Keynes published his general theory of employment in 1936, into a world where millions stood idle beside silent factories, and he wrote it as a rescue operation for an economic order he wished to preserve. Marx had died more than half a century earlier, leaving behind an account of crisis that treated breakdown not as an accident befalling capitalism but as one of its regular products. Placed side by side, the two men look like opposites arranged for a debating society: the Cambridge reformer against the revolutionary exile, demand against production, policy against transformation. That arrangement is the one most textbooks still use, and it is misleading in ways that cost the reader real understanding, because the two frameworks agree on more than either camp admits, disagree at a deeper level than the slogans suggest, and were joined, historically, by a figure whom the standard comparison almost never mentions.

The missing figure is Michal Kalecki, a Polish economist who worked out the central result of Keynesian economics independently, from Marxian starting points, before the General Theory appeared, and who then wrote a short paper explaining why the Keynesian remedy, though economically available, would be politically resisted for reasons that Marxian class analysis predicts. His existence changes the shape of the comparison. A contest between two sealed systems becomes a three-cornered inquiry in which the most instructive position may be the one that borrows from both. This article therefore proceeds by giving each framework its strongest form before comparing them, because a comparison of weakened versions settles nothing. Keynes will be presented as his ablest defenders present him, Marx as his ablest interpreters present him, and the verdicts, where verdicts are possible at all, will be assigned axis by axis rather than awarded to a camp. Readers who want the wider map of how Marxian economics differs from the mainstream tradition it argues with can find it in the comparison of Marxist and mainstream economics, which treats method across the whole field; the task here is narrower and sharper, confined to the single problem of crisis.

What is the main difference between Marx and Keynes on crisis?

Both reject the idea that markets self-correct. Keynes locates crisis in deficient demand and volatile expectations, which policy can remedy. Marx locates it in the accumulation of capital and the production of surplus value, which policy can postpone but not remove without changing ownership.

Everything else in the comparison unfolds from that compressed answer. Notice what the compression already does. It refuses the familiar caricature in which Keynes stands for spending and Marx for revolution, as if one offered a technique and the other a temper. Both men were theorists of a monetary production economy, both denied that a slump is a punishment for past excess that patience will cure, and both built their accounts around the decisions of those who own capital and commit it in advance of any guaranteed sale. The difference lies in where each locates the failure. For Keynes the failure is a shortfall of spending against the output a fully employed economy could produce, driven by the unstable expectations of investors and correctable by public action. For Marx the failure germinates inside the production process itself, in the relation between the surplus extracted there and the conditions under which it can be realised, expanded, and extracted again. One theory looks at the circuit of capital from the side of its interruption; the other looks from the side of its engine.

A reader should also know, before the detail begins, what kind of difference this is. Some theoretical disputes are about facts, and observation settles them. Some are about which mechanism dominates an episode, and observation helps without settling. A few are about the depth at which explanation should stop: Keynes deliberately treated the distribution of income and the ownership of wealth as givens, data for his inquiry rather than objects of it, whereas for Marx those givens are precisely the things to be explained. When two inquiries stop at different depths, they can both be internally successful and still collide, because each takes as fixed what the other treats as the problem. Much confusion in the secondary literature comes from scoring one framework by the questions of the other. The discipline used throughout the pages ahead is to ask, of each claim, which question it answers, and only then to ask how well.

The shared refusal: Say’s law and the myth of self-correction

The agreement comes first, because it is substantial and because popular accounts erase it. Both frameworks reject the doctrine associated with Jean-Baptiste Say, under which production itself generates the income needed to purchase what has been produced, so that a general glut is impossible and downturns must be local, temporary, or caused by interference. That doctrine, refined by Ricardo and repeated in various forms ever since, treats money as a veil: goods exchange against goods, saving flows automatically into investment, and any income not consumed is spent by someone else through the credit system. On that view, an economy-wide failure to sell can only be a brief maladjustment of relative prices. Both Marx and Keynes built their mature work on the denial of exactly this.

Marx’s denial came first and cuts at the level of form. In a monetary economy, sale and purchase are separate acts, divided in time and person. The peasant who sells grain holds money and may wait; the merchant who buys must sell again before his capital returns. Marx argued in the first volume of Capital, and more expansively in his notebooks on earlier economists, that the mere separation of sale and purchase contains the abstract possibility of crisis, because nothing guarantees that the money realised in one sale flows back as demand for someone else’s product at the moment and in the proportion required. Money held back is not a neutral delay. It breaks a chain of payments in which each link’s solvency depends on the next link’s sale, and the breakage can generalise. What the classical doctrine described as a technical impossibility, Marx described as a standing possibility inscribed in the commodity form itself, awaiting only the developed conditions that turn possibility into periodic actuality.

Keynes’s denial arrived by a different road but reached a cognate place. His target was the classical theory of interest, under which saving and investment are equilibrated by the rate of interest, so that a decision to save more automatically finances more investment and total spending is preserved. Against this he argued that saving and investment are made by different people for different reasons, that income itself does the adjusting when the two plans diverge, and that an attempt by everyone to save more can shrink income until actual saving falls back to the level of a depressed investment: the paradox of thrift. Where the classical picture had a mechanism guaranteeing that abstention becomes construction, Keynes put a multiplier working in reverse. The economy, on his account, possesses no home position at full employment to which it gravitates; it can settle, and remain settled, at many levels of activity, most of them wasteful.

Two consequences follow from noticing the agreement. First, the popular contrast between a Marx who blamed the system and a Keynes who blamed bad luck is false at the root: Keynes treated crisis and chronic underemployment as endogenous products of a monetary economy left to itself, not as shocks arriving from outside it. His disagreement with Marx concerns the depth and remedy of the endogeneity, not its existence. Second, both men wrote against an orthodoxy that explained mass unemployment by wages being too high, and both denied, for different reasons, that cutting wages across the whole economy could restore prosperity. Those reasons repay attention, because they expose the machinery of each theory more clearly than the shared conclusion does. The Marxian analysis of the demand side of crisis, which runs through the family of arguments gathered under overproduction and underconsumption, shows how one branch of Marx’s inheritance approached territory that Keynes later occupied from another direction.

There is also an agreement about time. Neither thinker accepted the consoling doctrine that the long run heals what the short run breaks, because both saw the path itself doing damage: idle workers lose skills and standing, idle firms lose markets, debts incurred in boom conditions are denominated in money that slump conditions make heavier. Keynes compressed the point into a famous shrug about the long run and mortality. Marx’s version is structural: the crisis is itself part of how the system restores the conditions for another round of accumulation, destroying values, cheapening assets, and disciplining labour, so that recovery and damage are the same event viewed from different positions. That divergence inside the agreement, crisis as malfunction against crisis as mechanism, will recur at every stage of the comparison, and the reader should hold it in view from here onward.

The older battlefield: the glut controversy behind both men

The dispute between the frameworks has a prehistory that explains why each man wrote as he did, because the question whether capitalism can produce more than it can sell was fought once before, in the decades after the Napoleonic wars, and the losers of that first fight shaped the orthodoxy both Marx and Keynes later attacked. Thomas Robert Malthus argued against Ricardo that deficient demand was possible, that landlords’ unproductive consumption performed an economic service, and that a nation could save its way into stagnation; the Sismondi of the Nouveaux Principes reached a kindred position from the observation of crises themselves. Ricardo won the argument within the profession, and his victory installed the doctrine, later pinned to Say, that general gluts were impossible and that every production was a demand for something else. The victory was so complete that when crises recurred with metronomic regularity through the nineteenth century, the orthodoxy explained each one away as a local disproportionality or an abuse of credit, never as evidence about the system. Marx, who read the glut controversy closely and wrote hundreds of notebook pages on it, sided with the losers on the possibility question while rejecting their remedies and their reasoning: Malthus had seen a real phenomenon and explained it with a defence of parasitic consumption, which Marx treated as apologetics accidentally brushing against a truth.

That prehistory matters for reading Keynes, because his admirers and detractors alike sometimes present the General Theory as a thunderbolt from a clear sky, whereas its author knew he was reopening a buried case. Keynes praised Malthus’s intuition in his biographical essays, regretted that Ricardo’s line had conquered, and understood his own principle of effective demand as restoring, with proper machinery, what the glut controversialists had glimpsed. The parallel with Marx is exact in structure and opposite in allegiance. Both men dug beneath the Ricardian settlement; Marx radicalised the excavation until it reached the production relation, Keynes stopped it at the level of demand and expectations, and each regarded the other’s stopping place as the evasion. A comparison that begins in 1936 therefore starts too late: the frameworks are rival completions of a nineteenth-century argument that orthodoxy had declared closed, and their agreement against Say is less a coincidence than a shared inheritance from the losing side of that argument.

The nineteenth century also supplied both frameworks with their evidence base, and the character of that evidence shaped their different emphases. The crises Marx watched, in 1847, 1857, 1866, and 1873, were crises of a recognisably industrial capitalism bound together by bills of exchange, railway shares, and the gold standard: they began in speculation and credit, burst into bank failures and demands for gold, and resolved through bankruptcies and unemployment. A theorist formed by that sequence would naturally treat the monetary surface and the productive depths as one event with two faces, and would build credit into the theory’s core rather than bolting it on. The crisis Keynes watched was of another kind: not a violent liquidation followed by brisk recovery, but a long, grinding stagnation in which the economy settled below its capacity and stayed, with British unemployment insured above a tenth of the workforce for two decades and the American collapse of the early nineteen thirties destroying a quarter of output without generating the classical rebound. A theorist formed by that experience would naturally ask not what periodically wrecks the machine but what permits it to idle indefinitely. The frameworks differ, in part, because the pathologies differed: Marx theorised the storm, Keynes the doldrums, and much of the subsequent argument consists of each tradition insisting that its pathology is the fundamental one of which the other is a variant.

The road Keynes travelled: from the Treatise to the general theory

Keynes arrived at his crisis theory by failing first, and the failure is instructive. His Treatise on Money of 1930 attempted to explain cycles through divergences between saving and investment mediated by the banking system, and it contained, in the widow’s cruse image, the germ of the profit logic Kalecki was formalising in Warsaw at the same moment: entrepreneurs’ spending returns to them as profit, so investment validates itself in aggregate. But the Treatise lacked a theory of output. It could describe prices whirling around a given level of activity; it could not explain what set the level. The young Cambridge critics around the Circus, Richard Kahn and Joan Robinson and Austin Robinson and Piero Sraffa among them, pressed the gap, Kahn supplied the multiplier as the missing transmission, and Keynes rebuilt the argument around the determination of employment itself. The General Theory of 1936 is the rebuild: consumption governed by a psychological propensity, investment by the marginal efficiency of capital against the interest rate, interest by liquidity preference, and employment emerging where the schedules cross.

Two features of that journey explain choices the finished theory sometimes hides. The first is that Keynes’s target narrowed as his argument deepened. He began wanting a theory of money and prices for a world of banks; he ended with a theory of employment for a world of mass joblessness, and the apparatus bears the marks of the migration, above all in its treatment of the long period as somebody else’s problem. Marxian critics who complain that the theory cannot discuss accumulation’s laws are complaining about a deliberate renunciation, announced in the book’s architecture. The second feature is rhetorical. Keynes wrote to persuade a profession and a Treasury, and he framed his heresies as restorations, claiming classical ancestry for his conclusions wherever the claim could be stretched. The Marxian tradition, formed in exile and opposition, framed its conclusions as ruptures. Readers comparing the frameworks should discount both costumes: the Keynesian argument is more destructive of orthodoxy than its tailoring suggests, and the Marxian argument more continuous with classical political economy, with Ricardo and with the glut controversialists, than its own tailoring admits.

The political Keynes travelled with the theory and belongs in the record of the comparison. He had attacked the Versailles settlement’s economics immediately after the First World War, advised on the financing of the second while designing the postwar monetary order at Bretton Woods, and spent his career arguing that the state could be made intelligent enough to spare capitalism its self-destructive episodes. His famous impatience with Marx was of a piece with this project: he believed the class war framing obstructed the administrative reforms that could, in his view, dissolve its causes, and he said so publicly in his essays on Russia and privately with greater venom. Marxian writers return the compliment by classifying him as the most gifted defender the system produced, and the classification, intended as criticism, contains an acknowledgement that honest comparison should preserve: Keynes understood capitalism’s crisis tendency from inside its own policy class, and his remedy was designed by a man who knew exactly which interests would have to accept it. The 1943 Kalecki paper can be read as the reply of a man who knew the same interests from the receiving end.

Keynes at full strength: effective demand and the volatile investor

The Keynesian framework deserves presentation at full strength, since it remains the reference point for most macroeconomic policy discussion and since caricaturing it as a doctrine of deficits insults a subtle construction. Its core is the principle of effective demand. Entrepreneurs produce what they expect to sell at a profit, and employment settles where the aggregate demand price for output equals the aggregate supply price that just induces firms to provide it. Nothing in that determination guarantees that the meeting point coincides with full employment. If expected proceeds fall short, firms contract together, incomes shrink, and the economy finds a resting place with plant idle and workers queuing, not because anyone chose idleness but because each firm’s rational response to the others’ contraction confirms the pessimism that started it.

Investment does the decisive work in this machinery, and investment, for Keynes, is the least governable of magnitudes. A factory, a railway, a mine: these commit money for returns spread over decades, under conditions no calculation can domesticate. Keynes insisted that the knowledge on which long-term expectations rest is flimsy, that conventions substitute for knowledge, and that the state of confidence can swing with news, mood, and the behaviour of others. His phrase for the spontaneous optimism that carries enterprise forward despite the darkness was animal spirits, and the corollary matters as much as the phrase: when animal spirits dim, the collapse of investment is not a correction but a cascade, because each postponed project removes the income that would have justified the next one. The stock exchange, in his telling, amplifies the instability by pricing long-lived assets minute by minute, inviting enterprise to become a byproduct of speculation rather than its master.

What did Keynes mean by animal spirits?

Keynes used the term for the spontaneous optimism and urge to action that lead entrepreneurs to invest despite incalculable uncertainty. Because future returns cannot be computed from probabilities, investment rests on convention, confidence, and mood, which shift together and without warning, dragging output and employment down with them.

Money, in this framework, is no veil. Keynes’s theory of liquidity preference holds that wealth holders choose between money and less liquid assets according to their trust in the future, and that the rate of interest is the reward for surrendering liquidity rather than a price that balances saving against investment. In a frightened economy the desire to hold money can rise without limit at low interest rates, monetary policy pushes on a string, and the central bank loses its leverage precisely when leverage is most needed. This is the analytical root of the Keynesian case for fiscal action: where private spending collapses and the interest mechanism jams, public spending can substitute directly, employing the idle resources whose idleness was the whole problem. The doctrine is frequently summarised as spending one’s way out of a slump, but the logic is stricter than the slogan. The spending works because resources are unemployed; it borrows against capacity that would otherwise produce nothing at all.

Why do wage cuts fail to cure unemployment for Keynes?

Because wages are both a cost to each employer and the income from which most spending comes. A general reduction lowers prices and demand alongside costs, leaves real burdens of debt heavier, and may worsen expectations, so the economy can settle at lower employment rather than returning to full use of its capacity.

The wage question separates Keynes from his classical opponents as sharply as any other. For a single firm, lower wages mean lower costs and perhaps more hiring; the classical error, as he diagnosed it, was to generalise that arithmetic to the whole economy, where one firm’s wage bill is another firm’s market. Across the economy, a general wage reduction drags prices down with it, so real wages need not fall at all, while the real weight of every money debt grows and expectations sour. He concluded that a flexible wage policy was not merely cruel but unreliable engineering, and that a stable money wage combined with managed demand offered the safer route to sustained employment. Note the character of the argument: it is presented as a claim about how the machine behaves, not about justice, and its author regarded the distributional struggle over wages as a political given lying outside his model’s jurisdiction.

Why is liquidity preference not a doctrine about saving?

Liquidity preference concerns the form in which wealth is held, not the amount saved: it is the desire to keep assets saleable and safe when the future is distrusted. It sets the rate of interest as the reward for parting with liquidity, and it can rise sharply exactly when spending is collapsing.

Two further pieces complete the strongest form of the Keynesian case. The first is the treatment of uncertainty as irreducible. Keynes had written on probability as a young philosopher, and the General Theory carries that formation into economics: for the decisions that matter most, there exists no scientific basis on which to form calculable probabilities at all, so agents fall back on convention, on assuming the present will persist, until the assumption breaks and breaks suddenly. This is not irrationality sprinkled over a sound machine; it is the condition of the machine. The second piece is political in the plain sense. Keynes believed the argument’s conclusion in advance of its proof: he wanted a capitalism that could be kept, and he said plainly in the book’s closing chapter that a somewhat comprehensive socialisation of investment would prove the only means of securing an approximation to full employment, while leaving consumption and much enterprise private. Abolition of the rentier, the euthanasia of the functionless investor as interest rates fell, state guidance of the propensity to consume through taxation: the programme was radical by the standards of his class and profession, and it was offered entirely within the preservation of private property in the means of production. Any account that paints Keynes as a timid technician has not read his final chapter.

His view of Marx, for the record, was hostile and largely uninformed by close reading, a matter the frequently asked questions below take up in detail. The hostility matters here only because it confirms that the convergences traced in this article are structural rather than diplomatic: Keynes was not borrowing from Marx and would have disliked the suggestion. When two frameworks built in mutual indifference arrive at a shared rejection of self-correction, the agreement is evidence about the object they both studied, the monetary production economy, and not about influence. The steelmanning discipline cuts both ways, however, and the Keynesian framework carries genuine exposures that its ablest defenders acknowledge. Its theory of distribution is thin: profits and wages appear as outcomes of spending and bargaining, not as the structured product of a production relation, so the model can describe a slump’s arithmetic better than its social anatomy. Its state is largely an exogenous rescuer, assumed willing and able to act, an assumption the Kalecki bridge will put under strain. And its time horizon is the short period: given equipment, given technique, given institutions, which is why its deepest critics, friendly and hostile alike, describe it as a special theory of the slump rather than a general theory of the system’s motion.

The multiplier and the schedule: Keynes’s machinery in detail

The machinery deserves one pass at close range, because the comparison with Marx is often conducted between a detailed Marx and a sloganised Keynes, and the asymmetry corrupts the scoring. Begin with the consumption function. Keynes proposed, as a fundamental psychological law, that households spend a rising absolute amount as income rises but a falling proportion of it, so that a growing economy must find ever larger investment outlets for the saving its prosperity generates. The proposition turns Malthus’s intuition into a systematic gap: the richer the economy, the wider the chasm between what it can produce and what its consumption habits will absorb, and the more its fate depends on the volatile decision to invest. Notice how close this brings Keynes to a stagnation thesis. A mature economy with saturated consumption needs and a slowing population, he speculated in his later wartime writings, might face a chronic deficiency of investment opportunity, a speculation that Hansen and others developed into the secular stagnation doctrine and that Steindl, from the Kaleckian side, grounded in oligopoly. Marxian readers who meet the consumption function for the first time are often startled by the family resemblance to underconsumption, and the resemblance is real; the difference lies in the trigger, since Keynes’s crisis arrives when investment fails to fill the gap, not when consumption itself collapses.

The multiplier connects the trigger to the damage. Kahn’s insight, absorbed into the General Theory, is that an initial act of investment spending becomes income for its recipients, of which a proportion is respent, becoming income again, the chain converging to a total change in income that is a multiple of the original impulse. Run backwards, the same machinery is a wrecking device: a fall in investment annihilates income in cascades, tax receipts fall, relief spending rises, and the public finances deteriorate for reasons that have nothing to do with profligacy, a point Keynes’s followers pressed on treasuries that read slump deficits as misconduct. The multiplier also disciplines the comparison with Marx at an unexpected place. Marx’s reproduction schemes are, formally, an interdependence engine of the same family: expenditure in one department constitutes demand for another’s product, and a failure in one propagates through the table. The two frameworks’ shared hostility to the doctrine that demand failures are impossible rests, at bottom, on a shared mathematical intuition about circular flows, reached by the schemes in one case and by the multiplier in the other, and the historians of thought who have set the two engines side by side have generally concluded that the schemes are the more general construction and the multiplier the more usable one.

The schedule of the marginal efficiency of capital completes the machinery, and it is where the Marxian reader must resist a translation error. The marginal efficiency is the expected rate of return over cost on a new asset, formed under the uncertainty described in the animal spirits discussion. It is not the rate of profit on the existing capital stock, and Keynes’s theory contains no account of that magnitude at all. When investment collapses in his narrative, it collapses because expectations of yield have fallen below the rate of interest set by liquidity preference, a relation between a psychological schedule and a monetary price. Nothing in the mechanism requires profitability to have deteriorated in the past; the slump can begin in pure anticipation. Marxian critics regard this as the theory’s characteristic idealism, causality unmoored from production, while Keynesian defenders reply that anticipations are themselves facts with balance sheets attached, and that a theory which waits for profitability to deteriorate before permitting a crisis will be perpetually surprised by financial ones. The axis is genuinely contested, and it resurfaces below in the discussion of the profit rate, where the question becomes why Keynes built his theory without the variable Marx placed at its centre.

One more mechanism must be credited to the Keynesian side before the comparison proceeds, since its absence is the most common defect of Marxian summaries: the theory explains why the slump does not self-correct through falling interest rates even where wages and prices are flexible. As incomes fall, the demand for money for transactions falls with them, releasing balances that should depress interest rates and revive investment; but the same collapse that releases the balances also poisons the expectations on which investment depends, and at the limit the demand for liquidity becomes absolute, absorbing every release without effect on the rate. Deflation, meanwhile, redistributes wealth to rentiers, swells the real burden of debt in the manner Irving Fisher formalised in his debt deflation theory of 1933, and teaches entrepreneurs to postpone every commitment in expectation of cheaper prices tomorrow. The classical self-correction story fails, in short, not at one point but at every point, and Keynes’s systematic enumeration of the failures is the permanent achievement on which his framework’s claim to seriousness rests.

Marx at full strength: accumulation and the production of surplus value

The Marxian framework begins where Keynes declined to go: inside production. Capital, on this account, is not a stock of useful things but a process, money advanced to buy labour power and materials, set to work, and withdrawn enlarged. The enlargement, surplus value, originates in the gap between what labour power costs, its value set by the goods needed to reproduce the worker, and what labour performs during the working day. Everything else in the crisis theory hangs from that hinge. Because the surplus must first be produced and then realised by sale, the system has two distinct sites of failure, and because competition drives each capital to expand or die, the drive to accumulate is not a preference of greedy individuals but a compulsion enforced on every firm that wishes to survive. Crisis, in this construction, is not the interruption of an otherwise harmonious process. It is what the process periodically does to itself.

Where does crisis begin in the Marxian circuit?

It begins in production, where labour power is set to work and surplus value is created, and it matures as that surplus struggles to be realised through sale at prices that validate the capital advanced. Crisis is the violent reconciliation of production and realisation when the two drift too far apart.

Several mechanisms branch from the trunk, and the tradition has argued for a century and a half about their ranking, an argument this article deliberately does not settle, since the families of Marxian crisis theory are mapped in their own right in the series guide to the Marxist crisis theory. For the comparison with Keynes, three mechanisms matter most. The first is disproportionality: investment decisions are made by competing capitals without a coordinating plan, so branches of production expand at mismatched rates, and the interlocking structure of inputs and outputs, which Marx modelled in the reproduction schemes of the second volume of Capital, periodically fails to mesh. The second is the pressure on profitability: as accumulation proceeds, the mass of machinery and materials set in motion by each worker grows faster than the living labour that alone produces surplus value, pressing on the rate of profit and, with it, on the incentive and capacity to invest. The third is the demand constraint in its specifically Marxian form: wages are simultaneously a cost each capital strives to compress and the source of most consumption demand, so the system’s success at its own object, cheapening labour, chronically undermines the market for the swelling output of consumer goods.

Compare those mechanisms with the Keynesian machinery and the geography of the disagreement becomes visible. Keynes’s slump begins in the sphere of circulation and expectation: investment falters because confidence falters, and production follows spending down. Marx’s crisis begins in the sphere of production and profitability: the surplus that justifies production becomes harder to extract at the required rate, investment falters because its reward falters, and spending follows production down. The sequences resemble each other once running, since a modern economy transmits any failure everywhere within months, which is why the two accounts can look interchangeable in newspaper commentary. They are not interchangeable at the point of origin, and the point of origin decides what each theory regards as a remedy. A failure of confidence can be repaired by a determined borrower of last resort. A failure rooted in the conditions of surplus production can be palliated by the same borrower, on the Marxian view, but the palliative works by sustaining an accumulation whose difficulties then reappear in altered form: as public debt, as inflation, as asset inflation, or as a deferred and larger reckoning.

The Marxian account also assigns crisis a function, which the Keynesian account does not. In the third volume of Capital and in the notebooks, devaluation is the system’s brutal method of restoring proportionality: capitals that cannot valorise themselves are destroyed or sold cheap, debts are defaulted or inflated away, unemployment swells the reserve of labour and disciplines the employed, and the survivors resume accumulation from a lowered cost base. Crisis is thus both breakdown and reset, an involuntary audit of claims that had outrun the surplus available to honour them. This functional reading explains a feature of the theory that puzzles newcomers, its refusal to treat each crisis as a policy failure that better management would have prevented. Management can shift the timing and the incidence; the theory holds that it cannot repeal the audit, only decide who pays for it and in what currency, losses or inflation or public balance sheets.

Steelmanned, the framework’s exposures deserve the same candour given to Keynes’s. The theory of the falling profit rate depends on empirical magnitudes, the composition of capital and the counteracting tendencies Marx himself catalogued, and honest Marxian economists disagree about what the long record shows. The reproduction schemes demonstrate possibility conditions elegantly but are not a forecasting engine. Above all, the framework was built for an economy of nationally organised capitals and a state of limited economic reach; applying it to an era of central banks that underwrite asset markets and treasuries that can sustain deficits for decades requires the auxiliary arguments the tradition has in fact produced, about credit, fictitious capital, and the political management of devaluation. Those arguments exist and some are strong, but their necessity concedes the Keynesian point that the modern state has more room to manoeuvre than a reader of the nineteenth-century texts might expect. The objection section below gives that point its full force.

One more piece of the Marxian position must be stated precisely, because the comparison turns on it: the role of the state. Marx left no systematic theory of economic policy, and the tradition has filled the gap in divergent ways, but the structural claim is consistent across most of it. The state, however autonomous its managers, operates inside an economy whose investment decisions remain private, so its room to sustain demand is bounded by the confidence of capital, by the bond market’s terms, and by the electoral and workplace consequences of full employment itself. The guide to Marxist state theory and modern politics develops that structural picture; for the comparison at hand, the point is that the boundary Keynes treats as a technical matter of policy design appears, on the Marxian side, as a political relation that policy design cannot wish away. Kalecki, arriving later with tools from both frameworks, converted that boundary from a slogan into an argument.

The demand question inside Marx: reproduction, Luxemburg and the surplus

A stubborn misreading holds that the Marxian framework has no demand problem because its crises are born in production, and the misreading survives because Marx never assembled his scattered treatments into one chapter. The second volume of Capital, published by Engels from the manuscripts in 1885, is in large part a treatise on demand, though its vocabulary is that of reproduction. Marx divides social production into two departments, one making means of production and one making means of consumption, and asks what proportions between them permit the total product to be sold and replaced, period after period, first on the assumption of no accumulation and then with accumulation complicating every flow. The schemes demonstrate that smooth reproduction is possible, which is itself a result: they are no proof that crisis must occur. They demonstrate simultaneously that smoothness requires proportions that no mechanism enforces, since each department’s decisions are taken by competing capitals ignorant of the aggregate requirement. The demand problem inside Marx is therefore a proportionality problem in the first instance, and the schemes stand behind both the disproportionality family of crisis theories and, at a further remove, the input output methods that Leontief later built into planning and national accounting practice.

Rosa Luxemburg pushed the demand question to its sharpest Marxian form in her Accumulation of Capital of 1913, asking who, precisely, buys the portion of the product that embodies the surplus destined for accumulation. Workers cannot, beyond their wages; capitalists’ personal consumption covers only part; investment demand answers the question only by enlarging next period’s product and therefore next period’s problem. Her conclusion, that accumulation requires markets outside the capitalist relation itself, in the peasant and colonial worlds being absorbed and destroyed, was rejected by most of the tradition, and her critics, Bukharin most ferociously, showed that the schemes permit accumulation without the external buyer she demanded. The debate matters for the Keynes comparison because it locates the exact point where Marxian analysis brushes against effective demand and then turns aside. Luxemburg’s question is Keynes’s question in Marxian dress: what validates the surplus product as sold? The tradition’s orthodox answer, that investment demand validates it circularly and legitimately within the schemes, is formally correct and dynamically silent: it states the proportions that would work without explaining what holds the decisions to them. Kalecki, who began from the schemes and took the demand question as seriously as Luxemburg had, can be read as the thinker who accepted her question and rejected her answer, replacing the external market with the spending decisions of the capitalist class itself and the state at its elbow. That reading flatters neither camp’s mythology, which is some evidence for it.

The later Marxian literature on realisation adds a refinement the comparison needs. In the third volume, the conditions of production and the conditions of realisation are described as distinct and potentially divergent: surplus value is extracted under one set of constraints, governing the working day, intensity, and the technical composition of production, and realised under another, governing the distribution of income and the proportions of demand. Crisis erupts when the divergence can no longer be papered over by credit. Formulated that way, the Marxian framework contains a demand theory subordinated to a production theory, and the subordination is the whole disagreement with Keynes in miniature. Keynes’s framework contains, symmetrically, a production theory subordinated to a demand theory: supply adjusts passively to demand in the short period, capacity is given, and the conditions of production never push back within the model’s horizon. Each framework possesses the other’s territory as a province and governs it lightly. The frontier disputes of the literature, over whether profitability or demand failed first in a given episode, are border disputes between two empires each of which maps the whole continent and administers half of it.

Credit and fictitious capital: the monetary depth of the Marxian account

Any reader who knows the Marxian framework only through its production vocabulary will underestimate its monetary sophistication, and the underestimation flatters the Keynesian side of the comparison by default. The third volume of Capital contains an extended analysis of the credit system, interest bearing capital, and what Marx called fictitious capital: titles to future revenue, shares and bonds and government debt, which circulate as wealth while representing claims on surplus value that has not yet been produced and may never be. In the boom, these titles multiply on the strength of expected earnings, collateral validates credit and credit validates collateral, and the system’s real accumulation proceeds inside a hall of financial mirrors that Marx described with a reporter’s eye for the swindling that accompanied the railway promotions of his century. In the crisis, the mirrors are audited. Everyone demands money, the one asset that is not a claim on the future but settlement in the present, and capitals that were solvent on the boom’s valuations are insolvent on the crisis’s. The structural kinship with Keynes’s liquidity preference and with Minsky’s later balance sheet dynamics is not a resemblance imposed by sympathetic readers; it is a shared recognition that a capitalist economy denominates its promises in an asset it can refuse to honour collectively, and that the refusal, when it generalises, is the crisis’s most contagious form.

The difference, once again, is ordering. Marx’s fictitious capital expands because real accumulation, pressing against the limits of profitable production, seeks outlets in speculation and in the capitalisation of every revenue stream in sight; the financial froth is a symptom of the production economy’s difficulty in placing its surplus. Keynes’s financial instability arises within the sphere of valuation itself, as the conventions governing long term expectation prove flimsier than the commitments built on them, with no necessary prior deterioration in production. The empirical literature on modern crises finds material for both orderings, often in sequence: a real economy generating more claims than its surplus can service, and a valuation regime that discovers the fact all at once. What neither framework permits is the doctrine, standard in the textbooks both opposed, that finance is a veil draped over real magnitudes and that monetary disorder is always somebody’s policy error. On money, as on Say’s law, the supposed opposites are allies against the centre, and their quarrel is over which end of a single mechanism deserves causal primacy.

Two episodes under both lights: the nineteenth-century crash and the great depression

Frameworks earn their keep on episodes, and two canonical episodes show the division of labour between these two with unusual clarity. The crisis of 1857, the first to deserve the name world crisis, began with American railroad finance and the failure of a New York trust company, crossed the Atlantic through bills of exchange, broke banks in Hamburg and Glasgow, and idled mills in Lancashire within a season. Marx, who reported it for the New York Daily Tribune and filled notebooks in anticipation of the general reckoning he expected, read it as the system auditing its own overextension: overbuilt railways, overissued paper, overstocked warehouses, the whole apparatus of claims repriced against the surplus actually available. The Keynesian instruments, applied retrospectively, illuminate the same episode from the spending side: a collapse of investment in the leading sector, a liquidity panic in which the demand for gold and Bank of England notes became insatiable, and a multiplier working through the trading economies in reverse. Both readings organise real evidence; neither is reducible to the other; and the Marxian reading has the edge on the question why the claims had outrun the surplus in the first place, while the Keynesian reading has the edge on the mechanics of the contagion once the repricing began.

The great depression inverts the emphasis. The American economy of 1929 was not obviously overbuilt in the Marxian sense; its catastrophe was the depth and duration of the collapse that followed the financial break, output falling by roughly a third, unemployment reaching roughly a quarter of the labour force, and the partial recoveries relapsing, most instructively in 1937 when premature fiscal and monetary tightening demonstrated, in something close to laboratory conditions, that the economy’s resting place was wherever policy and expectations left it. Keynes’s framework was built for this episode and explains its signature facts, the persistence, the failure of wage and price flexibility to heal, the efficacy of public spending when finally applied at war scale, with an economy of means no rival matched. The Marxian literature on the depression, from the Comintern’s crude third period pronouncements to the serious later work on profitability and monopoly structure, has always had to work harder, and its best contributions concern the preconditions, the distribution of income and the fragility of the credit structure built in the twenties, rather than the collapse’s arithmetic. A fair scoring of the two episodes gives each framework one home fixture, which is exactly the result the axis by axis discipline predicts, and exactly why the search for a synthesis, rather than a champion, has attracted the strongest minds who worked the territory.

The framework comparison

With both frameworks at full strength, the comparison can be stated compactly. The framework comparison below sets the two positions against each other on the dimensions where they genuinely diverge, and its final row adds the position that combines them, because the history of the subject refuses to stay a duel. Cells are compressed by necessity; the prose around the table carries the qualifications, and no reader should score the contest from the table alone.

Framework	Source of crisis	Role of expectations	Treatment of class	What policy can achieve	Time horizon	Prediction for sustained full employment
Marx	Production of surplus value under competitive accumulation	Secondary; confidence follows profitability	Constitutive; classes defined by relation to production	Postponement and displacement of crisis, not removal	Long period; the system’s whole motion	Unstable; full employment strains discipline and profitability until accumulation or politics breaks it
Keynes	Deficient effective demand driven by volatile investment	Central; investment rests on convention and confidence	Taken as given; distribution outside the model	Restoration and maintenance of high employment	Short period; given equipment and institutions	Achievable and durable if investment is socially guided
The Kalecki bridge	Effective demand operating inside a class-divided economy	Important, but anchored in class income shares	Constitutive for distribution, Keynesian for output	Economically sufficient, politically self-undermining	The political cycle within the long period	Technically possible, politically resisted; cycles between adoption and abandonment

Read the columns slowly, because several apparent disagreements dissolve under pressure while a few harden. Expectations, for instance, are not absent from the Marxian picture: the credit system, which Marx analysed with a care that surprises readers who meet him only through summaries, runs on confidence, and its collapse converts latent disproportions into open crisis. The difference is one of causal ordering. In the Keynesian sequence, a shift of expectation is a prime mover, capable of starting a slump in an economy whose productive arrangements were sound. In the Marxian sequence, euphoria and panic are the forms in which underlying movements of profitability announce themselves, real enough in their effects but derivative in their origin. An examiner asking which framework assigns expectations the leading role has a clean answer; an examiner asking whether either framework ignores them has the opposite one.

Class shows the reverse pattern: present in both, constitutive in only one. Keynes wrote vividly about rentiers, entrepreneurs, and the working class, and his policy proposals had distributional edges he understood perfectly well. But the classes enter his model as characters, not as the mechanism. Remove them from the General Theory and the machinery still runs, driven by propensities and expectations of representative agents. Remove them from Capital and nothing remains at all, because the theory’s engine is a relation between classes: the purchase of labour power by the owners of money. That asymmetry explains why the two frameworks can be combined only by doing work, not by stapling: any synthesis must decide whether distribution is an input or an output of the analysis, and Kalecki’s originality lies in making it both, income shares determining demand while demand determines the surplus that shares divide.

The time horizon row deserves emphasis because it quietly governs many scoring disputes. Keynes announced a theory of the short period and defended the restriction as method: hold the stock of equipment and the institutional frame fixed, and ask what determines output within them. His critics’ charge that the theory cannot explain the frame itself is true and, from inside the method, beside the point. Marx’s theory is nothing but an account of the frame in motion, of how equipment, institutions, and class capacities are themselves produced and periodically wrecked by accumulation. To ask which is the better theory of a given recession is therefore sometimes to ask a well-posed question about mechanisms within the period, and sometimes to commit a category error, grading a theory of motion by the standards of a theory of rest. The strongest literature on both sides respects the distinction; the weakest, which is also the loudest, does not.

Axis by axis: money, saving, the wage and the state

The locus of crisis has been covered; four further axes complete the comparison, and each will matter when the objection and the bridge arrive.

Money first. For Keynes, money is the asset whose peculiar properties, negligible carrying cost and a liquidity premium that can outrun the prospective yields of real assets, allow it to set the terms on which investment occurs. The rate of interest is a monetary phenomenon, and a monetary economy can therefore refuse to employ its resources for monetary reasons: a flight into liquidity is a refusal to set labour in motion. For Marx, money is the necessary form of value, the medium in which the metamorphosis of commodities completes or fails, and the credit system built on it stretches that metamorphosis until the underlying claims on surplus value outrun any plausible realisation, at which point the demand for money as the only final means of payment returns with violence. The vocabularies differ, the phenomenon recognised is recognisably kin: in both frameworks, crisis wears a monetary face, panics are about liquidity, and the doctrine that money is a neutral token is rejected root and branch. Scholars who stress the kinship, among them theorists of endogenous money in the post-Keynesian tradition, have built a genuine research programme on the overlap; scholars who stress the difference reply that for Marx the monetary panic is the surface form of a production disorder, while for Keynes it can be the disorder itself. Both readings are documented positions in the literature, and the applied case of a modern financial collapse, examined in the series study of Marxism and the financial crash, shows each mechanism at work in the same episode at different depths.

Saving and investment divide the frameworks more subtly. Keynes’s revolution was to make saving a passive residual: investment, decided by entrepreneurs under uncertainty, determines income, and saving adjusts to investment through the change in income rather than through any price. Thrift campaigns in a slump are therefore arson dressed as virtue. The Marxian tradition reaches a structurally similar conclusion by a different path: the capitalist’s drive is not to consume but to accumulate, hoarding in Marx’s analysis is a phase of the circuit rather than a leakage from it, and the problem is never a shortage of abstinence but the conditions under which accumulated money can be profitably recommitted. Where Keynes sees investment blocked by expectation and the interest rate, Marx sees it blocked by the prospective rate of profit on new capital relative to the mass of claims already outstanding. The difference surfaces in policy: a Keynesian state attacks the blockage by lowering the cost and raising the expected return of spending, its own included; a Marxian analysis expects the blockage to migrate, since capital that will not invest at the going rate of profit can also refuse, strike, or exit, converting an economic policy into a political confrontation.

The wage axis has been touched already and needs only its comparative edge sharpened. In Keynes, the money wage is best kept stable, real wages move countercyclically as prices adjust, and the level of employment is set by demand rather than by bargaining over the wage. In Marx, the wage is the value of labour power, tugged downward by the reserve army of unemployed that accumulation itself replenishes and upward by workers’ organisation and by the prosperity that tightens labour markets. The two accounts agree that general wage cutting is no road to recovery, a conclusion each reaches from its own engine room, Keynes from the demand side and Marx from the observation that wages are advanced from capital and their reduction contracts the market for the product. They disagree about what prosperity does to the wage relation over time: Keynes treats a high-employment economy as one in which distributional conflict can be managed and gradually softened, while the Marxian tradition, and preeminently Kalecki, treats full employment as a condition that transforms the balance of power at the point of production and therefore generates its own opposition.

The state axis follows from the wage axis. Keynes’s state stands outside the machine it regulates, armed with the budget and the central bank, constrained mainly by the competence and nerve of its operators. The Marxian state stands inside the machine, its fiscal capacity resting on the surplus it must not frighten, its officials recruited from and answerable to the society whose class structure the economy continually reproduces. Neither picture is a cartoon: Keynes knew that treasuries face parliaments and markets, and Marxian writers from the state derivation debates onward have analysed the relative autonomy of governments with sophistication. The axis is a matter of default assumptions. When a Keynesian policy fails, the Keynesian literature asks first whether it was tried at sufficient scale and duration. When it fails, the Marxian literature asks first whose interests its continuation would have injured and what those interests did about it. The 1943 paper at the centre of the next section is the classic statement of the second question, written by an economist who accepted the first literature’s economics entire.

Recovery and its shape: what each framework expects after the fall

Crisis theories are usually graded on their account of the fall, but their accounts of the recovery separate them just as cleanly, and the separation is testable against the reader’s own memory of slumps. The Keynesian recovery is a restoration of spending: expectations revive or are revived, the multiplier reverses its work, and the economy climbs back toward the capacity it never lost, its workers rehired into the same industries with the same equipment. Nothing in the machinery requires the slump to have accomplished anything; it was waste from first to last, and the recovery’s speed is a verdict on policy rather than on any purgative process the economy needed to undergo. The Marxian recovery is a reconstruction: the slump has destroyed values, bankrupted the weak, cheapened assets and machinery, swollen the reserve of labour until wages and workplace resistance are beaten down to levels at which accumulation becomes profitable again, and the upturn begins on a restructured base, with ownership more concentrated and the survivors’ capital written down to values the available surplus can validate. On this account the slump did accomplish something, at a cost its victims paid, and a recovery that arrives without the restructuring, engineered by spending alone, carries the unliquidated claims forward into the next expansion.

The contrast generates divergent readings of a phenomenon both frameworks must explain: the jobless or credit laden recovery, in which output and profits revive while employment and household balance sheets lag. The Keynesian reading reaches for the composition of spending and the caution of firms burned by the slump, treating the lag as a demand shortfall in particular channels, remediable by better aimed policy. The Marxian reading sees the recovery’s shape as the point: accumulation resuming on the basis of a disciplined and cheapened labour force, with the surplus extracted per worker raised precisely by the insecurity the slump installed, so that the lag in employment is not a defect of the recovery but one of its mechanisms. Kalecki’s political cycle binds the two readings into a sequence: the recovery engineered by spending strengthens labour as it proceeds, until the strengthening itself becomes the political problem that ends the policy regime, at which point the reconstruction logic reasserts itself through a politically administered dose of unemployment. A reader equipped with all three accounts can watch an actual recovery and ask, in order, whether spending is returning, whether the base has been restructured, and whether the political weather around full employment is shifting, and will find that the three questions rarely have the same answer at the same time.

The open economy axis: mercantilism, Bretton Woods and the world market

Neither framework was built for a closed economy alone, and the international axis exposes a further instructive asymmetry. Keynes, in the General Theory’s most heterodox chapter, argued that the mercantilists had grasped something the classical doctrine missed: under a fixed monetary standard, a trade surplus was a rational national strategy for an economy short of investment outlets, since it imported demand and exported unemployment, and the scramble for surpluses was therefore a competition the classical harmony story could not see. His wartime design work at Bretton Woods attempted to civilise that scramble, with an international clearing union whose rules would pressure surplus countries to expand rather than allowing deficit countries to bear the whole adjustment. The Marxian tradition begins at the other end: the world market is not an extension of the national economy but its presupposition and its destiny, the arena in which capitals of different nations and different organic compositions confront each other, values are transferred through unequal exchanges of every contested kind, and crises synchronise across borders through trade and credit. Marx planned a book on the world market that he never wrote, and the tradition has been completing it ever since, from Luxemburg’s external markets to the dependency and world systems schools to the contemporary literature on global value chains.

Set side by side, the international axis repeats the domestic pattern at larger scale. Keynes treats the international order as a design problem: build institutions that penalise hoarding and surplus accumulation, and the demand failures of the world economy can be managed as the national ones are. The Marxian tradition treats the international order as a hierarchy problem: the institutions themselves register the power of the capitals and states that dominate them, surplus countries will not submit to discipline they can veto, and the adjustment will be imposed on the weak in the currency of unemployment and austerity, whatever the charter says. The postwar monetary order’s history, in which the clearing union was rejected in favour of a fund the strongest creditor could dominate, and in which adjustment has indeed fallen overwhelmingly on deficit countries, is the episode both frameworks cite, Keynesians mourning the road not taken and Marxians doubting the road existed. The reader will recognise the shape of the disagreement by now: a technical design and a structural veto, with Kalecki’s theorem, that the veto operates through the domestic politics of full employment in each country separately, supplying the mechanism the structural claim needs.

The missing variable: the rate of profit and its absence from Keynes

The FAQ section answers briefly why Keynes built no part of his theory on the rate of profit; the question deserves a fuller hearing in the body, because it is the point where the comparison stops being a matter of emphasis and becomes a matter of architecture. In Marx, the rate of profit is the system’s master gauge. It condenses the whole production relation, the rate of exploitation in the numerator and the accumulated weight of past investment in the denominator, into the single magnitude on which the decision to accumulate depends, and its tendential fall, driven by the rising mass of machinery per worker and checked by counteracting influences Marx enumerated himself, is the theory’s principal long run dynamic. Every other mechanism in the crisis theory can be stated as an influence on that gauge or a consequence of its reading. In Keynes, the gauge does not exist. There is a schedule of expected returns on new assets, formed asset by asset under uncertainty, and there is the interest rate, and investment proceeds while the first exceeds the second. The existing stock of capital yields what it yields, undistributed into any aggregate rate whose movement might govern the system’s behaviour. A Martian reading only the General Theory would learn that capitalism suffers from deficient demand and unstable expectations, and would never suspect that its theorists of the rival school regarded the profitability of the accumulated stock as the mainspring of its history.

Keynes’s omission was principled, and its principle deserves the steelmanning this article has practised throughout. He believed the aggregate capital stock was not a measurable thing independent of the rate of return used to value it, an intuition the Cambridge controversies later weaponised against neoclassical production functions, and he believed that a theory of employment for the short period should take the stock as it found it. There was also a strategic silence. To theorise the rate of profit is to theorise its source, and to theorise its source, in any vocabulary available in 1936, was to enter the territory of exploitation, where Keynes did not wish to go and where his policy audience would not have followed. His framework’s celebrated practicality, its ability to be operated by a Treasury official with no metaphysical commitments, is purchased with that silence. The Marxian counterclaim writes itself: a variable excluded for reasons of tact returns as the theory’s blind spot, visible whenever a demand expansion runs into a profitability wall, whenever investment refuses cheap money because the expected return on real projects, not the cost of funds, is the binding constraint, and whenever a government discovers that the confidence it must court is confidence in profits specifically. The postwar profit squeeze episode, examined in the objection’s aftermath above, is the standing exhibit. Scoring the axis, then: the Marxian framework possesses a theory of the magnitude that governs accumulation and pays for it with a century of empirical dispute about the magnitude’s measurement; the Keynesian framework declines the magnitude and pays for the refusal with a recurring explanatory gap at exactly the points where slumps are prepared. Neither payment is small, and a reader who has grasped both is in possession of the comparison’s most technical and most consequential result.

Uncertainty against contradiction: the deepest methodological divide

Beneath the axes lies a divide about what kind of thing a crisis is, and surfacing it clarifies why the debate has lasted a century without exhausting itself. For Keynes, the root disorder is epistemic. The future cannot be known, the conventions that substitute for knowledge are fragile, and an economy that must commit real resources on the basis of unknowable returns will periodically lose its nerve. Crisis, at the deepest level of his system, is what irreducible uncertainty does to investment when the animal spirits fail. For Marx, the root disorder is relational. Capital must expand or die, labour power is bought as a commodity and set to produce more value than it costs, and the system’s own success criteria, cheapen labour, expand output, outrun the rival, generate the disproportions and profitability pressures that crisis periodically resolves. Crisis, at the deepest level of his system, is what the capital relation does to itself when it works as designed. Uncertainty against contradiction: the phrases are slogans, but they mark a real fork. An epistemic disorder invites epistemic and managerial remedies, better conventions, steadier expectations, a state that lends its own nerve to the market. A relational disorder invites the conclusion that management, however skilful, administers a contradiction it cannot remove, and the political history of the twentieth century can be read, without much forcing, as the two invitations being accepted in turn by different societies at different times.

The methodological divide also explains the frameworks’ different relations to evidence. Keynesian claims are comparatively easy to test at the level of episodes, since the variables, spending, interest rates, expectations as proxied by surveys and asset prices, move in observable time, and the school’s empirical tradition, from the great postwar macroeconometric models to the narrative identification of policy shocks, is correspondingly rich. Marxian claims operate at longer wavelengths and through magnitudes, the rate of profit above all, whose measurement requires contestable decisions about valuation, depreciation, and the boundary between productive and unproductive labour, so that the school’s empirical tradition is a permanent argument about its own data. Partisans convert these asymmetries into verdicts: one side calls the other unfalsifiable, the other calls the first superficial. The fairer judgement is that the frameworks have different exposure profiles, Keynesian theory risking death by a thousand qualifications as each mechanism is refined, Marxian theory risking death by indefinite postponement as each test is deferred to better data. The Kalecki bridge, once more, sits instructively between them: its profit equation is an accounting identity with causal direction imposed, testable in national accounts within a year of any episode, while its political theorem operates at the long wavelength where Marxian claims live. A synthesis that inherits both exposure profiles is not thereby twice as vulnerable; it is, on the record of its use, the working equipment of the economists, in both camps’ penumbras, who have explained actual crises best.

The Kalecki bridge

Michal Kalecki was born in 1899 in Lodz, trained as an engineer, drifted into economic journalism and then into theory without a chair or a school, and arrived at the principle of effective demand by a road that began in Marx’s reproduction schemes. The biographical sequence matters because it certifies the independence of the derivation. He was not a Keynesian who added class for flavour, nor a Marxist who borrowed demand as an afterthought. Working in Warsaw in the early nineteen thirties, publishing in Polish and then in French, he constructed a theory of the business cycle in which output and employment are determined by capitalists’ spending decisions, while the division of the product between wages and profits is determined by the degree of monopoly, the pricing power of firms. Those two determinations, demand from the spending side and distribution from the power side, are the two halves of the bridge.

What did Kalecki publish before the General Theory appeared?

Working in Warsaw in the early nineteen thirties, Kalecki published studies in Polish, and in a French journal, that built a theory of the business cycle on Marx’s reproduction schemes and on the division of income between classes, reaching the principle of effective demand several years before Keynes’s book appeared in print.

The derivation of profit is the bridge’s load-bearing arch, and students should master its logic in words. Divide society into capitalists and workers. Workers, Kalecki assumed, spend what they receive; saving out of wages is negligible at the level of the whole economy. Capitalists decide how much to invest and how much to consume, and their profits turn out to equal exactly that investment plus that consumption, together with any budget deficit and export surplus in an open economy with a state. The aphorism the literature attaches to the result runs that workers spend what they earn while capitalists earn what they spend. The causality is the revolutionary part. Profits are not a reward that precedes spending and finances it; they are a consequence of spending decisions already taken. A capitalist class that invests boldly creates its own profits in the aggregate, however anxious each member feels individually, and a class that retrenches finds its profits shrinking however virtuous its intentions. Keynes reached the same inversion through the widow’s cruse image in his earlier Treatise on Money and placed investment at the centre of the General Theory; Kalecki reached it from the reproduction schemes, where the departments producing means of production and means of consumption must validate each other’s output, and where the question of who buys the surplus product receives an exact answer.

Joan Robinson, who knew both bodies of work as well as anyone alive, championed Kalecki’s priority in print and treated the independent derivation as a standing reproach to a profession that had ignored publications in Polish. Her judgement carries weight because she had every partisan reason to credit Keynes, whose circle she belonged to, and credited Kalecki anyway. The episode also instructs methodologically. Two derivations of one result, from different starting points, in different languages, by men working in mutual ignorance, is the kind of convergence that in other sciences counts as strong evidence for the result itself. The economics profession registered the convergence slowly and imperfectly, and the failure is part of why the bridge figure remains half known: his collected works appeared in full English dress only decades after his death in 1970, and the short paper that completes the bridge is still more cited than read.

That paper, published in the Political Quarterly in 1943 under the title Political Aspects of Full Employment, asks the question neither framework had posed squarely. Suppose the Keynesian technique works. Suppose governments learn to maintain full employment through spending, so that slumps become optional. Will the regime last? Kalecki’s answer was no, and his reasoning had nothing to do with economic technique. Business leaders, he argued, value influence and social position alongside profit, and a permanent regime of full employment dissolves the disciplinary foundations of the workplace. Dismissal loses its terror when another job waits across the street. Workers grow, in his deliberately flat wording, out of hand; demands extend from wages toward the organisation of production itself; and the captains of industry, who dislike state intervention in employment creation even when it raises their sales, discover a principled objection to deficits precisely when deficits have ceased to be necessary evils and become the permanent machinery of their own disempowerment.

What happens to the bargain between classes at full employment?

Sustained full employment strengthens the bargaining position of workers, since dismissal loses its sting when another job is easy to find, and it encourages demands that reach beyond wages toward control of the workplace. Kalecki treated that shift in power, not any technical impossibility, as the core political obstacle.

From this he derived the political business cycle: full employment would be engineered before elections and abandoned under the banner of sound finance when the social consequences ripened, so the economy would oscillate not because the technique failed but because it succeeded. The argument’s structure is what earns it the name this article gives the whole construction. Its economics is Keynesian without remainder: demand determines output, deficits employ the idle, no technical barrier stands in the way. Its politics is Marxian without remainder: the constraint on policy is the class relation itself, the dependence of workplace authority on a reserve of unemployed labour, and the organised capacity of capital to translate its social discomfort into doctrine, pressure, and electoral weather. The claim that names the article’s organising idea, the Kalecki bridge, is exactly this: the strongest position in the Marx and Keynes comparison is the synthesis in which Keynesian remedies work economically and falter politically for reasons that Marxian class analysis specifies. Neither parent framework states the synthesis. Keynes assumes a state that, once instructed, will act; the Marxian tradition asserts that it will not, but often asserts it as fate rather than deriving it from the workplace outward. Kalecki derived it.

The bridge also carries traffic in the other direction. Steindl, Kalecki’s collaborator and intellectual heir, developed the stagnation analysis of mature oligopolistic economies that fed directly into the Monthly Review tradition, whose account of monopoly capital and its chronic surplus absorption problem is the Marxian school that stands closest to Keynesian concerns, as its own profile in this series on monopoly capital and stagnation theory makes plain. And the political business cycle, once named, became a permanent instrument of analysis across the spectrum, used by writers who would reject every other element of Kalecki’s inheritance. A bridge, to deserve the metaphor, must be walked on by people travelling to different destinations; by that test this one qualifies.

Fairness requires registering the points where the bridge creaks. Kalecki’s assumption that workers save nothing is a simplification he adopted knowingly, defensible for the economies he studied and less innocent in an age of pension funds and mortgage debt, though his followers have generalised the model without breaking its spine. His degree of monopoly determines distribution with a directness that bargaining institutions, minimum wages, and public employment complicate in practice. And the political argument, powerful as a tendency, does not specify its own timing: regimes of high employment have persisted for decades in some countries, a fact the objection section now confronts head on, because it is the strongest evidence on the other side.

The bridge inspected: distribution, the cycle and the reception

A bridge metaphor obliges engineering detail, and Kalecki’s construction has three further members that the compressed account above deferred. The first is his theory of distribution. In Kalecki’s models, firms in imperfectly competitive markets set prices as a mark up over prime costs, and the average mark up, his degree of monopoly, fixes the share of wages in national income before any bargaining over money wages begins. Money wage increases, under this construction, are largely passed into prices, leaving the real distribution governed by pricing power, industrial structure, and the relative strength of unions in resisting the pass through. The elegance for the comparison is that distribution enters as a structural parameter with class written through it, exactly where Keynes’s framework has a blank: Keynes accepted the classical doctrine that real wages equal the marginal product of labour as a first approximation, a concession his followers at Cambridge spent decades regretting on his behalf. Kalecki never needed the concession retracted, because he had never made it. His economy is born divided, and its division does macroeconomic work from the first equation.

The second member is the business cycle theory proper, the work of the early nineteen thirties from which the effective demand result originally grew. Kalecki modelled investment as decided in one period, delivered after a gestation lag, and productive of capacity that then competes with the demand its own construction generated, so that the very success of an investment boom prepares the downturn: orders crest, capacity arrives, profits disappoint, orders collapse. The mechanism has a Marxian skeleton, since it turns on the dual character of investment as demand now and supply later, a duality Marx’s schemes encode in the two departments and that Keynes’s short period method suppresses by holding the capital stock fixed. It also has a prophetic quality that the later literature noticed: an economy can talk itself into trouble in Keynes’s telling, but in Kalecki’s it builds its trouble, in plant and equipment, during the good years. Empirical business cycle research in the Kaleckian tradition, and the financial elaborations Minsky added from the Keynes side, converge on that built character of fragility from opposite starting points, another of the convergences that make the bridge load bearing rather than decorative.

The third member is the reception, which is itself evidence in the comparison. Kalecki spent the late nineteen thirties in Cambridge, where his priority was acknowledged by Robinson and quietly resented elsewhere, then worked for the Oxford Institute of Statistics during the war, contributing to the practical economics of rationing and war finance, and returned to Poland after it, to advise, to be marginalised in the Stalinist years, to return to prominence after 1956, and to spend his last decade building a school in Warsaw while his international reputation climbed. Keynes, by contrast, was installed at the centre of two treasuries and a new international order before his death in 1946. The asymmetry of the afterlives is usually told as a story about language and personality, and it is partly that. But it is also a story about what each framework offered its decade: Keynes arrived with a policy for governments that existed, Kalecki with an analysis that told those governments why they would not keep using it. Professions reward the first gift more reliably than the second, and a reader of the comparison should factor the reward structure into any judgement about why one name fills the textbooks and the other requires an article like this one to be introduced.

One caution completes the inspection. The bridge is a construction of the literature as much as of Kalecki himself: he did not write a treatise reconciling Marx and Keynes, and his own polemics could be as dismissive of Keynesian vulgarisations as any Marxist’s. The claim that his work forms a bridge is a claim about the logical geography of his results, that his demand equations are Keynesian in function while his distribution theory and his political analysis are Marxian in content, and that a reader standing on his work can see both banks clearly. Claims about logical geography are the kind that comparisons live on, and this one has survived half a century of specialists trying to annex him wholly to either shore, which neither annexation has managed without amputating half the corpus.

The strongest objection: the postwar generation

No comparison in this series is complete until the hardest case against its organising claim has been stated at full strength, and here the case is formidable. It runs as follows. For roughly a generation after the Second World War, the advanced capitalist economies did the thing the Marxian framework says cannot durably be done. Under private ownership of the means of production, with markets and wage labour intact, governments committed to high employment maintained it: unemployment in the leading economies ran at levels their prewar history would have called impossible, real wages rose year after year, welfare states were built, and the profit system not only survived the arrangement but expanded through it. Demand management, the Keynesian technique, was the avowed instrument. If a theory predicts that policy can only postpone crisis, a quarter century of postponed crisis with rising living standards looks less like postponement and more like refutation. Moreover, when the regime was eventually abandoned in the nineteen seventies and eighties, the abandonment was a political choice, contested at elections and in cabinets, not a mechanical necessity: governments chose disinflation and accepted mass unemployment as its price. Choice, the objection concludes, is exactly what structural theories say is not on offer, and the postwar decades therefore stand as the Keynesian framework’s best evidence and the Marxian framework’s hardest fact.

The bridge position answers the objection by absorbing its second half and disputing its first. The abandonment, on Kalecki’s argument, was not an uncaused choice dropped from the sky of ideas; it was the predictable political harvest of the regime itself. Sustained full employment had, by the nineteen seventies, shifted bargaining power in the workplace, nourished wage militancy and a widening of demands from pay toward control, and squeezed profits between organised labour and intensifying international competition. The political coalition for sound finance assembled exactly the interests Kalecki identified, speaking exactly his predicted language of discipline and confidence, and the stagflation that discredited the old regime arrived through the distributional conflict the regime had incubated. On this reading the postwar generation confirms the bridge rather than refuting it: the technique worked for as long as its political preconditions held, and it generated the forces that withdrew them. The Swedish wage earner funds proposal of the nineteen seventies, in which organised labour sought to convert wage restraint into collective ownership stakes, reads in hindsight like the moment the logic surfaced into open politics: the boundary between managing demand and changing ownership, which Keynes had treated as the limit of his inquiry, presented itself as an agenda, and was defeated as one.

The concession must be stated with equal clarity, because the bridge’s defenders in print, from Rowthorn to the post-Keynesian literature, state it themselves. The reply concedes the economic point. It grants that demand management can sustain high employment and rising wages under private ownership for a generation, which is to say it grants that the strong Marxian claim, that policy is impotent and crisis unpostponable, is false as stated. What survives is a weaker and more interesting claim about politics: that the sustainability of the regime is decided not by technique but by the shifting balance of class power that the regime itself alters. A reader scoring the contest should register the relocation honestly. The disagreement between the frameworks, pursued to its end, migrates out of economics proper and into political science, where the Marxian tradition’s concepts of class power and the state’s structural dependence on private investment do the work that its crisis mechanics cannot. Whether that migration counts as a vindication or a retreat depends on what one thought the original dispute was about, and the literature contains both judgements, attributed and argued, rather than a settled score.

Two political philosophies: preservation against transformation

The frameworks carry political philosophies that their technical presentations half conceal, and an honest comparison brings them into the open without endorsing either. Keynes’s philosophy is a liberalism that has looked into the abyss: he held that the economic problem of scarcity was being solved by accumulation itself, that within a century or so the advanced world could afford to demote the profit motive from its pedestal, and that the interim task was to manage the transition humanely, by socialising investment, taxing the rentier into gradual extinction, and keeping employment high enough that the politics of desperation never took hold. His attitude to the working class was paternal and genuinely fearful in equal measure: he wanted its material grievances dissolved before they radicalised, and his economics is the engineering drawing of that dissolution. The Marxian philosophy inverts the valuation of the same facts. What Keynes calls the transition’s management, it calls the deferral of the only change that addresses the cause; what he calls dissolution of grievance, it calls the purchase of consent with a portion of the surplus, revocable when the surplus tightens; and where he sees the profit motive being gently retired by satiation, it sees a relation that reproduces itself through every satiation because accumulation, not consumption, is its object. Marx’s own horizon, sketched in the critique of the Gotha programme and the third volume’s remarks on the realm of freedom, lies beyond the wage relation altogether, in a society where the reduction of necessary labour time, not its profitable extension, becomes the measure of wealth.

A comparison that concealed this divergence behind technique would be a fraud on the reader, but so would a comparison that treated the divergence as a substitute for the technical argument. The philosophies explain the theories’ shapes: Keynes’s bracketing of distribution is the technical shadow of his belief that distributional conflict could be administratively dissolved, and Marx’s insistence on beginning in production is the technical shadow of his belief that nothing else was fundamental. They also explain the theories’ tempers. Keynesian writing at its best has the confidence of a reformer who expects to be obeyed by events; Marxian writing at its best has the patience of a tradition that expects vindication at the wavelength of decades and has arranged its concepts accordingly. When the two tempers meet in one economist, as they did in Kalecki, who combined a policy maker’s concreteness about the annual budget with a structuralist’s certainty about the decade’s politics, the result is the body of work this article has placed at the comparison’s centre. The reader should nonetheless resist converting temper into evidence. Confidence is not correctness, and neither is patience; the axes, the episodes, and the profit equation are where the scoring belongs.

Inflation: the meeting point neither framework wanted

Inflation is the phenomenon on which the comparison’s later history turns, and it deserves a section because the nineteen seventies are unintelligible without it. Keynes had distinguished, in his wartime pamphlet on paying for the war, between the demand deficiency of slump and the inflationary gap of an economy pressed beyond its capacity, and his framework handles inflation as the mirror image of the slump: spending outrunning real resources, to be closed by taxation, rationing, or deferred pay. What the framework handled awkwardly was inflation with stagnation, prices rising through an economy with idle plant and mass unemployment, because its basic geometry assigned those conditions to opposite regimes. The Marxian tradition had the opposite profile: distribution conflict was native to it, and the profit squeeze and monetary accommodation accounts it produced for the seventies, in which organised labour and organised capital each defended its share and the state validated both claims by expanding credit, read the stagflation as a distributional war financed by the printing press. Kalecki’s framework, once again, was waiting at the crossing: his mark up pricing made inflation a distributional variable from the start, cost increases passed through pricing power, wage increases chasing prices in a spiral whose terms were set by bargaining strength, and his followers produced conflict theories of inflation that required no Phillips curve and no mystery. The episode redrew the comparison’s map in the profession at large, since the Keynesianism that fell in the seventies was the domesticated synthesis version, and its fall carried the undomesticated Keynes down with it in the public mind, while the Marxian accounts gained a hearing their profitability data could not have earned alone. Historians of the period divide on the weights; the structural point stands that inflation is the axis on which the demand framework is weakest and the class framework strongest, and any complete scoring must enter it.

Debt and the balance sheets neither founder centred

One territory belongs fully to neither founder and has become central since their deaths: private debt as an macroeconomic engine in its own right. Keynes wrote brilliantly about debt deflation’s dangers through Fisher’s lens and about the rentier’s claims, but household and corporate leverage as the cycle’s driver belongs to Minsky’s development of his system. Marx analysed credit as the accelerator of accumulation and the form in which its contradictions mature, and his remarks on the joint stock company and the credit system remain startlingly alive, but the indebted consumer as a pillar of demand belongs to a capitalism of mortgages and credit cards that neither man inhabited. The postwar decades quietly rebuilt the demand problem on this new foundation: as wage shares stagnated in the leading economies from the nineteen eighties onward, consumption was sustained by household borrowing against appreciating houses, so that effective demand was, for a generation, privatised and collateralised rather than publicly underwritten. The arrangement has been described, in a phrase associated with Raghuram Rajan’s post mortem on the financial crisis, as letting credit substitute for the incomes policy could not raise, and it collapsed in 2008 into the largest demand failure since the nineteen thirties, at which point both frameworks were exhumed by a public that had filed them as antiques.

The episode scores the comparison in a way neither camp scripted. The Keynesian machinery described the collapse and the required response with embarrassing precision, down to the liquidity trap conditions in which the world’s central banks found themselves, and the multiplier arithmetic of the austerity years afterwards supplied the framework with some of its cleanest empirical confirmations. The Marxian and Kaleckian machinery described the preconditions with equal precision: a distributional shift that hollowed out wage based demand, a surplus in search of outlets that found them in lending to the households whose incomes had stagnated, and a political order that preferred the credit solution to either the Keynesian public one or the distributional one, because credit, unlike wages or deficits, arrived without any visible shift in workplace power. Kalecki’s ghost presides over that preference. The debt economy can be read as the political business cycle’s longest evasion, the discovery that full employment’s demand effects could be simulated by leverage while its power effects were postponed, until the leverage itself detonated and the postponed politics returned in the crisis’s aftermath. Frameworks should be judged partly by the quality of the questions they leave behind, and the question this section’s material leaves, whose balance sheet carries the system’s demand when wages will not and the state dare not, is the form in which the Marx and Keynes comparison now presents itself to any honest student of the subject.

The political cycle in the record, and the modern descendants

Kalecki’s predicted cycle found its documentary illustrations quickly, though rarely in the pure form he sketched. British economic management in the nineteen fifties and sixties acquired the nickname stop go from contemporaries: expansions encouraged until the balance of payments or the exchange rate protested, then deliberate restriction, unemployment accepted as the instrument of correction, then expansion again as elections approached. The drivers were external constraints as much as workplace discipline, and honest application of the 1943 argument must register the difference. Yet the family resemblance is not superficial. In each stop, the official language ran through confidence, discipline, and the necessity of teaching wage bargainers the facts of life, and in each go phase the same governments rediscovered that unemployed resources were a waste no doctrine could prettify. Later formalisations of the political business cycle, beginning with Nordhaus in the nineteen seventies, modelled opportunistic electoral manipulation of the Phillips curve from premises innocent of class analysis, and arrived at oscillations Kalecki would have recognised while disowning their innocence. The convergence matters to the comparison for the same reason the 1930s convergence mattered: a result reached independently from antagonistic premises earns a credibility neither premise can confer alone.

The profit squeeze literature of the nineteen seventies supplied the objection and the bridge with their shared test case. Glyn and Sutcliffe’s study of British capitalism argued, from within the Marxian tradition, that the postwar settlement had loaded the system with commitments, full employment, welfare provision, and militant wage bargaining, that pressed the share of profits until accumulation faltered, a diagnosis that borrowed its mechanism from the class politics Kalecki had forecast and its remedy, in some versions, from nowhere at all. Kaleckians and post-Keynesians divided on whether the squeeze represented the system’s success being punished or its distributional accounts coming due, and monetarists exploited the disarray to install a rival common sense in which unemployment was natural and demand management the disease. The episode is the bridge’s hardest weather and its best advertisement at once: nothing in the sequence is intelligible as pure technique, since every actor’s move was a response to the balance of workplace and electoral power, and nothing in it is intelligible as pure class mechanics either, since the policy instruments worked exactly as the Keynesian manuals said they would, in both directions, whenever a government chose to pull them hard.

The modern descendant that most often forces the comparison back into public argument is modern monetary theory, and the FAQ traffic on the subject reflects a genuine classification puzzle. The school’s intellectual ancestry is explicitly charted by its founders: Keynes on effective demand, Lerner’s functional finance, Minsky on financial fragility, and, running underneath, Kalecki’s insistence that the constraint on employment policy is real resources and politics rather than finance. Its signature claims, that a currency issuing state cannot be forced to default on obligations in its own unit, that taxation regulates demand rather than funding spending, and that a job guarantee can anchor full employment without igniting the wage spiral, amount to an attempt to repeal by institutional design the constraint Kalecki derived from class relations: the guarantee is meant to preserve a buffer of employed labour whose discipline effect substitutes for the reserve army. Marxian critics reply, in a now substantial literature, that the class content of the buffer is the whole question, that a state guaranteeing jobs inside a private accumulation economy will meet the political business cycle at the level of the programme’s administration, its wage, and its survival across elections. Keynesians of the older school object from the other flank, doubting the inflation mechanics. The debate is the comparison continued under new management, and its existence, decades after both founders, is the best evidence that the territory the bridge spans has not stopped moving.

A second modern descendant deserves brief registration, since it shows the Keynesian framework completing a circuit back toward Marx. The secular stagnation revival of the twenty first century, in which Summers and others argued that advanced economies face a chronic excess of desired saving over investment outlets at any attainable interest rate, restates in mainstream vocabulary a proposition that the Marxian tradition had maintained in its own vocabulary for a century: that mature accumulation generates more surplus seeking placement than profitable production can absorb, with the surplus diverted into asset markets and debt. The vocabularies remain untranslated and the camps mostly ignore the echo, but the structural identity of the diagnosis has been noted by historians of thought across the spectrum. When the mainstream’s own crisis theory migrates toward the territory of overaccumulation, the comparison this article has staged ceases to be a museum piece and becomes a map of where the arguments were always going to meet.

Afterlives: post-Keynesians, Minsky and the schools between

The comparison did not freeze in 1943. Its afterlife runs through the post-Keynesian school, whose founders, Robinson and Nicholas Kaldor preeminent among them, attempted the synthesis deliberately: Keynes’s effective demand extended into the long period, distribution theorised through class saving behaviour, growth analysed as a contest between accumulation and the surplus available to fund it. Pasinetti’s models gave the Cambridge tradition its most rigorous dress, in which workers who save become owners of capital and the profit rate settles by laws that owe something to both ancestors. The Sraffian reconstruction, descending from Piero Sraffa’s 1960 book, supplied the tradition with a theory of prices independent of marginalist supply and demand, and the capital controversies of the nineteen sixties, in which the Cambridge Keynesians and their sympathisers attacked the coherence of aggregate capital in neoclassical theory, were fought with Marxian ammunition by economists who mostly declined the Marxian label. What the post-Keynesians share with the Marxian tradition, and what separates them from it, is a question the FAQ section answers in compact form; the body of the article notes only that the sharing is real, covering method, money, and the rejection of Say’s law, while the separation is equally real, covering the theory of value, the analysis of exploitation, and the attitude toward the system’s permanence.

How does Minsky carry the Keynesian argument into finance?

Minsky argued that stability itself changes financial behaviour: quiet periods encourage firms and banks to take positions that depend on refinancing, so the system migrates from robust to fragile finance without any external blow. Crisis then arrives from inside the balance sheets, a development of Keynes’s investment theory rather than a departure.

Hyman Minsky deserves his own paragraph in any honest map of the territory, because his financial instability hypothesis is the point where the Keynesian lineage grows something that reads, to Marxian eyes, like a native product. Minsky took Keynes’s theory of investment under uncertainty, joined it to the balance sheet structure of firms and banks, and showed how a period of tranquillity validates progressively fragile financial postures: units that cover their debts from income give way to units that must roll debts over, and those to units that must borrow to pay interest, until a small disappointment cascades. The mechanism is endogenous, cumulative, and indifferent to external shocks, three adjectives that describe Marx’s crisis dynamics as readily as Keynes’s. Minsky himself acknowledged Kalecki’s profit equation as the correct frame for his own work, a lineage admission that places him standing on the bridge whether or not his readers noticed. Scholars divide on the classification, some filing him as the completest Keynesian, others as an honorary member of the Marxian extended family, and the division is itself instructive: at the frontier where each tradition does its best work on financial crisis, the frontier between the traditions blurs.

How should a reader decide between the two frameworks?

Decide axis by axis rather than by allegiance. Ask which account better explains the episode under study, whether its mechanism can be observed in the record, and what each theory would have predicted beforehand. On some axes the evidence favours one side; on others the frameworks address different depths of the same event.

The remaining afterlives can be sketched briskly. The mainstream that Keynes conquered proceeded, within a generation, to domesticate him: the neoclassical synthesis translated his insights into a special case of the general equilibrium theory he had attacked, valid when wages were sticky, and monetarism then displaced even the special case in policy circles. Marxian economics, meanwhile, developed along lines the crisis literature knows well: the profit squeeze accounts of the nineteen seventies, the social structure of accumulation school, the regulation approach in France, and the long empirical argument over the tendency of the profit rate, all of them attempts to give the production side of the comparison its modern form. Attempts at a grand reunification recur at intervals, and the ones with staying power tend to be built, knowingly or not, on the bridge’s design: demand arithmetic from Keynes, distribution and political constraint from Marx, and Kalecki’s equations holding the joint. The question the whole comparison finally poses is not which economist won, since the trophy would have to be awarded by decade and by axis, but whether the reader can now see a slump with both sets of eyes and knows what each set is for.

Baran and Sweezy: the American crossing

The bridge had an American span, built by Paul Baran and Paul Sweezy in Monopoly Capital of 1966, and its design shows what the synthesis looks like when constructed from the Marxian bank outward. Their central concept, the economic surplus, the difference between what a society produces and the costs of producing it, generalises surplus value for an economy of giant corporations, and their central problem is Kalecki’s problem restated for the monopoly stage: a system that generates a rising surplus and cannot absorb it through consumption or productive investment at the required rate, and therefore wastes it, on advertising and the sales effort, on military spending, on the financial sector’s ingenuity, or stagnates. The Keynesian signature is unmistakable beneath the Marxian vocabulary: the problem is framed as absorption, a failure of demand at the level of the whole system, and the state’s spending appears as the absorber of last resort whose composition, guns rather than schools in their famous accounting, is politically determined. Sweezy had been among the first American economists to take Keynes seriously in the nineteen thirties and had written one of the earliest systematic English expositions of Marxian crisis theory in 1942, a biographical double citizenship that the book embodies on every page.

The school’s fortunes trace the comparison’s later history in miniature. Marginalised in the American profession during the Keynesian ascendancy it partly shared, it supplied the Monthly Review tradition with its analytical core, influenced the dependency writers, and was revived after 2008 by readers who found its stagnation thesis waiting for them in the used bookshops. Its critics, including Marxian ones, fault its underconsumptionist leanings and its replacement of the profit rate by the surplus as the master magnitude, faults the tradition’s own later work, including the financialisation studies of its recent editors, has worked to repair by rejoining Minsky. The repair is telling: the American bridge, like the Polish original, holds together only when both banks’ engineering is respected, demand arithmetic on one side and the production of the surplus on the other, and it sags toward whichever side its builders have neglected.

Robinson’s attempted marriage: reading Marx through Keynesian spectacles

Joan Robinson performed the comparison’s most deliberate experiment in her Essay on Marxian Economics of 1942, in which a leading Keynesian, fresh from the General Theory’s inner circle, read Capital as an economist and reported what she found with a candour neither camp has matched since. She found a theory of employment struggling to escape from a value theory she considered an unnecessary detour, praised Marx’s grasp of effective demand avant la lettre and his analysis of the reserve army, faulted the falling rate of profit argument as underargued, and concluded that Marx’s followers had buried his usable economics under his metaphysics while Keynes’s followers were in danger of the opposite burial. The essay enraged orthodoxies in both directions, which she counted a success, and her later work spent three decades building the post-Keynesian long period theory that the essay sketched: growth through class saving, distribution through conflict, history rather than equilibrium as the medium of analysis. Her famous impatience, expressed in her methodological writings, with theories that treated time as a logical device rather than as the irreversible medium in which expectations are disappointed, is the deepest point of contact between her Keynes and her Marx, both of whom, unlike the equilibrium tradition she opposed, wrote about processes that happen once and cannot be rerun.

The experiment’s lesson for the comparison is about translation costs. Robinson could read Marx as a Keynesian because the demand side of his system is genuinely extractable, and her extraction produced real insights that Marxian specialists have quietly used ever since. But the extraction had a price she acknowledged: the labour theory of value, discarded as a detour, was for Marx the load bearing proof that profit is unpaid labour, the proposition that converts his economics into his politics, and a Marx without it is a theory of capitalist malfunction rather than of capitalist exploitation. Keynesians who adopt Robinson’s Marx gain a colleague and lose the adversary; Marxians who adopt her Keynes gain the demand machinery and risk the same conversion in reverse. The bridge traffic, in other words, changes the travellers, and the honest comparison, this one included, must keep reporting the toll.

A guided passage: the General Theory read with Marxian questions in hand

Readers who know one founder and not the other often ask for a route into the General Theory that keeps the comparison alive, and the route matters because the book is badly served by its own reputation for difficulty and by summaries that stop at the slogans. Begin with the third chapter, the principle of effective demand in twenty pages, and read it asking the Marxian question of what has been assumed away: the answer, technique, distribution, and the capital stock, is the map of the book’s jurisdiction. Proceed to the eleventh and twelfth chapters, on the marginal efficiency of capital and the state of long term expectation, and ask the Keynesian question in return of any Marxian text at hand: where in this passage is the mechanism by which expectations, once formed, are disappointed together? The sixteenth and seventeenth chapters, on the nature of capital and the essential properties of interest and money, contain the book’s most radical and least read material, including the argument that capital assets are valuable for their scarcity and that a society which made them abundant would dissolve the rentier’s power without confiscating anything, a passage that startles readers who arrived expecting a technician’s manual. Close with the twenty fourth chapter’s social philosophy, read against the Gotha critique’s refusal of distributive slogans, and the two books’ deepest difference, whether the system’s evils are accidents of its management or expressions of its form, will have been encountered in the founders’ own words rather than in any intermediary’s, which is where a comparison of this weight deserves to be settled in the reader’s mind.

The hidden workshop: Marx’s notebooks on the economists

The Marx who faces Keynes in the comparison is usually the Marx of Capital’s three volumes, but the anti-Say argument that founds his side was worked out in the notebooks posthumously published as Theories of Surplus Value, and the workshop shows the argument’s quality better than the finished volumes do. There Marx takes Ricardo’s version of the doctrine apart sentence by sentence, granting its internal rigour and locating its error with a precision his popular reputation never credits: Ricardo, he argues, abstracts from money, from the separation of sale and purchase, and from the specific social form of production, treating capitalist production as if it were barter production with better tools, and therefore cannot see the crisis possibility that the form itself contains. The notebooks also contain his engagement with the glut controversialists, in which Malthus is simultaneously vindicated on the possibility of general overproduction and dismantled for explaining it by the consumption needs of an unproductive class whose preservation Marx regarded as the doctrine’s real purpose. Reading the workshop changes the comparison’s temperature. Keynes’s demolition of the classical doctrine in the General Theory, conducted at high speed and with a controversialist’s shortcuts, covers ground Marx had surveyed at walking pace eighty years earlier, and the parts of the Keynesian argument that Marxian critics call shallow, the treatment of money as a store of value that can interrupt circulation, are precisely the parts where the notebooks had already dug deepest. It does not follow that Marx anticipated Keynes, a claim the FAQ section handles with care in the Kalecki case where it belongs. What follows is that the frameworks’ shared refusal has a seniority the Keynesian tradition rarely acknowledges and the Marxian tradition rarely proves, because the proof lies in notebooks that were published in full only in the twentieth century and are still read mainly by specialists. The scholarly moral generalises: in a comparison between a published system and a posthumous archive, the archive’s owner will always appear, on any given page, both more prescient and less systematic than he was, and fairness consists in grading the published system against the archive’s considered results rather than against its most quotable fragments.

Say’s law in modern dress: loanable funds and the return of the veil

Doctrines of this depth do not die; they change their clothes, and part of the comparison’s continuing use is as a field guide to the reappearances. The classical claim that saving determines investment through the interest rate survives in the loanable funds apparatus of standard textbooks and in the crowding out doctrine that dominated policy argument in the decades after Keynes’s ascendancy: public borrowing, on the revived view, competes for a fixed pool of saving, raises interest rates, and displaces the private investment it ostensibly supplements, so that demand management is self defeating except as a palliative for the specific rigidities the doctrine concedes. The Keynesian replies are the ones this article has rehearsed, that the pool is not fixed because income moves, that in a slump the borrowing employs saving that would otherwise have leaked into idle balances, and that the empirical record of the austerity years, in which consolidation contracted economies faster than their official forecasters’ multipliers allowed, stands as the doctrine’s most expensive examination. The Marxian replies operate at the by now familiar deeper register: the crowding out story presupposes that the constraint on accumulation is a scarcity of saved funds, whereas the recurrent historical problem has been a scarcity of profitable placement for funds that exist in embarrassing abundance, the condition the secular stagnation revival rediscovered. A reader who has followed the comparison this far possesses the equipment to grade each new costume on its appearance: ask whether the argument treats saving as prior to investment, whether money enters as a veil or as an asset, and whether the state’s constraint is located in a market for funds or in the politics of distribution, and the three answers will assign the argument its lineage more reliably than its author’s self description. Lineage is not refutation, and the guide works both ways: a Marxian argument that quietly assumes demand failures impossible, as some vulgar versions do when they treat every slump as a profitability event by definition, has crossed to the other bank without announcing the journey, and the same three questions will catch it.

The ministry and the audit: policy design under both frameworks

Suppose a finance ministry, in the depths of a slump, adopts the Keynesian programme entire: deficits sized to the output gap, public investment brought forward, the central bank instructed to hold rates at the floor, and relief designed to reach the households likeliest to spend it. The comparison’s practical content can be stated as the audit the other framework then performs on the ministry’s own documents. The Kaleckian audit asks first about the programme’s time consistency under political pressure: which of its instruments create constituencies that will defend them, which create facts, a tightened labour market and a bolder workforce, that will be cited as reasons to reverse it, and whether the ministry has any theory of its own survival beyond the next election. The Marxian audit asks the deeper question about the programme’s incidence on the capital relation: if the spending restores profits by restoring demand, it restores also the accumulation whose dynamics prepared the slump, and the audit wants to know what has changed in the structure, in ownership, in the distribution of the surplus, in the dependence of livelihoods on the wage, such that the next cycle should end differently from the last. The Keynesian rejoinder, which the ministry will certainly make, is that survival and incidence are luxuries when a quarter of the industrial economy stands idle, that a framework which cannot act in the emergency forfeits the right to grade the action, and that the restructuring the audit demands was not on any ballot the ministry could have won. Between the programme and the audit stands the whole practical history of twentieth century economic management, and the comparison’s final service to the policy reader is to make the exchange explicit: every actual demand management regime has been a negotiation between the ministry’s arithmetic and the audit’s questions, conducted in the dark by participants who mostly knew only their own side’s brief. Kalecki, who served governments and watched them flinch, is again the figure who knew both, and his 1943 paper can be read, without distortion, as the audit’s founding document, written before the ministry’s programme had even been fully tried.

Labour supply and the reserve: the long run of the workforce

One axis remains that the standard comparison neglects and that both frameworks treat with surprising richness: the supply of labour itself across the long period. Marx’s reserve army is not merely a pool of unemployed that disciplines wages; it is a population produced by accumulation in layered forms, floating, latent, and stagnant, drawn from ruined peasants, displaced artisans, and the children of the employed, and its existence means that capitalist growth manufactures its own labour supply rather than bidding for a fixed endowment. Keynes, whose short period took the workforce as given, nonetheless wrote with great penetration about population in his interwar essays, fearing in the nineteen twenties that a declining population would deepen the investment famine by removing the surest outlet for new capital, housing and equipping a growing people, and speculating in the Galton lecture of 1937 that a stationary population would raise living standards only if the demand problem were solved by the distributional and investment policies his theory prescribed. Put the two together and a genuinely synthetic proposition emerges that neither founder stated alone: the size and security of the workforce is an endogenous product of the accumulation process, while the demand required to employ it is a separate variable that the same process does not automatically supply, so that a mature capitalism can find itself with a reserve army it no longer needs to replenish and an investment famine it cannot cure by the old demographic engines. The contemporary literature on ageing societies and their stagnation tendencies, conducted mostly in ignorance of both founders, reproduces the synthetic proposition’s terms with each new cohort study, and the comparison claims it, with due modesty, as another of its bridges waiting for its engineer.

Exile and insider: the biographies inside the theories

Theories bear the marks of the lives that produced them, and the comparison sharpens when the biographies are set side by side without reducing either theory to its author’s address. Marx wrote as a stateless exile, expelled from Prussia, France, and Belgium in turn, working in the British Museum’s reading room on an income assembled from journalism, Engels’s subsidies from the Manchester cotton trade, and small inheritances, while his household endured the poverty illnesses and child deaths that his letters record without self pity and with frequent fury. The theory that emerged is a theory seen from underneath the system, by a man for whom the labour market’s disciplines were not an abstraction and whose political hopes were invested in the class his economics placed at the centre. Keynes wrote as the consummate insider, Eton and King’s, the Treasury in two wars, the board of the Bank of England, Bloomsbury at his dinner table and the Arts Council as his wartime creation, a man who dined with the governors whose confidence his theory made a macroeconomic variable. The theory that emerged is a theory seen from the bridge, by an administrator who believed the ship sound and its officers educable. Neither vantage is an argument; plenty of exiles have written nonsense and plenty of insiders have seen clearly. But the vantages explain the theories’ characteristic blind spots with uncomfortable symmetry. The exile underrated the state’s capacity for intelligent self preservation, having met the state mostly as policeman and censor. The insider underrated the resistance his reforms would meet from his own dinner table, having experienced his class’s power as a medium he swam in rather than as a wall. Kalecki, the engineer from Lodz who was an outsider in Cambridge, an adviser distrusted in Warsaw, and an émigré twice over, occupied neither vantage and was confined by neither, and the comparison has already drawn the moral: the bridge was built by the man whose biography placed him on both banks and at home on neither.

Machinery and technical change: the axis of the long transformation

Technical change is the axis on which the frameworks’ difference in time horizon does its greatest damage to mutual comprehension, and a section belongs to it. Marx’s theory is, among other things, a theory of machinery: the drive to cheapen labour by revolutionising the instruments of production is the inner history of his accumulation, raising the mass of means of production each worker sets in motion, displacing workers into the reserve even as it multiplies the wealth they produce, and generating the tendential pressure on the profit rate that the crisis theory consumes. Every element of that sentence is a claim about a process measured in decades. Keynes’s framework, by deliberate construction, has almost nothing to say about machinery: equipment is given in the short period, technical progress appears, when it appears, as an influence on the marginal efficiency schedule, and the theory’s few long run speculations, the euthanasia of the rentier and the satiation of wants in his essay on the economic possibilities for his grandchildren, treat technical abundance as the solvent of the economic problem rather than as a protagonist in its dramas. The Kaleckian tradition supplies part of the missing middle, since Kalecki and Steindl theorised innovation as a determinant of the trend, monopolistic firms generating the profits that fund the research that entrenches their monopoly, and the evolution of technique becoming one more channel through which the distribution of income and the pace of accumulation govern each other.

The scoring on this axis is the most lopsided in the comparison, and the Keynesian side’s honest defenders have never seriously contested it. A framework that brackets the transformation of the production apparatus cannot explain the long waves of capitalist development, the succession of leading sectors, or the geography of deindustrialisation, and must import those explanations from outside itself when history demands them. The Marxian framework, conversely, has sometimes paid for its long view by treating the short period’s demand failures as surface foam on the deep tide, a judgement the nineteen thirties refuted at the cost of millions of livelihoods, and its best modern practitioners have learned to run the two clocks simultaneously, profitability setting the tide and effective demand the weather. The two clocks image is worth keeping, since it converts the frameworks’ rivalry into a division of temporal labour that the evidence, with its decades long profitability trends and its months long demand collapses, seems almost designed to endorse. It also locates the remaining genuine dispute exactly: whether the tide governs the weather in the last instance, as the Marxian tradition insists, or whether the weather, skilfully managed, can hold the tide’s damage permanently below the threshold of crisis, as the Keynesian tradition’s practice has always assumed. That dispute is not closable by any evidence yet in, because the experiment it requires, generations of uninterrupted skilled management, keeps being interrupted, and the interruptions are themselves the disputed evidence.

How the textbooks froze the comparison: the synthesis and its discontents

Between the founders and the modern reader stands a layer of textbook sediment that the comparison must clear away, because most educated people meet both frameworks in their sedimented forms. Keynes reached the textbooks through the neoclassical synthesis, Samuelson’s grand reconciliation in which the General Theory governed the short run of sticky wages while classical principles resumed command of the long run, and through Hicks’s IS LM diagram, which translated the book’s shifting schedules into two crossing curves and quietly dropped the uncertainty that Keynes had called his theory’s heart. Joan Robinson, who had helped build the original, named the result bastard Keynesianism and spent her later career protesting the baptism. Marx reached the textbooks, where he reached them at all, through a mirror operation: the crisis theory reduced to a disproportionality diagram or an underconsumption paragraph, the falling rate of profit stated as a slogan with its counteracting tendencies amputated, and the whole filed under the history of thought, where systems go to be summarised rather than used. Each sediment served the institution that laid it down. The synthesis made Keynes safe for a profession that needed his policy conclusions without his epistemology; the history of thought chapter made Marx safe for curricula that needed his name without his analysis. The financial crisis of 2008 performed an unplanned excavation: central bankers quoted Minsky, finance ministers quoted Keynes in the original for a season, and Marx’s sales figures rose in German bookshops, three resurrections that embarrassed the sediments equally. A comparison written after that excavation owes its readers the unmediated founders, which is why the guided passage above routes the student to chapters rather than to summaries, and why the FAQ section answers the priority question with names and journals instead of with the sediment’s usual shrug.

What crisis is for: the traditions’ tempers toward breakdown

The last axis is temperamental and political at once: what each tradition, at its most reflective, believes a crisis accomplishes in history. The Keynesian temper is medical and meliorist. Crisis is a fever of a body that need not be ill, dangerous chiefly because it wastes the patient’s strength and because fevers, untreated, invite quacks; the physician’s task is to bring the temperature down and keep it down, and the tradition’s saints are the administrators who learned the dosage. Its corresponding vice is impatience with anyone who finds meaning in the fever, a suspicion that the crisis’s interpreters are secretly in love with the catastrophe, and the suspicion has been aimed at the Marxian tradition in every decade since the thirties. The Marxian temper is diagnostic and, in a sense its critics rarely grant, conservative about evidence: crisis is the system’s periodic confession, the moment when the accounts it keeps with itself are forcibly reconciled with the surplus actually produced, and the tradition’s claim is not that the confession is desirable but that it is informative, revealing in weeks the relations that decades of smooth functioning concealed. Its corresponding vice is a temptation toward schadenfreude and toward reading every downturn as the final one, a temptation the tradition’s history indulges often enough to have earned the Keynesian suspicion honestly. Kalecki’s temper, characteristically, is ironic and engineering minded at once: crisis is what happens when a workable machine is operated by interests that cannot afford to let it work, and the appropriate response is neither the physician’s confidence nor the diagnostician’s patience but a redesign of who holds the controls, which is why his short paper ends as a work of political prognosis rather than of economics. Three tempers, three theories of what the breakdown means, and a reader who can identify which temper is speaking in any commentary on any slump possesses a critical instrument that outlasts the business cycle it was sharpened on.

War finance: the accidental laboratory of full employment

History ran the comparison’s crucial experiment once, under conditions no ethics committee would approve, and both frameworks have been citing the results ever since. The Second World War drove the advanced economies to something beyond full employment: unemployment in the United States fell from its depression plateau to levels below any peacetime recording, British mobilisation absorbed the entire interwar reserve of labour and drew married women into the factories in millions, and the output thus commanded was produced under state direction on a scale that made the peacetime policy debates read like miniatures. Keynes supplied the experiment’s operating manual in his 1940 pamphlet on paying for the war, arguing that the problem had inverted, from demand deficiency to an inflationary gap, and that the gap should be closed by deferred pay and taxation rather than by the price mechanism’s brutal rationing of the poor. The pamphlet’s success as policy is a permanent exhibit for his framework: the same economist whose machinery explained the slump explained the boom’s opposite danger with the same instruments, a symmetry that theories built only for crisis cannot match.

The Marxian reading of the laboratory is equally instructive and begins by granting the facts. War, on this account, is the one form of state spending whose political preconditions Kalecki’s theorem suspends: the employer class does not resist full employment when the alternative is national defeat, workers’ bargaining power is conscripted along with their labour, and the surplus is absorbed by destruction on a scale that solves the realisation problem by the simple expedient of exploding the product. The war economy thus appears, in the colder Marxian literature, as the limiting case that proves the peacetime rule, full employment being entirely compatible with private property provided the employment is in arms and the discipline is military. The peace that followed posed the experiment’s real question, whether the laboratory’s conditions could be approximated without its horrors, and the postwar settlement, with its welfare states and its military Keynesianism held in reserve, can be read as the two frameworks’ joint, unstable answer: Keynesian technique operating inside a Kaleckian political compromise, at a level of armament neither founder would have chosen as the surplus’s permanent destination. The reader who wants the strongest single demonstration that the comparison is not an academic parlour game should sit with the war economy until its double character sinks in: it is simultaneously the Keynesian framework’s greatest practical triumph and the Marxian framework’s most sorrowful confirmation.

The unpaid economy: a blind spot the two frameworks shared

Honesty about the comparison requires registering the territory both founders left unmapped, because the correctors of the omission have drawn on each framework and the correction now belongs to the subject. Both Marx and Keynes built their crisis theories around the wage labour market and the monetary circuit attached to it, and both treated the household’s unpaid work, the cooking, cleaning, nursing, and child rearing that reproduce labour power daily and generationally, as a background condition rather than as an economic magnitude with crisis relevance. Marx’s value of labour power includes the worker’s maintenance and replacement as costs, and his analysis of the working day knows exactly whose unpaid labour stretches a wage across a family’s needs, but the household appears in the theory as a site of consumption, not of production. Keynes’s household is thinner still, a propensity to consume with a psychology attached. The feminist economics that arose in the later twentieth century, from the wages for housework debates within the Marxian tradition to Folbre’s and others’ reconstruction of care as an economic sector, demonstrated that the omission is not decorative: an economy can export its crises into the household, absorbing unemployment by intensifying unpaid labour, cutting public services whose withdrawal is silently replaced by women’s work, and reporting the absorption as resilience. A demand failure that lands on the kitchen table does not disappear from the economy; it changes its accounting address, and frameworks that cannot see the address will misread the slump’s depth and its incidence alike. The correction is one of the comparison’s living frontiers, pursued by scholars in both lineages, and it exhibits the pattern this article has found everywhere else: the frameworks are most fruitful where their heirs refuse to treat the founders’ silences as boundaries.

The comparison exported: development, Prebisch and the periphery

The comparison travelled, and its travels supply a final test of both frameworks’ generality. In the postwar decades, the economics of development posed the founders’ questions in economies they had not studied: capital scarce, labour superabundant, demand constrained by the poverty of the majority and by structures of land ownership inherited from colonial orders. Kalecki himself became a central adviser in this world, working on the planning problems of India, Cuba, and other new states, and his framework migrated almost without translation: the problem of financing development without inflation, in his formulation, turned on the supply of food and necessities, on whether investment’s demand effects would meet an elastic supply of wage goods or crash into shortages that taxed the poor through prices, an analysis in which the class structure of agriculture did the work his degree of monopoly did at home. The Keynesian inheritance arrived through other channels and was transformed in transit: Prebisch and the Latin American structuralists rebuilt effective demand for economies whose centres of cyclical impulse lay abroad, arguing that the periphery’s crises were imported through its terms of trade and that industrialisation behind protection was the demand policy of nations that could not devalue their way to employment. The Marxian inheritance arrived most powerfully of all, in the dependency school’s claim that the periphery’s underdevelopment was not an original condition but a product of its integration, the surplus drained through the same circuits the metropolitan theory described. Three exports, three transformations, and a pattern the comparison’s reader will by now predict: in each new territory, the frameworks that survived the journey were the ones that shed their founders’ local assumptions fastest, while the questions, who validates the product, whose spending governs, what the class structure permits the state to do, proved to be the portable part. Portability of questions, across a century and across every variety of capitalism the century produced, is not the least of the findings a scoring section should record.

Measuring the master gauge: how the profit rate argument is actually conducted

The Marxian framework’s empirical burden was flagged earlier, and fairness requires showing the reader how that burden is actually carried in the literature, since the argument about the rate of profit is not a shouting match but a measurement dispute with identifiable decision points. The first decision is valuation: whether the capital stock in the denominator is carried at historic cost, what was paid for it, or at current replacement cost, what it would fetch or cost to reproduce under present conditions. The two conventions can move the measured rate in opposite directions across a period of inflation or technical change, and whole schools have formed around the choice, the temporal single system interpretation insisting on historic cost and sequential determination, its critics defending current cost as the magnitude relevant to decisions. The second decision is the boundary of production: whether the labour of finance, commerce, and much of the state counts as producing surplus value or as consuming it, a choice that moves both numerator and denominator and that maps, not accidentally, onto the frameworks’ larger dispute about which activities are the economy’s engine. The third decision is geography: whether the relevant rate is national, in a world of nationally kept accounts, or belongs to a world economy whose capitals compete across every border, a question the data answer badly because the accounts were designed for the Keynesian ministry’s purposes, not for Marx’s. Through all three decisions runs the finding that gives the dispute its heat: on most careful constructions, the rate of profit in the leading economies fell from its postwar peak into the nineteen seventies, recovered partially under the distributional reversals of the following decades, and never regained its golden age level, a contour broad enough to be visible through several conventions and contested at every point of detail. Kliman’s work reads the contour as the tendency asserting itself through the countertendencies and faults the recovery’s dependence on debt; other Marxian empiricists, Duménil and Lévy prominent among them, read the neoliberal decades as a genuine restoration of profitability for the owning classes achieved through finance and globalisation, and therefore dispute the crisis theory built on the fall. The disagreement is internal, evidence based, and unresolved, and the comparison records it as the framework’s honourable scar: this is what a falsifiable core looks like when it is actually being falsified and defended at once, and the Keynesian framework, whose core expectations are harder to corner, has no equivalent scar to show.

Measuring the multiplier: the Keynesian framework’s empirical wars

The Keynesian framework’s flagship magnitude has had its own century of measurement warfare, and the parallel is closer than either camp’s rhetoric admits. The fiscal multiplier, the number by which a unit of public spending multiplies into income, is the General Theory’s most consequential bequest to practice, and its estimated size has swung with the decades and the methods: the early postwar macroeconometric models found values comfortably above one, the rational expectations critique argued that announced policy would be discounted into impotence, the narrative identification literature rebuilt estimates from the documentary record of tax and spending changes, and the state dependent school established what Keynes’s own logic predicted, that the multiplier is large in slumps, when resources stand idle and monetary policy cannot offset, and small or negative near capacity. The International Monetary Fund’s post mortem on the European austerity years, in which Blanchard and Leigh reported that forecasters had systematically underestimated the multipliers embedded in their own projections, functioned within the profession as something close to an official concession that the slump values were the large ones, and the episode deserves its place in the comparison as the Keynesian framework’s equivalent of the profit rate contour: a core magnitude, measurable in principle, fought over in practice, and periodically vindicated at the moment of maximum political inconvenience. The deeper methodological symmetry deserves one sentence of emphasis. Both frameworks stake themselves on a magnitude, the rate of profit and the multiplier, whose measurement requires the measurer to have already decided contested questions of theory, so that the empirics of each tradition are partly an argument with itself conducted by other means, and the demand that one side produce its number while the other keeps its principles is a debating trick the reader should decline from whichever direction it arrives.

The corporation between them: joint stock capital and the rentier’s long life

The modern corporation is a test object the two frameworks examine from opposite ends with results that illuminate each other. Marx, in the third volume’s remarkable chapters on joint stock capital, saw the form as capitalism transcending itself within its own limits: ownership separated from management, the functioning capitalist rendered superfluous by salaried directors, capital appearing as social property inside the shell of private appropriation, and he called the development the abolition of capital as private property within the confines of the capitalist mode of production itself. Keynes, contemplating the same separation two generations later and with Berle and Means’s documentation of it arriving in his own decade, drew the complementary moral from the other bank: as ownership diffuses into rentier shareholding, the active direction of enterprise passes to managers whose time horizons and incentives diverge from the textbook capitalist’s, and the state’s task of guiding investment becomes easier, because the socialisation he proposed could proceed by influencing a managerial stratum already half detached from ownership’s imperatives. The two readings agree on the phenomenon, the divorce of ownership from control, and on its direction, the growing impersonality and social character of production, and they divide, as by now the reader expects, on the moral: for one tradition the form is a transitional shape whose contradictions, speculation unbound from production, managers as a new site of agency and extraction, must sharpen until the shell breaks; for the other it is the very mechanism by which the shell can be quietly repurposed. The subsequent history, in which the managerial corporation was first celebrated as the system’s humane maturity and then, from the nineteen eighties, re-subordinated to shareholder value by the financial markets Keynes had hoped to tame, reads like a staged experiment between the two morals, and the shareholder value era’s harvest, buybacks prioritised over investment, wage shares compressed, horizons shortened to the quarter, is the exhibit each tradition now brings against the other’s reading. Kalecki’s degree of monopoly, indifferent to the ownership form and attentive only to pricing power and its distributional consequences, once again describes the terrain on which both morals must operate, whichever of them the future confirms.

Throughput and limits: the ecological axis neither founder priced

The newest axis of the comparison is one neither founder could have drawn, and it is recorded here as an axis in formation rather than as a scored result. Both frameworks are theories of a growing system: Keynes needs investment to absorb the saving that prosperity generates and treats the satiation of absolute needs as a distant blessing that would retire the accumulation drive altogether, while Marx’s accumulation is growth by definition, its crises arising from too much production of the wrong composition rather than from any physical ceiling on production as such. A finite biosphere enters neither construction as a binding constraint, and the heirs of both traditions have spent two generations arguing about what the constraint does to the inheritance. The ecological Marxian literature, descending from the metabolic rift reading of Marx’s remarks on soil exhaustion and developed by Foster and others, claims resources inside the founder’s text: a theory of the rupture between social production and natural cycles, in which capital’s drive to expand throughput treats the conditions of production, soil, climate, the worker’s body, as free gifts until their degradation returns as cost and crisis. The Keynesian inheritance has been claimed from the other side by the green investment programmes that recur under names like the Green New Deal, whose logic is the multiplier’s: a vast, state led investment wave, in generation, insulation, and transport, that solves the demand problem and the carbon problem in a single motion, and whose Kaleckian critics ask the by now familiar question of which interests will be permitted to veto it and what happens to its politics when it succeeds. The confrontation is the comparison’s oldest pattern wearing its newest clothes: a technical programme of undeniable internal coherence meeting a structural analysis of the interests ranged against it, with the atmosphere keeping the accounts. No verdict is entered on this axis, because the evidence is still arriving and because the article’s discipline forbids pretending otherwise; the axis is recorded so that the next generation of the comparison, which someone will certainly write, knows that the question was visible from here.

The settlement’s terms: Beveridge, the White Paper and the one full trial

The objection section weighed the postwar generation as evidence; the settlement that organised it deserves its own inspection, because its terms show the comparison’s forces negotiating in documentary form. William Beveridge’s report of 1942 had named Want as the first of five giants and proposed to abolish it by social insurance; his Full Employment in a Free Society of 1944 went further, demanding that the state accept permanent responsibility for a labour market in which jobs outnumbered seekers, a condition he understood perfectly well would transform the bargaining atmosphere of every workplace in the country. The government’s Employment Policy White Paper of the same year accepted the responsibility in diluted form, committing to a high and stable level of employment rather than to Beveridge’s full employment, the dilution being precisely the margin in which Kalecki’s predicted politics was already operating before the regime began. The settlement’s other pillars, the health service, the nationalised commanding heights, the commitment to the international architecture Keynes had designed at Bretton Woods, composed a package in which the Keynesian economics was embedded in institutions meant to make its politics survivable: welfare provision decommodified enough of life’s necessities to lower the temperature of the wage bargain, while private ownership of industry, the settlement’s untouchable premise, preserved the structure whose confidence the whole arrangement had to court.

The trial’s conduct over the following three decades is the closest thing the comparison possesses to a controlled longitudinal study, and its data points have been entered throughout this article: the low unemployment and rising wages of the long boom, the stop go management of the external constraint, the profit squeeze as full employment’s distributional harvest ripened, the widening of demands from wages toward control that Kalecki had forecast in 1943 down to the vocabulary his successors used for it, and the eventual change of regime in which the restoration of unemployment as an instrument was defended in public, by its administrators, in the disciplinary language he had quoted in advance. No honest reading of the trial can award it wholly to either framework. The Keynesian instruments demonstrably worked, for a generation, at the task they were designed for, and no Marxian crisis mechanics predicted the generation’s length or its rising living standards. The Kaleckian politics demonstrably governed the trial’s ending, and no Keynesian technique explained why a successful regime was dismantled by the class it had enriched. The bridge stands over the trial as its best single interpretation, which is the finding the objection section reached by argument and this section reaches by chronicle. A reader inclined to distrust interpretations that fit their evidence too neatly should apply the distrust symmetrically: the trial also fits, with equal neatness, the simpler story that all policy regimes end and that class had nothing to do with it, and the choice between the stories turns on whether the documentary record of the ending, the stated reasons, the organised interests, the sequence of the wage and profit movements, is admitted as evidence. The comparison admits it, on both sides, everywhere, and asks to be judged by that standard.

The slump at kitchen table scale: the two frameworks in one household

Abstract frameworks earn their authority, in the end, by what they can say about a single household’s arithmetic when the slump arrives, and the exercise disciplines both. Take a machinist whose plant shortens its week as orders fall. The Keynesian account narrates the household’s danger as a contagion of spending: the family’s cutbacks, multiplied across the town, become the shops’ falling takings and the council’s falling revenues, so that prudence at each table sums to ruin at all of them, and the remedy is whatever restores the flow of spending past the family’s door, a public works payroll, a benefit that arrives without a means test’s delays, a mortgage whose payments the state has taught the bank to freeze. The Marxian account narrates the same household from the other side of the wage: the shortened week is the reserve army being replenished at the family’s expense, the overtime that vanishes and the bargaining power that vanishes with it are the mechanisms by which the surplus’s extraction will be raised when the recovery comes, and the family’s fear, rational and corrosive, is itself a factor of production in the next boom’s profitability. Kalecki’s household sits at the crossing of the two narrations, spending what it earns because it must, its wage share set by the pricing power of the firms that sell to it, its security rising and falling with the political weather around full employment rather than with any quality of its own conduct. Three narrations of one shortened week, and the test of the comparison’s reader is to be able to give all three without confusing their vocabularies, to know which facts each seizes and which it steps around, and to recognise that the household, which experiences the three as a single event called hard times, is under no obligation to find the frameworks’ quarrel as interesting as their agreement that the hardship was neither necessary nor the family’s fault.

Open files: three questions the comparison leaves undecided

A comparison that has scored what can be scored owes its reader a list of what remains genuinely open, since the boundary between settled and unsettled is where honest frameworks earn their authority. The first open file is the direction of causation in the typical modern downturn between profitability and demand. The empirical literature contains well constructed studies in both directions, episodes in which the profit rate’s decline precedes the investment collapse by years and episodes in which a demand shock arrives with profitability intact, and the emerging suspicion, that the sequence alternates with the cycle’s phase and the economy’s financial structure, would, if confirmed, convert the rivalry on this axis into a research programme about conditions rather than a contest of essences. The second open file is the political theorem’s scope. Kalecki derived the resistance to full employment from the workplace’s disciplinary economy, and the postwar trial fits the derivation; but the theorem’s performance across regimes with strong welfare states, coordinated bargaining, and public employment of last resort is less completely documented, and the Scandinavian cases in particular have been claimed, by competent scholars, both as the theorem’s refutation, since high employment coexisted for decades with social peace, and as its confirmation, since the peace was purchased by institutions that removed the wage bargain from the market’s discipline in ways the theorem’s logic predicts would be necessary. The third open file is the state’s balance sheet capacity itself: how long, and under what monetary arrangements, a government can carry the debt that countercyclical policy accumulates before the constraint binds, and whether the binding, when it comes, arrives as an economic event in the bond market or as a political event in the electorate. The modern monetary literature claims the constraint is real resources alone; the older Keynesian literature admitted confidence as a second constraint; the Marxian literature treats confidence as the form in which the class constraint announces itself in the bond market’s accent. All three files are being worked, with data the founders never saw, by scholars who mostly know which tradition they belong to and increasingly know the other tradition’s file as well. That, finally, is the comparison’s most hopeful finding, and it is sociological rather than theoretical: the frontier of the subject is populated by bilinguals, and the founders’ quarrel survives in their work as a division of labour they administer jointly, whatever the title pages of their books declare.

There is a temptation, on reaching the end of a long comparison, to ask which founder a sensible person should simply follow, as if the article had been a consumer test with the verdict withheld out of politeness. Resist it on the evidence assembled here. The sensible person follows Keynes into the emergency, because no other framework tells the fire brigade where the water is; follows Marx into the inquest, because no other framework asks why the building keeps catching fire in the same corner; and follows Kalecki into the council meeting afterwards, because he alone predicted, in writing and in advance, why the brigade’s budget would be cut while the corner was still smouldering. A reader who can perform those three followings in sequence, changing vocabulary as the scene changes and never mistaking one scene for another, has acquired the skill this comparison exists to teach, and the founders, who agreed about almost nothing except the refusal to be consoled by doctrine, might have recognised in that skill the beginning of the science both of them were trying, from their opposite banks, to found.

Scoring the comparison without rigging the scales

The article has refused a single verdict on principle, and the refusal needs defending against the suspicion that it is mere even handedness, the critic’s traditional hiding place. It is instead a conclusion about the object. The two frameworks differ in the depth of their explanations, the wavelength of their dynamics, and the questions they take as settled, so a verdict unqualified by axis and episode would be a category error dressed as decisiveness. What can be stated without qualification is a set of scored findings. On the rejection of self correction and of Say’s law, the frameworks agree, and the agreement is among the best supported propositions in macroeconomics. On the mechanics of a demand collapse and its repair, Keynes’s account is the more complete and the more directly usable, and a century of policy experience is its evidence base. On the origins of the surplus whose realisation is at stake, on the distributional structure that conditions demand, and on the political constraints that govern the repair’s durability, the Marxian account, completed by Kalecki, is the more complete, and the postwar regime’s biography is its evidence base. On the long run motion of profitability, the Marxian framework makes the bolder claim and carries the heavier empirical burden, unresolved in the literature to a standard either camp’s outsiders would accept. On financial crisis, the frontier work of both lineages converges in practice while insisting on different genealogies, a convergence the next crisis will presumably ratify again. A reader who can reproduce those five findings, with their evidence and their limits, knows more about the subject than most of the combatants whose books line the comparison’s shelves, and is equipped to read the founders themselves, which remains, after every summary including this one, the only court whose verdicts do not expire.

For students and exam candidates

This comparison appears on examination papers across economics, politics, and sociology, and it is marked by a rule students rarely guess: the caricature answer scores badly in both directions. An answer that reduces Keynes to spending and Marx to revolution announces that its author has met neither framework, and examiners reward the essay that opens with the agreement, since the rejection of Say’s law and of self-correction is the fastest route to demonstrating real reading. Structure the strong answer in four movements. State the shared ground first. Then give each framework its mechanism in its own terms, effective demand and volatile expectations on one side, accumulation and the production and realisation of surplus value on the other, keeping the causal sequences distinct. Then compare axis by axis, using the framework comparison table in this article as a checklist rather than a script, and award each axis separately. Finish with Kalecki, because he converts a survey into an argument: the independent derivation shows the demand result does not belong to one camp, and the 1943 paper supplies the synthesis examiners describe as evaluation. Keep the two vocabularies quarantined. Liquidity preference is not hoarding, the profit rate is not the marginal efficiency of capital, and confusing the pairs costs marks that the argument itself had earned. A reliable revision method is to file each thinker’s claims under the question it answers, and the Marxism study notes companion gives a structured place to keep those files, the comparison table, and the Kalecki profit logic side by side while preparing.

For teachers and seminar leaders

The productive seminar on this material begins by confiscating the slogans. Set the two steelmanned accounts as paired readings a week apart, Marx on the separation of sale and purchase and the General Theory’s third and twelfth chapters, and require each student to state the other side’s mechanism to the satisfaction of its defenders before any criticism is heard. The framework comparison table then works best as a provocation rather than a summary: assign one row to each group and ask the group to find the strongest textual evidence against its own cell, because the class treatment row and the expectations row both crack interestingly under that pressure. Kalecki’s short paper is the ideal closing text, short enough to read entire in the room and sharp enough to divide it, and the division is the lesson: students discover that their disagreement has quietly moved from economics to the politics of the workplace, which is the bridge argument experienced from inside. Watch for the two recurring seminar failures. One treats Keynes as a technician without politics, against the evidence of his final chapter. The other treats Marx as a demand theorist manqué, against the evidence of the production analysis. Both failures are corrected by returning to the causal sequences and asking, of every claim, where in the circuit it begins.

For researchers, historians, and journalists

For researchers, the live problems are precise ones. The priority question is settled in outline and open in detail: Kalecki’s Polish publications of the early nineteen thirties and his French journal article establish independence, while the profession’s reception, mediated by language and by Robinson’s advocacy, remains a case study in how citation networks allocate credit. The empirical contest between the frameworks turns on magnitudes that better data keeps revising: profit rate series, capacity utilisation, and the decomposition of downturns into demand shortfalls and profitability declines, with careful studies often finding the two signatures in sequence within a single episode. Historians of the postwar settlement will find the bridge hypothesis testable in principle, since it predicts that the political temperature of full employment tracks workplace bargaining power rather than the deficit numbers, and the archival record of business organisations’ internal deliberations is the underexploited source. Journalists need one discipline above all: when a downturn arrives, the accurate one-sentence contrast is that Keynes explains how far spending can fall, Marx why the system was primed to fall, and Kalecki why the rescue will be politically temporary, and compressing the three into a single camp’s victory manufactures the confusion this article exists to retire.

Frequently Asked Questions

Q: What is the main difference between Marx and Keynes on crisis?

The difference is one of depth. Keynes locates crisis in the sphere of spending: investment, governed by expectations that cannot be stabilised by calculation, collapses, and output and employment follow it down, a failure that determined public action can repair. Marx locates crisis in the sphere of production: the accumulation of capital presses on the conditions under which surplus value is produced and realised, so breakdown recurs however spending is managed. Both deny that the economy heals itself, and both reject the doctrine that supply creates its own demand. They part on whether the failure is a malfunction of an otherwise viable order or the periodic expression of the order’s own engine, and that parting decides what each regards as a remedy: management of demand in one framework, transformation of ownership in the other, with the historical bridge between them arguing that the first remedy works until the second question reasserts itself politically.

Q: Did Keynes read Marx?

The documented record suggests very little, and nothing like a sustained study. Keynes’s references to Marx across his writings are sparse, dismissive, and largely directed at Marxism as a political creed, which he attacked in his essays on Russia in the nineteen twenties, rather than at the economic analysis in Capital. In private correspondence with George Bernard Shaw he described Capital as an obsolete economic textbook, a judgement delivered with a confidence that historians of economics, including his major biographers, have generally taken as evidence of unfamiliarity rather than of a refutation he never published. No annotated copy, no serious review, and no engagement with the reproduction schemes or the profit rate argument survives in his papers. The fair summary is that Keynes rejected Marx the way his class and generation of Cambridge economists mostly did, at a distance, and that the convergences between the frameworks were achieved independently, which makes them more interesting, not less.

Q: Do Marx and Keynes both reject Say’s law?

Yes, and the agreement is foundational rather than incidental. Say’s law, in the form both men inherited, holds that production generates the income to buy the product, so general overproduction is impossible and saving flows automatically into investment. Marx rejected it at the level of form: in a monetary economy sale and purchase separate, money can be held back, chains of payment can snap, and the abstract possibility of crisis is contained in the commodity itself. Keynes rejected it at the level of mechanism: saving and investment are decided by different agents, income adjusts to reconcile them, and the economy can rest at underemployment equilibrium with no force carrying it back to full use of its resources. The two rejections differ in construction, one turning on the metamorphosis of commodities and the other on the determination of income, but they demolish the same doctrine, and the demolition is the shared foundation on which the entire comparison stands.

Q: Who was Kalecki and did he beat Keynes to it?

Michal Kalecki was a Polish economist who derived the principle of effective demand independently of Keynes, publishing his first statement in Polish in the early nineteen thirties and a version in a French journal before the General Theory appeared in 1936. His starting point was Marxian: the reproduction schemes of Capital and a class division of society into workers who spend their incomes and capitalists whose spending decisions determine profits. From those premises he reached the result that output and employment are governed by investment and capitalist consumption, the analytical heart of Keynesian economics, by a road Keynes never travelled. Joan Robinson, placed to judge, credited his priority in print. Beat Keynes to it is the correct verdict on the core result, with the refinement that Kalecki’s version was a theory of the cycle in a class economy, lacking the General Theory’s apparatus of liquidity preference and its programme for policy, while possessing a theory of distribution that Keynes’s book never had.

Q: Why do capitalists oppose full employment?

Kalecki’s 1943 answer remains the canonical one, and it is political rather than economic. A lasting regime of full employment, he argued, abolishes the disciplinary function of unemployment: when dismissal carries no threat because another job is readily available, authority inside the workplace erodes, workers press demands that extend from wages toward control over production, and the social position of the employer class declines even where its profits hold. Business leaders also dislike the state acquiring a permanent role in sustaining employment, since the instrument, once built, escapes their influence. The predicted consequence is not a conspiracy but a climate: pressure for sound finance, hostility to deficits in principle, and support for policies that restore a margin of unemployment, producing a political cycle in which full employment is repeatedly achieved and repeatedly abandoned. The argument explains the pattern of adoption and retreat without claiming that any individual employer calculates it, and it remains the most cited short paper in political economy for good reason.

Q: Can Keynesian policy solve Marxist crises?

On the bridge argument, it can solve the crisis in front of it and cannot repeal the class of crises. Demand management demonstrably restarts spending, re-employs idle resources, and can sustain prosperity for long periods; the postwar generation is the standing evidence, and any Marxian answer that denies it forfeits credibility. The Marxian claim, in its defensible form, is different: that the difficulties policy relieves reappear in displaced forms, as public debt, inflation, asset bubbles, or profit squeezes, and that the political conditions for continuing the policy erode as full employment shifts workplace power. Keynesian policy therefore solves the Keynesian problem completely and the Marxian problem conditionally and temporarily, on this view. A stricter Marxian would add that the distinction is illusory because the displaced forms are the crisis continuing by other means. The reader should notice that the two answers disagree about the bookkeeping of postponement, not about whether the spending works while it lasts.

Q: Is Keynes closer to underconsumption theory?

He stands nearer to it than to any other Marxian family, and the distance remains large. Underconsumption theories locate crisis in the restricted purchasing power of the masses: wages lag behind output, so the product cannot be sold at profitable prices. Keynes shared the instinct that deficient consumption demand matters and built part of his system on the propensity to consume, which falls as income rises and leaves investment to carry an ever heavier load. But his crisis trigger is not consumption at all; it is the collapse of investment under failing expectations, and he explicitly resisted making consumption the villain, since raising the propensity to consume was among his remedies rather than his diagnoses. Marx’s own relation to underconsumption was double, respecting it as a partial truth about the conditions of realisation while refusing it as a complete explanation, because crises characteristically break at the height of prosperity with wages rising. Keynes converged on a similar refusal from the opposite bank.

Q: Are Marx and Keynes answering the same question?

Partly, and the overlap is what makes the comparison worth having. Both ask why a monetary production economy can fail to employ its resources and can fall into slumps that no self-correcting mechanism reliably ends. To that shared question they give rival answers at different depths: Keynes explains the level of output given the structure of ownership and distribution, Marx explains the motion of that structure itself and the crises its motion generates. Where the questions coincide, in the rejection of Say’s law and of wage cutting, the answers reinforce each other. Where they diverge, many apparent contradictions dissolve into differences of jurisdiction, though not all: the sustainability of full employment under private ownership is a genuinely contested question on which the frameworks, and the Kalecki bridge between them, take distinguishable positions that evidence can inform. The safest formulation for an essay is that they answer one question together and two questions separately, and that confusing the three is the source of most bad comparisons in print.

Q: What do post-Keynesians share with Marxists?

The post-Keynesian school shares method and several mechanisms with the Marxian tradition while remaining a distinct lineage. Shared ground includes the principle of effective demand, the rejection of Say’s law, the treatment of money as endogenous and non-neutral, the insistence that uncertainty is irreducible rather than a calculable risk, and the view that distribution is settled by power and institutions rather than by marginal products. Joan Robinson’s generation pursued the overlap deliberately, extending Keynes into the long period with class based theories of saving and growth. The separations are equally definite. Post-Keynesians generally work without the labour theory of value and without exploitation as an organising category, treat the capitalist order as reformable in perpetuity rather than as historically transient, and draw their theory of prices from Sraffa rather than from Capital. Kalecki is claimed by both traditions and belongs fully to each, which is the institutional form the bridge takes in the history of the schools.

Q: Which framework better explains recessions?

The honest answer is that it depends on the recession, and the dependence is diagnostic rather than evasive. Downturns that begin with a collapse of spending, a financial panic, or a policy induced contraction fit the Keynesian mechanism closely, and its variables, expectations, liquidity, and the multiplier, organise the evidence well. Downturns preceded by a long profitability decline, rising capital intensity, and an investment boom that exhausts its own conditions fit the Marxian sequence, and several careful empirical studies of particular episodes find the profitability signature arriving before the demand signature. Many real recessions display both in sequence, fragility built up on the production and balance sheet side and detonated by a failure of spending or credit, which is precisely the configuration the Kalecki and Minsky line of work was built to describe. The frameworks are therefore best used as instruments with known ranges: the examiner’s question is a fair one only when it names the episode, and any answer that crowns a permanent winner has mistaken the instruments for teams.

Q: How does the framework comparison separate Marx and Keynes on saving?

The framework comparison treats saving as a point where the two frameworks reach similar conclusions from opposite directions. For Keynes, saving is a residual: it does not finance investment in advance but adjusts to it afterward, through changes in income, so an economy wide attempt to save more in a slump contracts income until saving falls back, the celebrated paradox of thrift. For Marx, the relevant category is not abstinence but accumulation: capitalists are driven to convert surplus value into additional capital, hoarding appears as a temporary phase in the circuit of capital, and the obstacle to investment is never a shortage of saving but a shortage of profitable outlets for the saving that accumulation itself generates. Put side by side, the comparison shows Keynes abolishing the saver’s virtue from the demand side and Marx dissolving it from the production side, leaving thrift with no macroeconomic defence in either framework, while disagreeing completely about what replaces it: expectations and the interest rate in one, the rate of profit in the other.

Q: What does the Kalecki bridge claim about profits and investment?

It claims that the causal arrow runs from investment to profits, not the reverse. In Kalecki’s derivation, capitalists as a class receive in profits what they spend on investment and on their own consumption, supplemented in an open economy with a state by the budget deficit and the export surplus. Workers spend their wages, so consumption out of wages cannot be a source of profit in aggregate; it merely returns wage income to firms. The consequence is that investment decisions, taken under uncertainty by firms watching their rivals, determine the mass of profit available to validate them, so a bold investing class creates its own reward and a frightened one destroys it. The bridge claim adds the political corollary: because profits depend on spending that public action can sustain, the economic obstacle to full employment is removable, and what remains is the political obstacle of a class whose workplace authority full employment dissolves.

Q: Why did Keynes never build his crisis account on the profit rate?

Because his theory deliberately stopped at a different depth. Keynes took the existing distribution of income, the stock of equipment, and the technique of production as data, and asked what determines output and employment within that frame. The rate of profit on capital as a whole plays no role in the answer because the variables doing the work, the propensity to consume, the marginal efficiency of capital, and liquidity preference, are defined without it; the marginal efficiency is a schedule of expected returns on new investment, not a theory of where profit comes from. Behind the method stood a judgement about what economics could settle and what belonged to politics: distribution, for Keynes, was a matter for social philosophy and fiscal policy, not for the theory of output. Marxian critics reply that the omission is the theory’s founding evasion, since the profit rate is where the production relation enters the arithmetic. Keynesians answer that a theory is not refuted by the questions it brackets. The comparison is most instructive when both statements are kept in view.

Q: How do the two frameworks differ on the role of money?

For Keynes, money is an asset with a liquidity premium, held precisely because the future is unknowable, and the interest rate is the price of parting with it; a monetary economy can therefore stall for monetary reasons, as wealth holders refuse illiquid commitments and investment waits on confidence that may not return. Money is causally active, created within the banking system, and capable of blocking the real economy’s use of its own resources. For Marx, money is the necessary form in which value appears and the only final means of payment, so the credit system can stretch the distance between production and final sale until claims on surplus value vastly outrun what production can honour, at which point the scramble for money itself becomes the crisis’s visible form. The kinship is genuine: both deny that money is a veil. The difference is one of depth, a monetary disorder that can be primary in Keynes and that in Marx expresses, in monetary dress, a disorder already gestated in production and accumulation.

Q: What should a student separate between liquidity preference and hoarding?

The two ideas belong to different frameworks and different problems, and examiners reward the separation. Liquidity preference is Keynes’s theory of the demand for money as an asset: given uncertainty, wealth holders balance the safety and saleability of money against the yield of bonds and real assets, and their collective preference sets the interest rate and can choke investment. It explains why money is held instead of spent on financial assets, and through interest, why it may fail to become investment. Hoarding, in Marx’s analysis, is a phase in the circuit of capital: money withdrawn from circulation as a treasure or as a reserve, interrupting the metamorphosis of commodities, significant because sale without a corresponding purchase breaks the chain by which other producers realise their values. One concept prices the holding of money within a portfolio; the other locates the holding of money within a flow of production and sale. A sentence that uses them interchangeably announces a confused essay, however elegant its prose.