Is what we call “capitalism” one system or many?
The word does at least four jobs at once — a price system, a set of national types, a 500-year sequence, a class relation. The fights about capitalism are usually fights about which job the word is doing.
Capitalism as a price system
“Capitalism is the economic system in which the means of production are privately owned and goods are allocated through voluntary exchange at market prices. Everything else — the welfare state, the central bank, the corporate-governance code — is decoration on that core.”
— the modal definition in an introductory economics course
This is the answer most readers arrive carrying. It is correct as far as it goes — and the rest of this walkthrough is about how far that is. Start by taking the price-system frame at its strongest, because the best version of it is more interesting than the textbook gloss.
Strip the frame to its load-bearing parts. Productive assets are privately owned. People decide what to make and buy on their own account. No central office sets the allocation; prices do, by rising where things are scarce and falling where they are abundant, so that millions of uncoordinated decisions add up to a rough order. The formal capstone is the first welfare theorem: under competitive markets, full information, and no externalities, the price-coordinated outcome is Pareto-efficient — you cannot make anyone better off without making someone worse off.
In the competitive equilibrium $(p^*, x^*)$, every agent maximizes subject to their budget and all markets clear. The first welfare theorem states that this allocation is Pareto-efficient:
$$\text{competitive equilibrium} \;\Rightarrow\; \text{Pareto-efficient allocation}$$The theorem is exact — and exactly as strong as its assumptions, which is the opening the rest of the walkthrough walks through.
No one designs the bread supply of a city, yet the bread arrives. Prices carry just enough information — what is scarce, what is wanted — for strangers who will never meet to coordinate. That is the whole claim: capitalism is the arrangement that lets the price signal do the planning.
Mainstream economics already knows where this breaks. When one person’s production dumps costs on a bystander — pollution, congestion, an uninsured risk — the price is wrong and the efficiency claim fails. These are market failures, and the discipline has a whole apparatus for them. But notice what kind of qualification this is: it patches the price system from inside. It never asks the question the next stage asks — whether two economies that both satisfy the price-system frame can still be different systems in any way that matters.
The formal home of the apparatus — supply, demand, and the price mechanism — is Ch 2 §2.1 (Demand); the catalogue of where prices stop telling the truth is Ch 4 §4.1 (Externalities).
The strongest version: Hayek’s knowledge problem
“The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form, but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess.”
— Friedrich Hayek, “The Use of Knowledge in Society,” American Economic Review, 1945
Hayek gives the price-system frame its sharpest form, and it is worth following exactly because it is not the textbook’s. The economic problem society faces, he argues, is never simply the allocation of given resources; it is the use of knowledge that no single mind ever holds. The relevant facts — that a particular machine is idle, that a particular tin of grease can be substituted, that this customer wants delivery on Tuesday — exist nowhere as a list. They live in the heads of the people on the spot, and they change by the hour. A central planner would need to gather all of it, in real time, and would fail not for want of computing power but because the knowledge is constitutively dispersed and partly tacit. The price system is Hayek’s answer to that impossibility. A change in one corner of the economy — a mine floods, a new use for tin appears — ripples out as a change in price, and everyone who deals in tin adjusts without ever learning why. The price is a sufficient statistic for everything they need to know. For Hayek this is not one feature of capitalism among others; it is what capitalism is. A system is capitalist to the degree that it lets prices, rather than commands, carry the coordinating knowledge. Everything Stage 2 will call “institutional variation” he would file under the engineering of the same machine: so long as the price signal does the planning, the rest is detail.
Hold onto how much weight that puts on a single mechanism — because the next stage is going to ask why, if price coordination is the whole story, Germany and the United States run on visibly different machinery and stay different for decades. Hayek’s frame has no place to put that question. It is not built to see the difference between two systems that both let prices clear; it is built to see the difference between price coordination and central command. That is a real and important distinction. It is also not the only one.
Where this leaves us
The price-system frame is right as a necessary condition and incomplete as a sufficient one. No arrangement that anyone in any of the literatures ahead calls “capitalism” lacks price coordination and private ownership of productive assets; that much is genuinely common to all the cases. But knowing that a system has those two features does not tell you whether the German social-market economy and the Chinese state-capitalist economy are the same system in any sense worth the word. Keep the price-system frame, then, as the layer on which “one system” is true. The walkthrough’s wager is that there are other layers, and that “many systems” becomes true on them. The deep root of this whole question — what kind of system is this? — is the one Adam Smith first put to commercial society; the genealogy runs through History of Economic Thought Ch.3 (Classical political economy), and the price-coordination frame itself was sharpened into formal apparatus by the marginalist revolution (Ch.5) in whose lineage Hayek stands. There is a rival answer Smith’s frame never quite engages — the one that defines capitalism by wage labour rather than by prices — waiting in Ch.4 (Marx); Stage 3 lets it speak.
If price coordination were the whole of it, why does the German economy run on long-term bank finance and patient cross-shareholding while the American one runs on equity markets and arms-length capital? Why does Japan train workers inside firms and Britain through portable credentials? Surface variation on one system — or different machines that happen to share a name? The next stage hands the question to the people who built careers measuring the difference. The historical ground for it — the Atlantic commercial world before industry, and Britain’s first industrial capitalism — sits in Economic History Ch.5 and Ch.7, and you can see the long-distance trade that Braudel called “capitalism proper” on the Atlantic trade map.
Capitalism as a comparative type
“In any capitalist economy, firms are the crucial actors… The most important institutions distinguishing one political economy from another will be those conditioning the interaction among firms and the other actors. We assign these many possibilities to two broad types: liberal market economies and coordinated market economies.”
— Peter Hall & David Soskice, Varieties of Capitalism, 2001
This is the founding statement of the Varieties of Capitalism program — the modal answer to our question on the comparative-politics side. The claim is not that one country is more capitalist than another. It is that rich capitalist economies sort into a small number of stable, internally consistent types. The program has since grown well past the original two.
The analytical engine here is institutional complementarity: the payoff to one institution depends on the presence of another. Long-term employment is worth more to a firm when patient capital lets it ride out downturns; patient bank finance is sustainable when cross-shareholding insulates managers from hostile takeover; firm-specific worker training pays off when wage-setting is coordinated so a rival cannot poach the trained worker at a small premium. Each piece props up the others. That is why these configurations are not menus of policy choices a country can mix freely — they are equilibria, sustained because the parts reinforce each other, and sticky for the same reason. Institutions, in this telling, behave less like rules a government picks and more like the institutional foundations economics studies under property rights and the rules of the game.
The apparatus for treating institutions as the binding constraint — not as background — lives in Ch 18 §18.2 (North’s institutional framework), with the inclusive-versus-extractive distinction at §18.4; the claim that different institutional bundles yield different innovation patterns connects to Ch 13 §13.5 (Schumpeterian growth).
Three reads of the same comparative evidence
Varieties of Capitalism: two equilibria
On the Hall–Soskice account, the rich capitalist world resolves into two coherent ways of organizing the firm’s relationships, and each is an equilibrium in the strict sense. In a liberal market economy (the United States, Britain, Ireland, Australia, Canada), firms coordinate mainly through arms-length, competitive markets: equity finance from dispersed shareholders, fluid labour markets, general and portable skills, wage-setting decentralized to the firm. In a coordinated market economy (Germany, Austria, the Nordics, Japan), firms coordinate through non-market institutions: long-term bank finance and cross-shareholding that free managers from the quarterly horizon, employer associations and works councils, sector-wide vocational training, and wage-setting bargained across the economy. The point is not that one is softer or kinder. It is that each bundle hangs together and produces a distinct comparative advantage. Where the liberal economy throws capital and labour at radical, market-making innovation (biotech, software, the venture-funded bet), the coordinated economy excels at incremental, high-quality improvement in machine tools, specialty chemicals, premium cars, where patient capital and a deep stock of firm-specific skill compound over decades. The logic is tight: a worker will only invest years in a narrow, firm-specific skill if institutions (employment protection, coordinated wages, a training system that certifies the skill) make that bet safe, and a firm will only finance the long apprenticeship if patient capital lets it wait for the return. Pull one institution and the others lose their footing, which is exactly why these bundles do not travel: you cannot graft German apprenticeship onto an American labour market and expect it to take. Predict from the institutions and you predict the innovation profile, the dispersion of wages, the way each economy rides a recession — and the predictions have largely held: liberal economies really do dominate the radical-innovation sectors, coordinated ones the high-spec manufacturing export niches. Two systems, both fully capitalist, both stable, doing different things on purpose.
The developmental state: a type the binary missed
Chalmers Johnson, Robert Wade, and Alice Amsden looked at the postwar ascents of Japan, Korea, Taiwan, and Singapore and found a configuration the liberal/coordinated binary cannot hold. Here the state is not a referee that sets the rules and steps back; it is a player. A pilot bureaucracy — Japan’s MITI, Korea’s Economic Planning Board — allocates credit, picks sectors, and steers investment through what Johnson called administrative guidance. Crucially, this is not the old import-substitution story of subsidy without discipline. Amsden’s decisive observation was that the East Asian states got prices wrong on purpose but tied every subsidy to a hard performance test: export, meet world-market standards, or lose the support. Wade called it governing the market — using state power to build comparative advantages that did not yet exist rather than accepting the ones a country was born with. This is a fourth equilibrium type, with its own complementarities (high state capacity, disciplined finance, export targeting), and it is the friendly amendment the Varieties program eventually had to absorb. It also plants a doubt the next thread sharpens: if the typology has to keep growing a new type for every region that does not fit, how clean were the original types?
Dimensions, not types: Amable and Esping-Andersen
Bruno Amable ran the question the other way around. Instead of starting from two ideal types and sorting countries into them, he scored rich economies on five institutional dimensions — product-market competition, labour-market institutions, the financial system, social protection, education — and let the clusters fall out of the data. They did not fall into two. They fell into five: a market-based model, a Continental European model, a social-democratic model, a Mediterranean model, and an Asian model, with boundaries softer and fuzzier than the crisp LME/CME line implied. Gøsta Esping-Andersen had reached the same shape a decade earlier from the welfare side, ranking countries by how far the welfare state decommodifies — lets a person survive a spell of unemployment, sickness, or old age without selling their labour at whatever price the market offers — and finding three worlds of welfare capitalism: a liberal world of means-tested minimal safety nets, a conservative world of status-preserving social insurance, and a social-democratic world of universal high benefits. Crucially these three worlds cut across the production typology, so a country can sit in one cluster on finance and another on welfare — which is precisely the evidence that the dimensions move somewhat independently rather than locking into two master types. The lesson from this corner is methodological and it bites: real institutional clustering exists, but whether you see two types or five or a smooth gradient depends on which dimensions you weight and how hard you draw the lines. The clusters are real; the number two was an assumption, not a finding.
What survives the evidence
The honest synthesis is dimensions with cluster structure. There is genuine clustering — the Anglo-liberal centroid and the Continental-coordinated centroid are robust, they show up however you cut the data, and the prediction that the two do different kinds of innovation holds up reasonably well. So “types exist” survives. What does not survive is the clean binary of the original 2001 statement: the developmental state and the post-Soviet hybrids forced the typology to keep adding rooms, and Amable’s method shows the walls between rooms are thinner than first drawn. Push to the live parameter question — how much have these economies converged since 1980? — and the answer is partial. Corporate-governance and financial institutions have drifted toward the liberal model almost everywhere; shareholder value, equity markets, and footloose capital spread. But the welfare state and the labour-market institutions proved durable, anchored by domestic politics that did not melt. So at this layer the comparative-politics literature has been right that variation is deep and equilibria are stable — and wrong, in its first confident form, that the variation comes in two clean kinds. The grounding evidence for all of it is in the divergent industrializations of Germany, Japan, and the US in Economic History Ch.8, the postwar variants — the Trente Glorieuses, the Wirtschaftswunder, the Japanese miracle — in Ch.14, and the Chinese state-capitalist case that broke clean out of the 2001 boundaries in Ch.17 §17.1; you can put the long growth trajectories of these economies side by side on the GDP map (compare the DEU, JPN, USA, KOR, and CHN annotations). One honest gap: Hall, Soskice, Amable, and Esping-Andersen are comparative-political-economists, not figures in the history of economic thought, so the closest the lineage book reaches is the institutional tradition from Veblen to Acemoglu in History of Economic Thought Ch.15 — these primary sources are cited directly because the field sits outside that book’s scope.
Every word of that argument was pitched at the altitude of the present — rich countries, the last forty years, a typology of co-existing forms. Step back to a longer horizon, to capitalism as a thing with a five-hundred-year history rather than a forty-year taxonomy, and the question changes shape entirely. Stage 3 hands it to the historians and the heterodox — the people who think “capitalism” names not a set of contemporary types but a sequence, or an anomaly, or a class relation that the institutional detail barely touches.
Capitalism as a historical sequence
“Capitalism without opposition is left to its own devices, which do not include self-restraint. The capitalist system is at present stricken with at least five worsening disorders for which no cure is at hand: declining growth, oligarchy, starvation of the public sphere, corruption, and international anarchy.”
— Wolfgang Streeck, How Will Capitalism End?, 2016
Streeck is not an outsider throwing stones. He helped build the Varieties program from inside, then moved to the historical-sociology frame after 2008 — and that move is itself a piece of evidence. If the comparative-types picture of Stage 2 fully answered our question, one of its own architects would not have gone looking for a different one.
There is no formal model to compress here, and pretending otherwise would be dishonest — periodization is a question for comparative political economy and economic history, not for the equation-bearing chapters of formal economics. What this stage owes you instead is a clean distinction between two shapes of claim. Stage 2 made a typology claim: capitalism comes in a few co-existing configurations, each internally consistent, all present at the same moment. This stage makes a periodization claim: what we call capitalism is a sequence of regimes, each with its own accumulation logic, separated by breaks that arrive when a regime’s internal contradictions accumulate past the point the institutions can hold. These are not rival answers to one question. They answer different questions — “what does capitalism look like now?” versus “what has capitalism been, across centuries?” — and both have evidence. Keeping them apart is half the work of the synthesis to come.
The periodizers
Streeck: three postwar regimes
Read the rich world since 1945 as a single story and it breaks cleanly into three. The first regime, roughly 1945 to 1973, was capitalism married to democracy: full employment, a rising wage share, an expanding welfare state — capital and labour bound together in a settlement that delivered both profit and broad-based prosperity. The marriage broke in the stagflation of the 1970s. The second regime, from around 1980 to 2008, was the unilateral release of capital from that bargain — the Hayekian turn, in which markets were freed, finance was unleashed, the wage share fell, and inequality climbed, with each crisis of the old settlement bought off by a new expansion of credit: first public debt, then private debt, then the central bank’s balance sheet. And the third regime, the one opened in 2008, is not a new settlement at all but an interregnum — an entropic phase in which neither the embedded-liberal compromise nor the neoliberal growth machine can be restored, and the system limps on without a stabilizing logic. The breakpoints are not Streeck’s decoration; they are visible in the data on labour share, financialization, and inequality, and they do not fall where a smooth-growth story would put them.
Arrighi: four systemic cycles
Giovanni Arrighi pulls the camera back four hundred more years. The long history of capitalism, he argues, is a succession of systemic cycles of accumulation, each organized around a hegemonic power and each ending the same way. A leading centre — Genoa, then the Dutch Republic, then Britain, then the United States — presides over a material expansion of trade and production; as that expansion exhausts its profits, capital retreats from goods into finance, and a financial expansion crowns the cycle just before the centre’s power passes to a rising challenger. “Capitalism” in this telling is not any one of the four hegemonies but the pattern of their succession — the recurring autumn in which the incumbent turns financier. The contemporary signal-flare, for Arrighi, is exactly that: the financialization of the American economy reads as the autumn of the American cycle, with the open question being whether an Asian-centred cycle follows or whether the four-cycle pattern itself is reaching its limit.
Braudel: the upper floor
Fernand Braudel offers the most disorienting move of all, because he splits the thing in two. Below, there is the everyday market economy: the transparent world of local fairs and small traders, of many buyers and many sellers, where competition is real and prices are made in the open. This is the world Stage 1’s price-system frame describes — and Braudel insists it is not what he means by capitalism. Capitalism proper lives on the upper floor: the shadowy zone of long-distance trade, monopoly, and high finance, where the great merchant houses operated on privileged information, where the rules were bent by political power, and where the real fortunes were made precisely by escaping competition rather than submitting to it. If Braudel is right, then most of what an economics course studies under the name “capitalism” is the ground floor, and the word has been quietly attached to the wrong storey. That is the most radical claim in this stage: not that capitalism has variety or sequence, but that the standard definition points at the wrong floor of the building.
The heterodox-historical frames
Polanyi: the great transformation and the double movement
Karl Polanyi denies the starting premise that everyone else takes for granted — that a self-regulating market is a natural condition that capitalism merely lets flourish. For the whole span of human history before the nineteenth century, he argues, the economy was embedded in social relations: production and exchange were governed by kinship, custom, religion, and obligation, with markets a marginal and supervised feature rather than the organizing principle. What happened in nineteenth-century Britain was an attempt, historically unprecedented and ultimately utopian, to invert that order — to disembed the market and make society serve it instead. The mechanism was the creation of three “fictitious commodities,” land, labour, and money, things never produced for sale that the market nonetheless insisted on pricing as if they were. The attempt was self-undermining. To treat human beings, nature, and purchasing power as pure commodities is to threaten to grind them to nothing, and society recoils. That recoil is the double movement: the spontaneous, many-sided defence — factory acts, trade unions, central banking, the welfare state, and in its darker forms fascism — by which society protects itself from the unrestrained market. Read through Polanyi, the postwar settlement was the double movement succeeding; the neoliberal turn was the market tearing loose again; and the convulsions after 2008 are the next protective countermovement still searching for its political form. Under this frame capitalism is not a system among systems but a brief, violent, and unstable nineteenth-century experiment whose consequences we are still living inside.
Wallerstein: one world-system
Immanuel Wallerstein turns the entire question on its head. The error, he holds, is to take the nation-state as the unit of analysis at all — to ask whether the German and American economies are different capitalisms. There has only ever been one capitalism, and it is not a country; it is a world-system, a single trans-national division of labour that took shape in the long sixteenth century and has been the same thing in its essential structure ever since. That structure is a hierarchy of zones: a core that captures high-wage, high-technology, high-profit production; a periphery locked into low-wage extraction of raw materials and labour; and a semi-periphery caught between, performing both roles and stabilizing the whole. The wealth of the core is not separable from the poverty of the periphery — the two are produced together by the same flows of value across the same system. So the LME-versus-CME variation that Stage 2 measured so carefully is, for Wallerstein, real but second-order: it is variation within the core. The structurally decisive line runs not between Boston and Frankfurt but between the core and the periphery it lives off — and who occupies which zone, and how the boundary shifts over centuries, is the only question about “varieties” that matters.
The Marxian frame: a class relation
For the Marxian tradition the defining feature of capitalism is neither its price mechanism nor its institutional bundle but its class structure: a society is capitalist when most people live by selling their labour-power to others who own the means of production and who appropriate the surplus that labour produces. Everything Stage 2 catalogued — the patient bank, the works council, the developmental ministry — is, on this view, variation in the legal and institutional dressing worn over an unchanged underlying relation between owners and workers. A coordinated market economy and a liberal one are both arrangements for the private appropriation of surplus value; that they appropriate it through different financial plumbing is not nothing, but it is not a difference in what kind of system they are. This is also why the Marxian frame is alive rather than antiquarian, and the post-2008 revival makes the point: the live questions it asks are about new mechanisms of accumulation — the financialization that turns ordinary life into a stream of interest payments, the platform that converts unwaged attention and gig labour into surplus — not about whether today’s capitalism is “more” or “less” capitalist than 1900’s. The relation is the constant; the question is always how it is being extended into new ground.
What the historians are actually claiming
Notice that none of these six voices is answering Stage 2’s question. The periodizers and the heterodox-historical frames are not asking “how do today’s configurations vary?” They are asking “what kind of historical entity is this thing?” — and on that question, “many systems sharing a name” is closer to right than “one system with variation.” The price-system frame of Stage 1 does not engage this question at all; it operates inside whichever regime the historical-sociology frames have identified, taking the regime as the given backdrop against which prices clear. There is a tame, mainstream version of periodization — merchant capitalism, then industrial, then financial, then platform — and it borrows Streeck’s shape while leaving out his apparatus: the regime-break mechanism, the bond between hegemonic power and accumulation pattern, the role of crisis. It is the periodization claim with the analysis removed. The strong forms are Streeck and Polanyi, and the walkthrough sides with them at this layer — with one proviso that the final stage turns into its main point: no version of the periodization claim replaces the price-system frame or the comparative-types frame. It operates on a different layer. The empirical chronology that grounds all of this lives in the history book — the 1970s break in Economic History Ch.16, the financialization regime in Ch.18, the 2008 rupture that closes Streeck’s third regime in Ch.19, and the slavery that the Marxian and world-systems frames read as constitutive of industrial capitalism in Ch.9 — and you can trace the core-periphery flows Wallerstein describes on the Atlantic trade map. In the lineage book, Marx stands at full strength in History of Economic Thought Ch.4, and the post-2008 return of Polanyian and Wallersteinian framings into live discourse is the modern-pluralism chapter (Ch.17); the whole network of schools, including the Marxian tradition and its revival, is browsable on the history-of-thought graph. Streeck, Arrighi, Braudel, Polanyi, and Wallerstein themselves are historical sociologists and Annales historians, not figures the history-of-economic-thought book owns — an honest boundary, and why they are cited here directly.
Three stages, three different verdicts. Stage 1: one system, on the layer where prices coordinate. Stage 2: many types, with soft cluster boundaries. Stage 3: many systems sharing a name, on the layer of long history. The last stage asks the only question left — are these competing answers, so that the reader must choose one, or layer-different answers, so that the reader needs all three to see the question at all?
The synthesis
You are now carrying five claims that pull against each other: Hayek’s price coordination as the definitional feature; Hall and Soskice’s complementarities as stable types; Streeck’s regime-breaks; Polanyi’s market society as a nineteenth-century anomaly; the Marxian wage relation as the constant beneath the institutional dressing. The easy move from here is to pick one and discharge the question — declare that capitalism “really is” the price system, or “really is” the class relation, and let the others fall away. That move feels like clarity. It is the one mistake the whole walk through the threads was meant to earn you out of. The harder, truer move is to see why the threads disagree — because once you see that, the disagreement stops looking like a fight and starts looking like a set of answers to different questions.
The tool that makes the disagreement legible is a single distinction: when two people disagree, they can be disagreeing at three different depths. They can disagree at the frame layer — about what the word even picks out, what work the noun is doing — and no evidence settles that, because frames are chosen for the questions they answer, not proved. They can disagree at the method layer — granting the same frame, about how to measure or model the thing — and that is settleable in principle, by which method explains more out of sample. Or they can disagree at the parameter layer — granting the frame and the method, about the magnitude of some quantity — and that is an ordinary empirical question with a rolling best estimate. Almost every public argument about capitalism mixes these three together and so becomes unwinnable. Pull them apart and most of the heat turns out to be misdirected.
The three layers of the disagreement
At the frame layer: a real disagreement that no evidence resolves
Mainstream economics says capitalism is the price system — private property plus coordination by prices. The Marxian tradition says it is the wage relation — the private appropriation of surplus from people who must sell their labour. Polanyi says it is the disembedded market — the nineteenth-century attempt to make society serve the market rather than the reverse. These are three different things to mean by one word, and they do not adjudicate against each other, because they are not making the same kind of claim. The price-system frame answers “what is the coordination mechanism?” The Marxian frame answers “what is the class structure?” Polanyi answers “what kind of historical entity is this?” Each is the right tool for its own question and the wrong tool for the others. Refusing to crown one of them is not a hedge or a both-sides dodge — it is the only correct treatment of a frame-level dispute. The reader who learns to ask “which frame is this claim using?” has gained more than any verdict could give them.
At the method layer: a productive debate that has mostly converged
Drop inside the comparative-political-economy frame — grant that the question is “how do contemporary capitalist configurations vary?” — and the remaining dispute is about method, not meaning: do you carve the world into a few discrete types, Hall–Soskice style, or score it on continuous dimensions and let clusters emerge, Amable style? This is the kind of disagreement evidence can move, and it has moved. The field has converged toward dimensions with cluster structure: there is real clustering, the centroids are robust, but the boundaries are softer and the count of types is less fixed than the original binary claimed. That is a settled-enough methodological result, reached the way method disputes are supposed to be reached — by which approach kept explaining the cases as new ones arrived.
At the parameter layer: live measurement, honest estimates
Drop one layer further, fix both the frame and the method, and what is left are quantities still being measured. How far have liberal and coordinated economies actually converged since 1980? The modal estimate is partial — substantial convergence in finance and corporate governance, durable divergence in welfare states and labour markets. How much of the East Asian developmental model survives the headwinds of the 2020s? How does Streeck’s post-2008 interregnum finally resolve — into a restored growth regime, a new settlement, or prolonged entropy? These are ordinary empirical questions with answers in principle and best-current-estimates in practice. The intellectually honest move is to name the modal estimate and flag its uncertainty, not to pretend the number is settled or that no number exists.
One system or many
“Is capitalism one system or many?” is not a question with one answer, because it is not one question. It is three questions wearing the same six words, operating at three layers at once, and naming the layers is what makes the question come apart in your hands. At the frame layer there is a genuine, unresolvable disagreement between the price system, the wage relation, and the disembedded market, and the right response is to recognize which frame a given claim is using rather than to pick a winner. At the method layer there is a real but largely settled debate that converged on dimensions with cluster structure. At the parameter layer there is ongoing measurement with partial convergence as the modal estimate. Each of the three traditions is right where it claims: mainstream economics at the frame layer it occupies, comparative political economy at the method layer it works on, historical sociology at the frame-and-long-horizon layer it reaches for. None of them is wrong; each overreaches only when it forgets which layer it is standing on. So what you carry away is a habit rather than a verdict: the next time a tweet or a column or a chapter tells you what capitalism “really is,” ask first which layer the claim is operating at, and most of the argument will resolve before you have taken a side.
Where this leaves us
We started with the word most readers carry in from an economics course — capitalism as private property plus prices — and took its strongest form, Hayek’s knowledge problem, seriously. Then each stage took something away from the unconditional version. The comparative-politics literature showed that economies which all satisfy the price-system frame still cluster into durable, differently-behaving types — though the clusters are softer than the first confident binary claimed. The historians and the heterodox showed that across centuries “capitalism” names a sequence of regimes, or a single world-system, or a class relation, or a nineteenth-century experiment — questions the price-system frame is not built to engage. The synthesis is not a tie-breaker among these. It is a disambiguation: the disagreement sits at three layers, and naming the layer is the whole move.
So carry the habit, not the verdict. One system, on the layer where prices coordinate. Many types, on the layer where institutions configure. Many systems sharing a name, on the layer of long history. The question only looked unanswerable because three questions were folded into one set of words. Pull them apart and you can hold all three answers at once without contradiction — which is what it means to think clearly about capitalism rather than to fight about it.