace where things are bought and sold while studiously not entering the hidden abode of production, on whose threshold hangs the notice, No admittance except on business, and where the secret of profit-making is finally laid bare.

— the argument of Karl Marx, Capital, Vol. 1 (1867), in his own register.

The frame did not die with Marx. The Frankfurt School — Adorno and Horkheimer’s Dialectic of Enlightenment (1944), and later Habermas — extended the analysis from the factory to culture and consciousness, reading the culture industry and the colonization of everyday life as the means by which the relation reproduces itself ideologically. And it remains a live research program: David Harvey reads neoliberalism and the urban form through accumulation and dispossession; Robert Brenner traces the long downturn since the 1970s to capitalist competition itself; the New Left Review circle keeps the structural-contradictions frame current; and Thomas Piketty, no Marxist, nonetheless put the relation between the return on capital and the rate of growth back at the center of distributional argument in Capital in the Twenty-First Century (2014). The post-2008 revival of this register in economic argument is in History of Thought Ch.17 (modern pluralism); the Frankfurt School itself sits outside the C-book’s scope and is cited directly. The deeper question of which of Marx’s specific predictions actually held up — immiseration, the falling rate of profit, revolution — is the work of the live sibling walkthrough on whether Marx was right about anything; here the point is only that the class-relation framing is one disciplinary lens among several, not a relic.

Now to the place all four framings collide without anyone announcing which one they are using: 2020s political discourse. Four voices, each engaging the word in good faith from inside its own disciplinary mix.

When Bernie Sanders says “we live under capitalism” — a line from his stump speeches and his 2019 defense of democratic socialism — the preposition is doing the analytical work. You do not live under a price mechanism. The framing is primarily critical-theory: capitalism as a class structure that subordinates, a system one is positioned within rather than a market one participates in. Secondarily it is historical-sociology, because “under capitalism” quietly periodizes — it names a particular era with a particular structure, implicitly one that began and could end. What the framing does not invoke at all is the CPE question of which kind of capitalism, or the economics question of how the price system is performing.

When Klaus Schwab and the World Economic Forum launched “stakeholder capitalism” — the 2020 Davos Manifesto, the book of the same name in 2021 — the word is doing CPE-typology work with a public-relations overlay. The whole premise is the plural: there are kinds of capitalism, and we should choose a better one, a stakeholder kind rather than a shareholder kind. Underneath, it is largely a rebrand of the mainstream-economics market-economy frame with institutional-variation features added back in — which is why it reads, to a critical-theory ear, as a way of changing the adjective while leaving the class relation untouched. At the rhetorical level, that is the move: invoke the typology, leave the structure alone.

When a national-conservative voice — Oren Cass’s The Once and Future Worker (2018), or the broader “save American capitalism from the globalists” register — takes up the word, the framing is a hybrid. It is CPE-typological, in that it treats capitalism as a national institutional configuration that can be the right kind or the wrong kind, and historical-sociological, in that it diagnoses the current configuration as in crisis and decline, with a culturally-conservative overlay about what an economy is for. The same noun, mobilized to argue that a particular national variety has been hollowed out and must be rebuilt — an argument the economics frame, with its indifference to national-institutional kinds, has no vocabulary to make.

And when a libertarian-popular voice — the Patrick Bet-David, Joe Rogan, Jordan Peterson register — says “capitalism lifted billions out of poverty,” the framing is economics, exclusively. The substantive content is the price system, private property, and market coordination; the evidence base is the cross-country poverty-reduction record since 1980; and the overlay is Friedman’s political-philosophical claim that this same system is the guarantor of freedom. It is the cleanest single-framing voice of the four — and precisely because it is single-framing, it cannot hear the historical-sociology question about which phase of which cycle produced that growth, the critical-theory question about how the gains were distributed, or the CPE question about which institutional configurations delivered them.

“Capitalism lifted billions out of poverty.”

The operative framing is economics: market economy plus price coordination as the substantive content, the post-1980 cross-country poverty-reduction record as the evidence. What it does well: it connects to a real and large empirical literature, and the headline fact — hundreds of millions out of extreme poverty, concentrated in China and India — is genuine. What it leaves out is everything the other three see. The historical-sociology question of whether the post-1980 episode is a durable feature of the system or one phase of one accumulation cycle. The critical-theory question of how those gains were distributed, within countries and between them, and at what cost to whom. The CPE question of which institutional configurations actually produced the growth — because “capitalism” in the abstract did not raise Chinese incomes; a very specific, very non-liberal Chinese institutional package did. Translating the claim across framings does not refute it. It locates it — as true within economics’s scope, and incomplete the moment it is taken as the whole story.

“We live under capitalism.”

The operative framing is critical theory, with historical sociology underneath. The defining content is the wage-labor and private-appropriation relation, and the preposition — under — carries the claim of structural subordination; the implicit periodization (we live in a particular era characterized by this structure) is the historical-sociology layer. What it does well: it connects to a real literature on the persistence of the class relation across every institutional variety the rich world has tried. What it leaves out: the CPE question of which kind of capitalism we live under — the differences the framing flattens are exactly the ones a Swede and an American would feel — and the economics question of how the price-coordination mechanism is actually performing for the people inside it. Same disambiguation, different disciplinary load.

“America needs a new kind of capitalism.”

The operative framing is CPE, with historical sociology secondary. The giveaway is the plural mobilized as a choice: there are kinds, and a new kind can be selected — the Hall-Soskice noun put to political use. The secondary periodization is the implicit claim that the current kind has run its course. What it does well: it connects to a real literature on institutional configurations and a Streeck-style reading of regime exhaustion. What it leaves out: the critical-theory question of whether changing the institutional configuration changes the underlying class relation at all — the suspicion that a “new kind” reshuffles the dressing — and the economics question of what the proposed new configuration’s actual allocation and welfare properties would be. Three claims, three disciplinary mixes, one move that makes each of them legible.