What survives of classical political economy?
Marginalism was supposed to retire Smith, Ricardo, Mill, and Marx in the 1870s — replace political economy with a tighter machine for allocation and welfare. Then in 2014 a French economist published a 700-page book called Capital, framed inequality in nineteenth-century terms, and the mainstream press called it the return of classical political economy. So which is it? The honest answer isn’t yes or no. It’s three answers at once — and telling them apart is the whole job.
以辩论图谱查看Why ask now
“Piketty’s is a book that nobody interested in a defining issue of our era can afford to ignore… Capital in the Twenty-First Century is an extremely important book on all fronts. It will be a long time, if ever, before anyone writes a sequel.”
— Paul Krugman, “Why We’re in a New Gilded Age,” New York Review of Books, May 2014
A Nobel laureate, reviewing for the most mainstream of intellectual venues, treating a book that opens with Ricardo and Marx as the most important economics of the decade. Reviewers reached for one phrase over and over: the return of classical political economy. That phrase is a puzzle. The discipline spent a century professionalizing away from political economy. So what, exactly, came back — and what never actually left?
Start with the cast. “Classical political economy” names a body of work built roughly between 1776 and 1867 by four figures who thought economics was inseparable from politics, morals, and history. Adam Smith founded the discipline with the Wealth of Nations (1776) and grounded it in the Theory of Moral Sentiments (1759). David Ricardo gave it its first real theoretical engine in 1817 — comparative advantage, and a tripartite division of national income among wages, profits, and rents. John Stuart Mill closed the tradition with his 1848 Principles, including a strange prophecy about a stationary state. And Karl Marx pushed the whole apparatus to its most ambitious extension and its most direct failure. Then, between 1871 and 1874, three men working separately — Jevons, Menger, Walras — rebuilt economics on marginal utility, and the political-economy tradition was declared superseded.
“What survives” is the wrong shape for a yes/no answer, and most public arguments founder by forcing it into one. The honest answer splits into three. First, folded-in durable insights: specific classical results that became permanent fixtures of the modern apparatus, often with the lineage scrubbed off — comparative advantage, the division of labor as a source of increasing returns, Ricardian equivalence as a benchmark. Second, framing imports the post-2008 turn brought back: classical ways of posing questions that the marginalist break had pushed out of bounds, and that twenty-first-century distributional and political-economy problems dragged back in. Third, a parallel heterodox tradition that never stopped running the classical apparatus at all — Sraffian, Marxian, world-systems. Three shapes. Keeping them apart is the whole job.
The lineage graph clusters the classical-era nodes — Smith, Ricardo, Malthus, Mill, Marx — against the marginalist rupture that follows them; the place to see the whole tradition mapped at once is the History of Economic Thought lineage view, and the redrafted era chapter that carries it is Ch.3 (Classical political economy: Smith, Ricardo, Malthus, Mill).
Why ask this in the 2020s rather than the 1970s? Because each of the four figures has a live contemporary signal, and they all got louder after 2008. Smith is in the middle of a quiet recovery as an institutional thinker: read the Theory of Moral Sentiments and the Wealth of Nations together, as Ronald Coase argued in 1976 and Amartya Sen has argued for decades, and Smith looks less like a prophet of unregulated markets and more like a theorist of how markets work inside moral and legal institutions.
Ricardo is having the most visible moment of the four. Piketty’s $r > g$ framing put functional income distribution — the split between capital and labor — back at the center of mainstream debate, in terms a Ricardian would recognize. And quietly, for sixty years, the Sraffian neo-Ricardians have kept Ricardo’s distributional apparatus running as a credentialed academic tradition.
Mill is the unlikely one. His 1848 vision of a “stationary state” — an economy that stops growing in scale and redirects its energy toward human flourishing — is the explicit ancestor of the contemporary degrowth movement (Jason Hickel’s Less Is More, 2020; Tim Jackson’s Prosperity Without Growth, 2009). Mill is being read as a policy prophet, with serious mainstream-econ pushback.
Marx never went away, but the academic Marxism doing real work now is not the textbook caricature: Anwar Shaikh’s quantitative classical-Marxian framework (Capitalism, 2016), the analytical Marxism of G.A. Cohen and John Roemer, and Immanuel Wallerstein’s world-systems analysis, alive in comparative sociology if not inside mainstream economics. Four figures, four living signals. The question isn’t nostalgic; it’s about what the discipline is currently doing.
Here is the position this walkthrough will defend, stated up front so the engagement can earn it. Classical political economy survives in more than the marginalist break suggests — but the interesting content is in the shape, not the slogan. It survives as folded-in durable insights, as framing imports the post-2008 turn recovered, and as a parallel heterodox tradition that kept the apparatus running unbroken. The next four stages make that decomposition concrete: Smith, then Ricardo, then Mill and Marx at the tradition’s terminus, then the marginalist rupture that frames what was lost and what came back.
Start with Adam Smith. He’s the right place to begin, because there are two Smiths in circulation — the libertarian patron saint and the moral-institutional theorist — and the one that survives in serious economics is not the one on the coffee mugs.
Smith at strongest
“How selfish soever man may be supposed, there are evidently some principles in his nature, which interest him in the fortune of others, and render their happiness necessary to him, though he derives nothing from it except the pleasure of seeing it… We endeavour to examine our own conduct as we imagine any other fair and impartial spectator would examine it.”
— Adam Smith, The Theory of Moral Sentiments, Part I & Part III, 1759
This is the opening of Smith’s other book — written seventeen years before the Wealth of Nations, and never disavowed. The man popular culture remembers for “the invisible hand” began his career arguing that human beings are wired for sympathy and that we judge ourselves through an imagined impartial spectator. Hold that next to the pin factory, and you get the two Smiths. The argument of this stage is that the surviving Smith is the one who wrote both books.
Smith’s single most-cited economic idea is the division of labor: the pin factory of Wealth of Nations Book I, where splitting one job into eighteen specialized operations multiplies output far beyond what the same workers could manage separately. For most of the twentieth century this read as a charming illustration of specialization. Then growth theory caught up with it. The modern home of the idea is increasing returns — the engine of endogenous growth, where the productivity of an economy rises with the scale and accumulated knowledge of its own activity rather than hitting the diminishing returns of the standard model.
In Paul Romer’s 1986 formulation, aggregate output depends on the stock of knowledge $K$ with a production function that does not exhibit the diminishing returns of the Solow model: $Y = A\,K^{\alpha} L^{\beta}$ with $\alpha + \beta > 1$ once the knowledge spillover across firms is counted. Because knowledge produced by one firm raises the productivity of all, the social return to investment exceeds the private return, and growth need not slow as capital accumulates. Smith’s “the division of labour is limited by the extent of the market” is the verbal ancestor of exactly this mechanism: a bigger market supports finer specialization, which raises productivity, which expands the market.
Smith’s claim was that specialization feeds on itself: a bigger market lets you slice work more finely, finer slicing makes each worker more productive, and the extra output grows the market again. The standard growth model assumed the opposite — that piling on more capital yields steadily smaller gains. Romer’s move was to put Smith’s self-reinforcing loop back in, by treating knowledge as something that spills over and never gets used up. Growth doesn’t have to peter out.
The formal home of increasing-returns growth is Economics Ch.13 §13.4 (Romer’s Endogenous Growth Model), and the institutional reading of Smith — markets as creatures of rules, firms, and property regimes rather than their absence — lives in Ch.18 §18.2 (North’s Framework).
The institutional Smith. The crude reading takes a few sentences from Wealth of Nations Book IV and turns Smith into an apostle of leaving markets alone. The full corpus refuses it. The Theory of Moral Sentiments spends a long book establishing that human conduct is governed by sympathy and the judgment of an impartial spectator — that self-interest operates inside a thick web of moral expectation. Ronald Coase, who knew something about how markets actually function, argued in his 1976 essay “Adam Smith’s View of Man” that the popular Smith is a fabrication: the real Smith assumed markets are embedded in institutions, law, and shared morality. Coase’s own 1937 “The Nature of the Firm” founded the modern study of why economic activity organizes itself into firms rather than spot markets — the lineage that runs through Williamson and North and quietly preserves the institutional Smith inside mainstream economics. Amartya Sen has spent a career insisting on the same point.
Smith as growth theorist. The other surviving Smith is the one whose division-of-labor insight became the backbone of how economists now think about long-run growth. Romer’s 1986 “Increasing Returns and Long-Run Growth” is recognizably Smithian even where it doesn’t cite him; so is Paul Krugman’s 1979–1980 new trade theory, which showed that countries trade not only on Ricardian comparative advantage but to exploit the increasing returns that come from larger markets and product variety. The lineage from “the division of labour is limited by the extent of the market” to modern increasing-returns growth and trade theory is one of the longest unbroken intellectual threads in economics — and it’s the subject of its own walkthrough, which traces the growth idea from the classical stationary state all the way to endogenous growth.
What didn’t survive. Two things. Smith’s labor-embodied theory of value — the idea that the “real price” of a thing is the labor it commands — was displaced by marginalism for reasons Stage 5 lays out; it was the wrong tool for the allocation questions the discipline ended up wanting to answer. And the “system of natural liberty” as a blanket policy prescription — the libertarian Smith — is a twentieth-century construction, assembled by quoting Book IV against the rest of the corpus. The honest reading is not that Smith was secretly a statist; it’s that he was a theorist of markets-within-institutions, and the institutional reading is the one that has held up.
“Adam Smith was the founding father of free-market economics… The invisible hand of the market, left unhampered by government, channels self-interest into the common good.”
— the standard op-ed invocation of Smith (Cato Institute / WSJ editorial-page register)
Was Smith the original libertarian?
The op-ed Smith is a free-market prophet who wanted government out of the way. The Smith who wrote two books is a theorist of markets embedded in moral and legal institutions. The gap between them is a selective Book IV excerpt versus the whole corpus.
Smith survives in two registers, and they line up with the first two of our three shapes. The institutional Smith is a framing the discipline keeps quietly recovering — the Coase–Williamson–North tradition that treats markets as creatures of institutions descends directly from the Smith who wrote the Theory of Moral Sentiments, and the lineage is traced in Ch.15 §13.3 (Coase as bridge). The growth-theorist Smith is folded in — the division of labor became increasing returns, load-bearing in endogenous growth and new trade theory, usually without the lineage named. The libertarian Smith is the one the discipline rejected when it actually read both books. The figure-level lineage for all of this is Ch.3 §6.1 (Smith and the founding of a discipline), and the growth-thread depth lives in the sibling walkthrough Why do economies grow? (classical stationary state through endogenous).
Smith built the cathedral. Ricardo wrote the single deepest chapter inside it. Comparative advantage is the durable Ricardo everyone knows. But there’s a second Ricardo — the one who put the distribution of income between landlords, capitalists, and workers at the center of economic theory — and that Ricardo’s framing is the one that came roaring back in 2014.
Ricardo at strongest
“To determine the laws which regulate this distribution [of the produce of the earth among the three classes of the community — rent, profit, and wages] is the principal problem in Political Economy.”
— David Ricardo, preface to On the Principles of Political Economy and Taxation, 1817
Read the sentence slowly. For Ricardo, the principal problem of economics was distribution — how the national product divides among landlords, capitalists, and workers. The marginalist revolution would later relegate that question to a corollary of factor-price theory. Then Piketty wrote a 700-page book putting Ricardo’s principal problem back at the center, called it Capital, and the mainstream press called it the return of classical political economy. The question for this stage: which Ricardo did they actually bring back?
Ricardo gave economics three things that are still in use, and they have very different survival stories. Comparative advantage (1817) is the one undergraduates meet: a country gains from trade by specializing where its opportunity cost is lowest, even if it’s worse at making everything. Formally unmodified in two centuries, it’s the foundation of every trade-theory chapter and the ancestor of Heckscher-Ohlin, Krugman’s new trade theory, and the gravity-model empirical program. Ricardian equivalence is the second: the proposition that, with forward-looking households, financing government spending by debt rather than taxes makes no difference, because households just save against the future tax bill.
Robert Barro’s 1974 formalization makes the benchmark precise. If households are infinitely-lived (or linked to their heirs by bequests) and capital markets are perfect, then for a given path of government spending the timing of taxes is irrelevant: a tax cut financed by debt, $\Delta T < 0$ today against $\Delta T > 0$ in present value later, leaves lifetime resources unchanged, so consumption is unchanged and private saving rises one-for-one with the deficit. The fiscal multiplier on a pure tax cut collapses to zero. The equivalence fails empirically — roughly half of households are credit-constrained or hand-to-mouth — but it survives as the benchmark against which real-world multipliers are measured as deviations.
Ricardian equivalence says: if the government hands you a tax cut today but you know it just means higher taxes later, you don’t spend the windfall — you save it to cover the future bill, and nothing changes. In the real world about half of households can’t do that; they’re living paycheck to paycheck and they spend the cut. So the equivalence is wrong as a description — but it’s the clean reference point that tells you how much the credit-constrained half is moving the multiplier.
The third Ricardo is the distributional one — the tripartite division of income that the provocation names as the principal problem. The formal apparatus for Ricardian equivalence is Economics Ch.16 §16.4 (Ricardian Equivalence); the labor-share side of the distributional question lives in Ch.21 §21.8 (the Labor Share); and the modern descendant of comparative advantage runs through the open-economy chapter.
Comparative advantage: folded in, unmodified. This is the cleanest case in the whole walkthrough. Ricardo’s 1817 demonstration that two countries both gain from trade even when one is more efficient at everything is a mathematical theorem, not a conjecture, and it survives every generalization — many goods, many countries, varying factor endowments. The textbook trade chapter that opens with “Ricardo” and ends with “gravity equation” is the single longest unbroken lineage in modern economics. Nobody calls it classical political economy anymore; it’s just trade theory. That’s what folded-in survival looks like — total absorption with the nameplate removed.
Ricardian equivalence: folded in as a benchmark. Ricardo floated the idea and doubted it; Barro’s 1974 “Are Government Bonds Net Wealth?” turned it into the modern reference case. What survived is not the empirical claim — that failed, because half of households can’t smooth consumption against future taxes — but the benchmark. Every estimate of a real-world fiscal multiplier is implicitly measured as a deviation from Ricardian equivalence. The full apparatus, including the credit-constraint rescue that brings the multiplier back to life, runs at depth in the sibling walkthrough Does government spending help the economy?. A classical result surviving precisely as the thing the modern apparatus measures departures from — that’s a distinctive mode of survival, and Ricardo owns the clearest example of it.
Distribution: the framing import, recovered via Piketty. Here is the contested one. Ricardo’s principal problem — how the product divides among classes — was pushed out by marginalist factor-price theory, which derived each factor’s share from its marginal product and treated the resulting distribution as the efficient by-product of allocation, not a problem in its own right. Before Piketty, mainstream economics had largely stopped asking what determines the long-run capital–labor split as a distributional question. Piketty’s 2014 data — two centuries of wealth concentration, the $r > g$ regularity — did the work the textbook had ruled out, and the framing he used to interpret it is recognizably Ricardian: capital-owners and wage-earners as distinct populations with distinct income dynamics. Acemoglu and Robinson’s institutionalism did the same recovery on the political-economy side. The load-bearing claim here is not that Piketty re-ran old data through a nostalgic lens; it’s that the question — what governs the functional distribution of income over the long run — is the classical question marginalist theory had quietly retired, and the post-2008 turn put it back in play. The intellectual-history version of this recovery lives in History of Economic Thought Ch.17 §12.4 (Institutions and inequality — the empirical turn).
The calibration flag. Be honest about the most-contested move in this walkthrough. Whether Piketty’s framing genuinely recovers the Ricardian tradition or merely decorates new data with an old vocabulary is a live disagreement. The position here is “recognizably Ricardian recovery” — the question Piketty re-opened is the one Ricardo posed and marginalism closed — but the flag stays on: the strength of the recovery, as opposed to the fact of the lineage, is what reasonable economists dispute.
The Sraffian parallel. And running underneath all of this, for sixty years, is the neo-Ricardian tradition that never stopped. Piero Sraffa’s 1960 Production of Commodities by Means of Commodities rebuilt Ricardo’s distributional apparatus on a rigorous footing, and the Cambridge capital controversy of the 1950s and 1960s — Robinson, Sraffa, and Pasinetti against Samuelson and Solow — won the technical points (reswitching, capital reversing) decisively enough that Samuelson conceded them in 1966. The mainstream agreed and then declined to restructure. Garegnani, Pasinetti, Kurz, and Schefold are the living lineage. This is the third shape — a parallel heterodox tradition keeping the apparatus running — and it’s real, credentialed, and small. Stage 5 returns to what the Cambridge controversy actually established.
“Piketty has written a truly superb book… a watershed in economic thinking. He has reminded us that the central questions of classical political economy — the distribution of income and wealth among the classes — never went away.”
— the 2014 reception of Capital in the Twenty-First Century (Krugman, NYRB; Solow, New Republic)
Did Piketty bring back classical political economy?
The reception called Capital in the Twenty-First Century the return of classical political economy. It did put Ricardo’s distribution question back at the center — but it did not revive the labor theory of value, and Piketty himself rejects it. So what came back, exactly?
Ricardo is the cleanest figure-level instance of the whole three-shape verdict, because all three shapes appear in one man’s work. Comparative advantage is folded in, unmodified and unnamed. Ricardian equivalence is folded in as the benchmark the apparatus measures deviations from. And the distributional Ricardo is the framing import the post-2008 turn recovered — with the calibration flag on the strength of the Piketty recovery, and the Sraffian neo-Ricardians as the parallel heterodox tradition that kept the apparatus running the whole time. The figure-level lineage is Ch.3 §6.2 (Ricardo: trade, distribution, rent). The full distribution lineage — Ricardo through Sraffa to Piketty across eras — is the subject of a thread-tracing sibling.
Smith built the cathedral. Ricardo wrote its deepest chapter. Mill closed the classical tradition with a strange prophecy that a growing movement is taking seriously again — and Marx kicked the doors off. The two of them bracket the tradition at its end: Mill on the liberal-reformist edge, Marx on the systemic-critique edge.
Mill and Marx at strongest
“I cannot… regard the stationary state of capital and wealth with the unaffected aversion so generally manifested towards it by political economists of the old school… the best state for human nature is that in which, while no one is poor, no one desires to be richer, nor has any reason to fear being thrust back, by the efforts of others to push themselves forward.”
— John Stuart Mill, Principles of Political Economy, Book IV, Chapter VI, 1848
Mill closes the classical tradition not with a growth model but with a value judgment: an economy that has stopped expanding in scale might be the best condition for human life, not a catastrophe. For a century this read as a Victorian eccentricity. Then a degrowth movement adopted it as a manifesto. Mill is one terminus of classical political economy — the liberal-reformist edge. Marx is the other — the systemic-critique edge. This stage takes them at their strongest, together, because they bracket where the tradition ends.
Both figures are evaluated against the same modern apparatus: the marginalist account of value, welfare, and production that replaced the classical labor-embodied theory. Mill’s stationary state is judged against the Solow steady state — an economy at constant per-capita capital, growing only with technology — and the welfare framework that defines what “better off” means. Marx’s labor theory of value and exploitation theory are judged against marginal-product factor pricing and the two welfare theorems, which deliver allocation-and-efficiency results the labor theory was never built to generate.
The contrast in one line. The labor theory sets the value of commodity $i$ as the socially necessary labor time embodied in it, $\lambda_i$, with prices supposed to track values up to a transformation. Marginalism sets the equilibrium price as the market-clearing $p_i$ where, at the optimum, $p_i = \mathrm{MU}_i/\mathrm{MU}_{\text{num}}$ for the consumer and $p_i = \mathrm{MC}_i$ for the producer. The First Welfare Theorem then certifies that competitive equilibrium is Pareto-efficient — a result with no analogue in the labor theory, because the labor theory was answering “where does surplus come from?” not “is this allocation any good?”. Mill’s stationary state, in Solow terms, is a steady state with the growth rate set to zero by choice rather than by the exhaustion of investment opportunities.
The labor theory and marginalism are answering different questions. Marx asked what, deep down, makes a coat trade for so much linen — his answer was labor. Marginalism asked a narrower question — given what people want and what firms can make, what prices clear the market, and is the result efficient? — and answered with marginal utility and marginal cost. The welfare theorems are the marginalist payoff: proof that competition reaches an efficient allocation. The labor theory can’t prove that, because it was never trying to.
The marginalist welfare baseline is in Economics Ch.11 §11.6 (The First Welfare Theorem); the production-and-cost foundation it rests on is Ch.5 §5.4 (Production Functions).
Mill’s stationary state, at strength. Take Mill on his own terms before judging him. His claim is not that growth is bad or that recession is good — the stationary state is a destination reached after material wants are met, not a depression. It is the condition where an economy stops chasing more scale and redirects its surplus capacity toward what Mill calls “the Art of Living”: leisure, culture, the moral and intellectual cultivation that endless accumulation crowds out. The contemporary degrowth movement — Jason Hickel’s Less Is More (2020), Tim Jackson’s Prosperity Without Growth (2009) — reads Mill as its ancestor and asks his exact question: is growth-without-end the right premise for an economy on a finite planet? That is a serious moral-political challenge, and the steelman is Mill’s, not the strawman where degrowth means engineered immiseration.
Mill as marginalist precursor. The complication that the rupture-narrative tends to hide: Mill was already pivoting toward marginalism from inside the classical tradition. His Principles draws the positive/normative distinction the discipline later formalized, and his treatment of value leans toward utility in ways that anticipate the 1871 revolution. Mill is the bridge figure — classical political economy handing off to what becomes marginalism. The discipline did not simply snap in 1870; it had been bending toward the marginal margin for decades, and Mill is where you can watch it bend.
Marx as the tradition’s terminus. Marx takes the classical labor theory of value — inherited from Smith and Ricardo — and pushes it to its most ambitious extension: surplus value, exploitation as a structural relation rather than a moral complaint, and the falling rate of profit as capitalism’s internal contradiction. It is the classical apparatus run at maximum range, and also its most direct failure point: the transformation from labor-values to equilibrium prices does not close cleanly (Bortkiewicz 1907; Samuelson 1971), which shuts the labor theory’s route to being an allocation theory. What survives is real but reconstructed. Class as an analytic unit folds into Acemoglu–Robinson institutionalism, heterogeneous-agent macro, and Piketty’s wealth data. The surviving academic Marxism — Anwar Shaikh’s quantitative work, the analytical Marxism of Cohen and Roemer that detaches exploitation from the labor theory entirely, Wallerstein’s world-systems — runs as a parallel tradition. This walkthrough takes Marx only as far as the classical-survival accounting needs; the full figure, five claims tested one at a time, is the dedicated sibling walkthrough Was Marx right about anything?
What didn’t survive, in both. Marx’s immiseration thesis — that capitalism would systematically drive real wages down — was refuted in the population he was theorizing about: real wages in advanced economies rose four-to-six-fold between his death and 1970, the standard-of-living debate that grounds the number is in Economic History Ch.7 (the Industrial Revolution). The proletarian revolution arrived in agrarian peripheries, not advanced industry — the inverse of the prediction. Central planning as a superior production technology failed by margins no serious economist now defends, the record in Economic History Ch.15 (Communist Economies). Deterministic historical materialism — each mode of production necessarily yielding to the next — the record refuses the necessity. And Mill’s stationary state, as a near-term trajectory prediction for advanced economies, has not been adopted by mainstream growth theory; it survives as a live ethical-political position, not a forecast.
Mill survives as the moral-political prophet whose stationary-state framing is alive in contemporary degrowth discourse — a framing import — without being adopted by mainstream growth theory as a trajectory; and as the marginalist precursor who complicates the clean rupture story. Marx survives in pieces: class as analytic unit is folded in (Acemoglu–Robinson, heterogeneous-agent macro, Piketty’s implicit class accounting); the surviving Marxian apparatus — Shaikh, analytical Marxism, world-systems — is the parallel heterodox tradition; and the post-2008 Marxian revival is the framing import the distributional turn brought back. The labor theory’s allocation route closes, but analytical Marxism’s exploitation-without-the-LTV move keeps the substantive Marxian claim running on different machinery. The lineage anchors are Ch.3 §6.5 (Mill: the classical synthesis), the dedicated Marx chapter Ch.4 §7.6 (what survived of Marx and why), and the modern-Marxian-readings stream in Ch.17 §12.4. Depth on Marx-as-individual-figure lives in Was Marx right about anything?; the Mill-stationary-state node sits inside the growth thread Why do economies grow?
Four figures, four pieces of survival accounting. But one question runs underneath all of them: what did classical political economy lose when the discipline professionalized in the 1870s? That loss is the other half of the ledger — and naming it is what keeps the three-shape verdict honest.
The marginalist break and the verdict
“Repeated reflection and inquiry have led me to the somewhat novel opinion, that value depends entirely upon utility… Labour is found often to determine value, but in a wholly indirect manner, by varying the degree of utility of the commodity through an increase or limitation of the supply.”
— William Stanley Jevons, preface to The Theory of Political Economy, 1871
This is the moment the rupture is announced. Jevons — and, in the same few years, Menger in Vienna and Walras in Lausanne — declared the classical labor theory of value wrong at its root: value comes from utility at the margin, not labor embodied. Within a generation the discipline had rebuilt itself around constrained optimization and market-clearing, and “political economy” became “economics.” Before tallying what survived, this stage asks the symmetric question: what did the break cost?
No new model here — the apparatus is the rupture itself. Between 1871 and 1874, Jevons, Menger, and Walras independently arrived at marginal utility as the foundation of value. Walras built the first general-equilibrium system — every market clearing at once; Marshall’s 1890 Principles made it teachable; Pareto turned welfare into a formal criterion; and von Neumann–Morgenstern (1944) and Arrow–Debreu (1954) completed the toolkit. What the marginalists replaced: the classical labor-embodied theory of value, and with it the political, dynamic, aggregate, and class-distributional framings that came bundled with classical political economy. What they put in its place: constrained optimization, market-clearing, marginal-product factor prices, and the welfare theorems as the discipline’s grammar for what counts as a good argument. The lineage chapter for the rupture is History of Economic Thought Ch.5 §8.1 (Three revolutions at once: Jevons, Menger, Walras).
The marginalist break at strength. Resist the temptation to cast 1871 as the year economics went wrong. The marginalists did real intellectual work the classical apparatus could not. They produced a coherent, unified theory of allocation and welfare; mathematical tractability that let arguments be checked rather than asserted; and the welfare theorems, which are genuine theorems proving that competitive equilibria are efficient and that any efficient allocation can be reached by competition plus redistribution. The labor theory of value could generate none of these results, because it was answering a different question. The replacement of the LTV for allocation and welfare was correct, and pretending otherwise concedes the argument to nostalgia.
What the discipline lost when it professionalized. The break was not free. It cost four framings, and naming them keeps the survival accounting symmetric. The political-and-moral framing: classical PE wrote about state and market jointly, and treated the justice of distribution as in-scope; professionalization bracketed the political register as outside economics proper. The dynamic-and-aggregate framing: Smith on long-run growth, Ricardo on long-run distribution, Mill on the stationary state, Marx on capitalism’s dynamics — the break centered the discipline on static equilibrium and demoted dynamics to a subfield. The class-distributional framing: marginal-product factor pricing and the representative consumer replaced class as the analytic unit. And the moral question the LTV was built to ask: whether the distribution of factor ownership is just, beyond whether factors are paid their marginal products — a question marginalist welfare economics deliberately set aside.
The critique from inside. The sharpest moment when the marginalist apparatus’s incompleteness was named from within the discipline is the Cambridge capital controversy (1953–1966). Joan Robinson, Piero Sraffa, and Luigi Pasinetti against Paul Samuelson and Robert Solow, over whether “capital” can be aggregated into a single quantity with a well-defined marginal product. Cambridge UK won the technical points — reswitching and capital reversing are real, and Samuelson conceded them in his 1966 QJE paper — and the mainstream agreed and then declined to rebuild its production functions around the result. That is the cleanest case of the discipline being shown an incompleteness from inside and choosing to keep its apparatus anyway, on grounds of tractability. The controversy lives in the marginalist-revolution and counter-revolution chapters; what the later counter-revolution sacrificed to push distribution back out of bounds is catalogued in Ch.10 §11.5 (what the counter-revolution sacrificed).
The three-shape integration. Against that loss, here is the survival, made concrete. (i) Folded-in durable insights: comparative advantage (Ricardo, unmodified); the division of labor as scale economy (Smith → Romer 1986); Ricardian equivalence as benchmark (Ricardo → Barro 1974); marginal cost in production, which crossed the rupture intact; the institutional Smith (Coase, Williamson, North). These are permanent fixtures of the modern apparatus, usually with the lineage scrubbed off. (ii) Framing imports the post-2008 turn brought back: the distributional framing (Ricardo’s tripartite division, recovered via Piketty 2014); political economy (Acemoglu–Robinson, the state-capacity literature); and the moral question the LTV asked, alive in Cohen and Roemer’s analytical Marxism and in political philosophy if not in welfare economics — with the calibration flag on the strength of the Piketty recovery. (iii) A parallel heterodox tradition keeping the apparatus running: Sraffian neo-Ricardian (Sraffa 1960; Garegnani, Pasinetti); Marxian political economy (Shaikh 2016; analytical Marxism); world-systems (Wallerstein 1974, alive in comparative sociology). Real, credentialed, and small relative to the mainstream — documenting its survival honestly is not the same as claiming it runs the discipline. The post-2008 absorption arc that ties these together is the subject of Ch.17 §12.6 (Convergence or fragmentation?).
The position, plainly. Classical political economy survives in more than the marginalist break suggests — but the content is the shape, not the slogan. The losses are real and named: the political-and-moral framing, the dynamic-and-aggregate framing, the class-distributional framing, and the moral question marginalism set aside. The recoveries are real and named: folded-in durable insights that became the apparatus, framing imports the post-2008 turn recovered, and a parallel heterodox tradition that never stopped. The walkthrough lands roughly where Brad DeLong’s Slouching Towards Utopia (2022) lands — classical political economy as live, load-bearing, partially absorbed — with the three-shape accounting made explicit. The answer is neither “everything survived” nor “the marginalist break broke it”; it is the decomposition, and the decomposition is the point.
Short reading list: Smith’s Wealth of Nations and Theory of Moral Sentiments read together; Sraffa’s Production of Commodities by Means of Commodities (1960); Piketty’s Capital in the Twenty-First Century (2014); Shaikh’s Capitalism (2016); Wallerstein’s The Modern World-System (1974); Coase’s “Adam Smith’s View of Man” (1976); and DeLong’s Slouching Towards Utopia (2022). For the full Marx accounting, the reading list of Was Marx right about anything? carries it.
This walkthrough does the four-figure-plus-rupture integration; no sibling does that. But each piece of it opens onto a deeper treatment elsewhere:
Where this leaves us
We started with a Nobel laureate calling a 2014 book the return of classical political economy, and a phrase that didn’t obviously make sense, because the discipline had spent a century professionalizing away from political economy. Five stages later, the phrase has a precise meaning — three meanings, actually. Smith survives as the institutional theorist and the growth theorist, not the libertarian prophet. Ricardo is the cleanest case of all three survival shapes in one body of work: comparative advantage folded in, Ricardian equivalence folded in as a benchmark, the distribution question recovered as a framing import. Mill survives as the stationary-state prophet alive in degrowth and the marginalist precursor who blurs the rupture. Marx survives in reconstructed pieces and a parallel academic tradition. And the marginalist break that was supposed to end all of this turns out to have lost as much as it gained — the political, dynamic, class, and moral framings whose post-2008 return is half of what “survival” means.
The honest verdict refuses both slogans. Classical political economy did not survive intact, and it was not retired in 1871. It survives in three identifiable shapes — folded-in durable insights, framing imports the post-2008 turn recovered, and a parallel heterodox tradition keeping the apparatus running — and the only answer that beats “more than the textbook admits, less than the revival hopes” is the one that says which shape each surviving piece belongs to. The next time someone tells you Smith was a libertarian, or that Piketty proved Marx right, or that the marginalists settled everything, you have the decomposition to push past all three.